Evidence-Based Power Workouts:

How to Combine Strength and Power Exercises (and Why Strength/Power Supersets Are Probably a Bad Idea)

By Dr. Brent Brookbush, DPT, MS, BS, CPT, HMS, SPC, IMT

Introduction

Should you do strength and power exercises in the same workout, and how much rest do you need between them? A systematic review of the 33 relevant studies on combining strength and power exercises answers both questions: yes, and rest 4 to 8 minutes between a heavy strength set and the matched power exercise. Power performance decreased when the power exercise followed within 30 seconds, was unchanged at 1 to 4 minutes, and improved most at 4 to 8 minutes. Building on that finding, the best power workout performs strength exercises first, uses a circuit format, and skips back-to-back supersets entirely.

The confusion is understandable. A search for "power workout" returns endless lists of jumps and throws and almost nothing on session design. Strength and conditioning resources disagree on rest, offering ranges as wide as 3 to 12 minutes and conceding that "opinions vary." Popular personal training certifications prescribe the opposite: a power exercise performed immediately after a heavy lift. This article resolves the question with research: 21 studies comparing rest intervals between a heavy strength set and a power exercise, 10 studies comparing session structures acutely and over 4 to 9 weeks of training, and 3 studies comparing circuit and traditional set structures.

What is the difference between complex training and contrast training (and supersets)?

The fitness industry uses these terms interchangeably, and the research literature is not much better, so the definitions used in this review are stated here.

• Complex training: Strength and power exercises are combined within the same training session, most commonly with all strength sets performed before all power sets.
• Contrast training: Strength and power sets are alternated one for one (a heavy set, then a power set, then a heavy set).
• Strength and power supersets: A power exercise is performed immediately after a heavy strength set, with little or no rest.

The research supports complex and contrast training when the rest between a strength set and the matched power exercise is adequate. The research does not support strength and power supersets.

Are strength and power supersets effective? (NASM Phase 5 style training)

If you studied the NASM Phase 5 power recommendations, or similar popular personal training models, you were taught to build power routines from supersets: a power exercise performed "immediately," "back-to-back," or "directly after" a heavy strength set for the same muscle group (e.g., a bench press followed by a medicine ball chest pass). The rationale is post-activation potentiation (PAP): the heavy lift recruits more motor units, and the subsequent explosive exercise should benefit.

However, this recommendation cites PAP as its rationale and then fails to align with the research on PAP. When a power exercise was performed within 30 seconds of a heavy strength set, power performance decreased significantly in 6 of the 8 samples tested, was statistically similar in 2 samples, and did not improve in any sample. When the same exercise pairing was separated by 4 to 8 minutes of rest, improvements in power performance were more consistent and were the largest identified in this review. Fatigue and potentiation coexist following a heavy set, and performing the power exercise immediately captures the fatigue before the potentiation can be expressed. In short, strength and power supersets are likely a bad idea.

How long should you rest between a strength exercise and a power exercise? (And does post-activation potentiation actually work?)

Post-activation potentiation works, and timing decides whether you see it. Rest intervals of 30 seconds or less resulted in decreased power performance. Rest intervals of 1 to 4 minutes generally resulted in performance similar to pre-exercise values. Rest intervals of 4 to 8 minutes resulted in the largest acute improvements in power performance, and findings for intervals of 10 minutes or longer were conflicting.

Note that the response varies. In the studies that reported individual responses, 56 to 80% of participants exhibited a positive response. Note that the samples that demonstrated significant improvements generally included stronger or professional athletes, which may imply that complex training is an advanced strategy, or that volume should be carefully monitored when introducing these programs to less experienced exercisers. One study demonstrated that stronger participants performed best after shorter rest; however, 1 finding among 21 studies may reflect chance, or a sample of exceptionally strong athletes. The 4 to 8 minute window is the most reliable prescription, but individual testing is recommended when maximal potentiation is the goal.

Should you do strength or power first in a workout? Can you do them in the same session?

Strength and power exercises can be, and should be, combined within the same session. Combined sessions resulted in outcomes similar to separate training days in the 2 studies that compared them, and combined sessions offer potential advantages: heavy strength sets may potentiate the power exercises that follow, and training both qualities in each session increases scheduling flexibility.

Strength exercises should be performed before power exercises. Research demonstrated that 9 weeks of block training with all squat sets before all jump squat sets resulted in a larger improvement in running vertical jump height than the reverse order, and another study demonstrated that during 8 weeks of concurrent soccer and resistance training, sprint speed decreased only when all plyometric sets were performed before strength sets. Lastly, 1 study demonstrated that relatively weak individuals exhibited lower jump power when heavy sets and jump sets were performed in alternating order. The trend is consistent but not strong; performing strength sets before power sets is likely to result in better power performance.

So, how do we take these facts (no supersets, 4 to 8 minutes of rest between matched pairs, and strength before power) and build a session that does not take 2 hours to complete? Circuit training resulted in similar set performance and similar multi-week adaptations when compared to traditional set structures, in approximately one third to one half of the session time. The routine below is built on that finding.

Source Material

This article is a focused review answering a specific question; the research was previously covered in the following courses and larger reviews:

1. Evidence-based Resistance Training Model (EBRTM)
2. Power Training: Evidence-based Model
3. Acute Variables: Complex Training
4. Acute Variables: Post-activation Potentiation
5. Acute Variables: Circuit Training

Caption: Power routine construction

Evidence-based Power Routine Construction

Summary Statement

• Strength and power exercises may be combined within a single training session (complex training) without compromising power performance or long-term adaptations. However, strength-and-power supersets (performing a power exercise immediately after a heavy strength set) are likely a poor choice. Research on rest between a heavy strength set and a power exercise demonstrated that intervals of 30 seconds or less decreased performance, intervals of 1 to 4 minutes resulted in performance similar to pre-exercise values, and intervals of 4 to 8 minutes resulted in the largest improvements in performance. Additionally, circuit training, with 30 to 60 seconds of rest between exercises, resulted in similar set performance and similar multi-week adaptations when compared to traditional set structures, in approximately one third to one half of the session time. Further, strength exercises should be performed before power exercises, as some research suggests this order results in better performance, although this trend is relatively weak. Based on these findings, the Brookbush Institute recommends performing strength and power exercises in the same session, in circuit format, with 2 to 4 strength exercises followed by 2 to 4 matched power exercises to maintain 4 to 8 minutes of rest between matched pairs, and 30 to 60 seconds of rest between exercises to maintain set performance.

Additional Details

• Combined sessions (complex training): Strength and power exercises should be combined in the same session. Research demonstrated similar outcomes for combined sessions and separate training days; however, combined sessions offer potential advantages: heavy strength sets may potentiate the power exercises that follow when rest is sufficient, and training both qualities in every session increases scheduling flexibility and may better align with linear and daily undulating periodization models, as opposed to potentially less effective block periodization approaches that develop each quality in separate phases.
• Circuit training: Sessions should be performed in circuit format. Circuit structures resulted in similar repetitions per set, bar velocity, and power output during heavy sets, and similar multi-week improvements in strength, power, lean body mass, and shuttle-run performance when compared to traditional set structures, in approximately half to two-thirds less session time.
• Rest: Rest between exercises should be 30 to 60 seconds (assuming a circuit training format), and rest between a strength exercise and a power exercise for the same muscle group should be approximately 4 to 8 minutes. Rest intervals of 4 to 8 minutes resulted in the largest acute improvements in power performance, and the optimal interval varied among individuals (56 to 80% of participants responded positively), so individual testing is recommended when maximal acute potentiation is the goal.
• Exercise order: Strength exercises should be performed before power exercises. Running vertical jump height improved more with strength-first training, and sprint speed decreased only when all plyometric sets were performed first.
• Format: The recommended format is 2 to 4 strength exercises followed by 2 to 4 power exercises matched by muscle group and movement pattern. This format is the culmination of the available research: it performs all strength exercises before power exercises, it provides the time needed for recovery and potential potentiation between each strength exercise and its matched power exercise, and it maintains the session efficiency of circuit training.
• Acute variables: Load, reps/set, sets/exercise, and tempo should match the training goal; the included studies improved power using strength exercises with 3-RM to 6-RM loads (approximately 85 to 90% of 1-RM) and power exercises with body weight to 30% of 1-RM loads — see "Power Training: Evidence-based Model" for more.
• Cautions: Power exercises should not be performed in a fatigued state (e.g., immediately following a strength set, following reps-to-failure sets, or without sufficient recovery). Further, some consideration should be given to volume (number of sets and exercises) for weaker and less experienced lifters. In this review, performance decreased when relatively weak individuals alternated heavy and jump sets set by set, and when sprints were performed shortly after fatiguing jump sets.

Recommendations

• Combine strength and power exercises in the same session.
• Perform sessions in circuit format.
• Perform 2 to 4 strength exercises followed by 2 to 4 power exercises, matched by muscle group and movement pattern.
• Rest 30 to 60 seconds between exercises, with approximately 4 to 8 minutes between a strength exercise and a power exercise for the same muscle group.
• Match load, reps/set, sets/exercise, and tempo to the exercise and training goal.

Caption: Squats and Box Jumps are great leg exercises for a power routine, but they should be spaced out so that there is 4 - 8 minutes rest between exercises.

Abbreviated Example: Power Routine

• See the full example with integrated corrective warm-up and daily undulated program here - Power Training: Evidence-based Model
• You can also create new programs in seconds: Evidence-based Program Generator

Acute Variables

• Cycle: 4-8 weeks
• Frequency: Upper/Lower Split 2 sessions each/week
• Load: Heavy (80 - 95% of 1-RM); Light (50-70% of 1-RM or Less)
• Reps/set: 3-8 reps-to-failure/set for heavy load strength exercises; 3-10 reps-not-to-failure/set for power exercises
• Reps/set (for Active Rest): 20 reps-to-failure/set
• Sets/exercise (or circuits): 2-4 circuits (sets/muscle group)
• Rest between exercises: 30-60 seconds between exercises, circuit training
• Rest between Circuits: 1 - 3 minutes
• Rep Tempo: As fast as can be controlled (X:X:X); Explosive (X:X:X)
• Training Time: 30 – 50 minutes (excluding warm-up).

Upper Body: Max Strength/Power Training (PAP) Routine

• Back (Strength): Pull-up (progress by increasing load, increasing sets/muscle group; alternative exercise: machine rows)
• Chest (Strength): Bench press  (progress by increasing load, increasing sets/muscle group; alternative exercise: machine chest press)
• Active Rest (Corrective/Core): Banded external rotation (progress by increasing band resistance, increasing sets/muscle group, and/or exercise progression: banded external rotation plus cervical retraction)
• Back (Power): Power chop (progress by increasing speed, distance, and last weight of the ball; alternative exercise: Power row sled pull )
• Chest (Power): Plyo push-up  (progress by plyo push-up onto two boxes, increase speed; alternative exercise: Medicine ball chest pass )
• Active Rest (Corrective/Core): Quadruped crawl (progress by increasing duration, increasing sets/muscle group, and/or exercise progression)

Lower Body: Max Strength/Power Training (PAP) Routine

• Legs (Strength): Back squat (progress by increasing load, sets/muscle group, alternative: leg press)
• Active rest/Corrective: Standing anti-rotation chops (progress by increasing load, increasing sets/muscle group, and/or exercise progression: single-leg standing anti-rotation chops )
• Legs (Power): Depth jump  (progress by increasing the height of the depth jump box, increase height of the landing box, alternative: Box jump )
• Active rest/Corrective: Miniband shuffles (progress by increasing band thickness, increasing sets/muscle group, and/or exercise progression: monster walks )

Note: For strength goals, progressing load should be the priority. Sets/exercise can also be increased, especially when lifting with very heavy loads, resulting in 1-4 reps/set.

Note: For power goals, the goal is to increase speed, height, or distance in the same amount of time. Additional reps may be added, but only if speed can be maintained. Adding load should only be considered when there is no other option.

Note: Active Rest (Corrective Exercise): Active rest is appropriate when aiming to ensure at least 4 minutes of rest between exercises that target the same or similar muscles in a circuit. Generally, the Brookbush Institute recommends using a core or corrective exercise during active rest to maximize the utility of the selected exercise. We do not recommend foam rolling or stretching techniques because some studies have shown that their inter-set use reduces performance in subsequent sets.

Caption: Strength exercises should precede power exercises in a power routine

Summary of Research Findings

Timing and Rest Between Strength and Power Exercises

Summary Statement: Twenty studies compared power performance (vertical jumps, sprints, cycle sprints, kicks, or bench press throws) following a strength exercise, with rest intervals of 0 seconds to 24 minutes. During power exercise performed 0 to 30 seconds after a heavy strength set, performance decreased significantly in 6 samples (2 of which may share participants), was statistically similar to pre-exercise values in 2 samples, and no sample exhibited an increase in performance. When significant improvement in power performance was demonstrated following a heavy strength set, the increase occurred after approximately 4 to 12 minutes of rest, and performance frequently returned to pre-exercise values by 10 to 24 minutes. That is, performing a power exercise immediately after a heavy strength exercise for the same muscle groups likely reduces or fails to enhance power performance, and a rest interval of 4 to 8 minutes is most likely to include any potential increase in performance.

• Less than 30 seconds rest: Research has demonstrated that jump height and peak ground reaction force decreased 10 seconds after a set of squats with 5-RM loads, vertical jump flight time decreased 30 seconds after a set of squats with 5-RM loads, power output and jump height decreased 15 seconds after 3 sets of squats with 87% of 1-RM, countermovement jump and bench press throw peak power decreased 15 seconds after a set of squats or bench press with 3-RM loads, jump height was significantly lower than a control group 15 seconds after a set of squats with 3-RM loads, and countermovement jump average rate of force development decreased approximately 12% at 0 seconds after a set of squats with 3-RM loads (significantly lower than at 2 and 5 minutes). Additionally, studies have demonstrated that sprint times and jump height were statistically similar to pre-exercise values 20 seconds after squat, weighted jump, and isometric protocols, and that jump performance was similar 30 seconds after a set of back squats with 3-RM loads with rest redistributed between repetitions.
• Rest of 1 to 4 minutes: Research has demonstrated that intervals of 1 to 4 minutes generally result in performance similar to pre-exercise values, with some studies showing trends toward improvement that failed to reach statistical significance.
• Rest of 4 to 8 minutes: Most of the significant improvements in power performance occurred following rest intervals of approximately 4 to 8 minutes. Following 10 sets of 1 rep/set of squats with 90% of 1-RM, cycle sprint average power output increased significantly with a 5-minute rest when compared to a 20-minute rest, and 10- and 30-meter sprint times also improved significantly with 5-minute rests when compared to 3-minute rests. Further, countermovement jump height increased significantly at 4, 6, and 8 minutes after a set of squats with 5-RM loads, and countermovement jump power output, jump height, and peak rate of force development increased significantly at 8 minutes after 3 sets of squats with 87% of 1-RM. Additionally, protocols that used self-selected rest intervals (averaging approximately 3.5 and 6 minutes in the 2 studies) after a set of squats with 5-RM loads resulted in improvements equal to or greater than fixed rest intervals.
• Rest of 10 minutes or more (conflicting findings): Some studies demonstrated that improvements were not maintained with longer rest intervals: countermovement jump power output returned to values statistically similar to pre-exercise values at 12 to 24 minutes after 3 sets of squats with 87% of 1-RM, cycle sprint average power output increased less following a 20-minute rest when compared to a 5-minute rest after 10 sets of squats with 90% of 1-RM, jump height was statistically similar at 10 minutes after a set of squats with 5-RM loads, and flight time decreased significantly at 6 minutes after a set of squats with 5-RM loads in 1 sample. However, other studies demonstrated improvements at 10 minutes or later: loaded half squat jump power output increased significantly at 10 minutes (but not 5 minutes) after 2 sets of half squats with 4-RM loads or 2 sets of half squat jumps, jump and bench throw peak power increased significantly at 12 minutes (and 16 minutes for throws) after a single set with 3-RM loads, isometric rate of force development increased significantly at 15 and 20 minutes after 5 single squats with 1-RM loads (after decreasing at 2 and 10 minutes), sprint times were fastest at 12 to 16 minutes after an isometric squat protocol, and jump performance remained improved at 4 to 20 minutes following a set of flywheel squats at the power-maximizing load. Note that this pattern implies the window for improved performance closes as well as opens, and that the timing of the window varies with the conditioning exercise, training volume, and sample.
• Jump height maintained with an altered jump strategy: During jumps performed with little rest after a set of back squats with 3-RM loads, propulsive impulse (the primary determinant of jump height) remained statistically similar to pre-exercise values, while propulsion time increased and rate of force development decreased or was not enhanced. This pattern was demonstrated in 2 samples. Note that this implies an unchanged jump height following a heavy squat set may conceal a reduction in rapid force production, and that jump height alone may not detect the cost of an insufficient rest interval.
• Low training volumes: The protocols that resulted in group-level performance statistically similar to the control, regardless of the rest interval, included lower training volumes (e.g., a single set of squats with 3-RM loads or 5 reps with 85% of 1-RM, or a single set limited by velocity loss to approximately 4 reps). However, a single set with 3-RM loads resulted in significant changes in professional rugby players, and each participant's best rest interval after a single set with 85% of 1-RM resulted in significantly higher power than the control. That is, a single low-volume set may not provide a sufficient stimulus to change the performance of a subsequent power exercise at the group level, particularly in less trained samples.
• Individual variation and training status: The rest interval that resulted in each individual's best performance differed among participants in the 6 studies that analyzed individual responses, and 56-80% of participants exhibited a positive response in the 3 studies that classified responders. Additionally, the samples that exhibited significant improvements generally included professionals or stronger athletes; stronger participants exhibited their best performance after shorter rest intervals in 1 study, and changes in sprint times correlated with relative strength in another. The samples with a relative 1-RM back squat of 1.45 times body mass or less did not exhibit group-level improvement at any rest interval, whereas improvement was demonstrated in samples from approximately 1.6 to 2.4 times body mass. Note that relative strength was reported as 1-RM, estimated 1-RM, 3-RM, or 5-RM across studies, and these metrics should not be compared directly.
• Upper body (bench press): Bench press exercises exhibited the same general pattern, with the rest interval required for improvement increasing with the load of the strength set. Sets of bench press with 65% of 1-RM resulted in improved power output following a 4-minute rest; sets with 85% and 100% of 1-RM loads required 7 to 8 minutes of rest, and a set of bench press with 85% of 1-RM followed by a 4-minute rest resulted in a decrease (approximately 1%) in power output. Additionally, bench press throw peak power decreased 15 seconds after a set of bench press with 3-RM loads and increased significantly at 8 to 16 minutes.

Exercise Order, Alternating Pairs, and Circuits

Summary Statement: These studies demonstrate that the structure used to combine strength and power exercises, blocked in either order, alternated set by set, paired within the same set, or separated into different sessions or days, results in largely similar acute power outputs and multi-week improvements in strength and jump performance, implying structure can generally be selected for practicality (e.g. equipment, space, and schedule). Four exceptions were noted: alternating heavy and jump sets acutely reduced jump power in relatively weak individuals, sprint times were slower when sprinting was performed shortly after countermovement jumps, performing all heavy sets before all jump sets during multi-week training resulted in a larger improvement in running vertical jump height, and sprint speed decreased over 8 weeks of concurrent soccer and resistance training only when all plyometric sets were performed before strength sets.

• Acute exercise order: Performing all heavy squat sets before or after all jump sets resulted in statistically similar session jump height, peak power, and force in resistance-trained athletes, although peak power during the 1st jump set was significantly lower immediately following a strength-first block in 1 study. Alternating heavy and jump sets set by set resulted in significantly lower session jump power than jumps alone in recreationally trained men who could half squat approximately 1.5 times their body weight, whereas stronger athletes who could half squat approximately 2 times their body weight exhibited increased force output with the alternating order, suggesting the acute response to alternating pairs may depend on relative strength. Additionally, 5- and 20-meter sprint times were significantly slower when performed shortly after countermovement jumps and were statistically similar when performed after heavy squats. Note that formats preserving 3 to 4 minutes of rest between a heavy set and the subsequent jump set did not reduce jump power, consistent with the rest interval findings in section 1.1.
• Multi-week session structure: Combining strength and plyometric or jump exercises within the same session, within the same set, in blocked orders, or in separate sessions or days for 4 to 9 weeks resulted in statistically similar between-group outcomes for nearly all measures in every study; lower-body 1-RM strength increased significantly with all formats tested, and vertical jump height increased significantly regardless of format in 3 studies, only with alternate-day training in 1 study, and in no group in 1 study. Three format-specific differences were noted: 10- and 20-meter sprint speed decreased significantly over 8 weeks of concurrent soccer and resistance training when all drop jump sets preceded half squat sets, but not with strength-first or alternating sequences; running vertical jump height improved significantly more when all squat sets preceded jump squat sets during 9 weeks of block training; and 6 weeks of paired sessions resulted in greater percentage increases in box squat and bench press strength and increases in muscle fiber cross-sectional area when compared to alternate-day training, although the paired group was stronger at baseline and performed more sets per session. In that same study, countermovement jump and overhead throw performance increased significantly only with alternate-day training, without significant between-group differences.
• Circuit structure and training density: Sequencing different-muscle-group exercises between sets with 35 seconds of transition, rather than resting passively, preserved approximately 3 minutes of recovery for each muscle group, reduced session duration by approximately two thirds with the same exercises, loads, and volume, and resulted in significantly higher oxygen consumption, heart rate, blood lactate, energy cost, and excess post-exercise oxygen consumption than a traditional set structure.

Session Structure - Exercise Order, Alternating Pairs, and Circuits

Summary Statement: These studies demonstrate that heavy resistance exercises performed in circuit structure result in similar set performance and similar multi-week adaptations when compared to traditional set structure, in approximately half the session time. Repetitions per set, bar velocity, and power output were statistically similar during heavy sets with 6-RM loads whether different-muscle-group exercises were performed between sets or the rest was passive, although heart rate was significantly higher during the circuit condition. Following 8 weeks of training with 6-RM loads, bench press and half squat 1-RM strength, bench press peak power, lean body mass, and shuttle-run performance improved significantly and similarly with circuit and traditional structures. Following 6 weeks of a mixed Olympic lift and strength exercise program in high school football players, strength increased significantly and similarly with blocked and circuit orders, and sprint and agility times did not improve with either.

Caption: 4 - 8 minutes of rest between strength and power exercises during a power training routine.

Annotated Bibliography

Section 1.1: Studies Including Immediate Rest Intervals (30 Seconds or Less)

Countermovement jump height and peak ground reaction force decrease significantly 10 seconds after a set of back squats with 5-RM loads, and return to pre-exercise values at 1 to 4 minutes; jump performance is not significantly enhanced at any interval.

Jensen and Ebben compared 21 NCAA Division I athletes 11 males (age: 21.4 ± 1.9 years) and 10 females (age: 19.6 ± 1.0 years) participating in collegiate sport. Following a warm-up and a pre-exercise countermovement jump, all participants performed back squats for 1 set/exercise, 5 reps/set, with 5-RM loads, followed by 5 additional countermovement jumps at 10 seconds, 1, 2, 3, and 4 minutes after the squat set, during 1 session. Outcome measures included jump height and peak ground reaction force (force platform, sampled at 1,000 Hz); an additional analysis compared the 8 strongest participants to the 8 weakest participants by 1-RM squat. The findings demonstrated that jump height at 10 seconds after the squat set (0.66 m) was significantly lower than the pre-exercise jump (0.74 m) and all other post-squat jumps, and that peak ground reaction force at 10 seconds was significantly lower than at all other repetitions for males and females. Jump height at 1, 2, 3, and 4 minutes (0.72 - 0.76 m) was statistically similar to the pre-exercise jump, with a trend toward improvement that failed to reach statistical significance. Additionally, jump height did not differ across repetitions between the high- and low-strength groups, and males jumped significantly higher than females at all time points. Note that pilot testing demonstrated no effect of repeated jumping alone, which implies the changes resulted from the squat set and recovery period rather than the preceding jumps. Note that the exercise protocol included 1 set of squats and single jumps at each interval during a single session, and the significant decrease occurred only at the interval in which the jump immediately followed the squat set.

• Jensen, R. L., & Ebben, W. P. (2003). Kinetic analysis of complex training rest interval effect on vertical jump performance. Journal of Strength and Conditioning Research, 17(2), 345-349.

Countermovement jump flight time decreases significantly 30 seconds after a set of back squats at 5-RM loads, is similar to baseline at 2 and 4 minutes, and decreases again at 6 minutes; the rest interval producing each individual's best and worst performance varied among participants.

Comyns et al. compared 18 athletes (9 males, 9 females) from predominantly anaerobic sports (sprinting, jumping, and rugby) with weight lifting and plyometric training experience and a back squat of at least 1.5 times body mass (females: 1.9 ± 0.3; males: 2.1 ± 0.2 times body mass). Following 1-RM testing and familiarization during a first session, all participants completed 2 testing sessions, each including a warm-up, 3 pre-exercise countermovement jumps, and 2 complex pairs. Each complex pair included back squats for 1 set/exercise, 5 reps/set, with 5-RM loads (87% of 1-RM loads), followed by a 30-second, 2-minute, 4-minute, or 6-minute rest (randomly assigned), followed by 3 countermovement jumps, with a very long (10 min) rest between complex pairs. All jumps were single-leg countermovement jumps performed on an inclined (30°) sled apparatus with the trunk and arms secured by a harness (trial-to-trial and day-to-day intraclass correlation coefficients: 0.98 - 0.99). Outcome measures included flight time and peak ground reaction force (force platform, sampled at 1,000 Hz). The findings demonstrated that, for the entire group, flight time was significantly reduced at the 30 second and 6 minute intervals when compared to pre-exercise values, and was statistically similar to pre-exercise values at the 2 and 4 minute intervals. For females, flight time was significantly reduced at the 30-second and 6-minute intervals, and no interval resulted in improvement; for males, no interval resulted in a significant reduction or improvement, with a trend toward improvement at the 4-minute interval that failed to reach statistical significance. Peak ground reaction force was statistically similar to pre-exercise values at all intervals. Additionally, an analysis of each participant's best and worst interval demonstrated that the largest decreases in flight time and ground reaction force were statistically significant for males and females, the largest increase in flight time was significant for males, the largest increase in ground reaction force was significant for females, and the interval producing each participant's best and worst performance differed between participants. Note that the significant reduction at 6 minutes may imply the window for improved performance closes as well as opens, and that single-leg jumps performed on a sled apparatus are not directly comparable to free countermovement jumps.

• Comyns, T. M., Harrison, A. J., Hennessy, L. K., & Jensen, R. L. (2006). The optimal complex training rest interval for athletes from anaerobic sports. Journal of Strength and Conditioning Research, 20(3), 471-476.

Countermovement jump power output and jump height decrease significantly 15 seconds after 3 sets of back squats with 87% of 1-RM loads; power output, jump height, and peak rate of force development increase significantly at 8 minutes and are statistically similar to pre-exercise values at 4 and 12 to 24 minutes

Kilduff et al. compared 20 professional male rugby athletes (age: 25.4 ± 4.8 years; body mass: 102.5 ± 11.5 kg; estimated 1-RM squat: 201 ± 41 kg, approximately 2.0 times body mass) with at least 2 years of structured weight training experience. Following a warm-up and a pre-exercise countermovement jump, all participants performed back squats for 3 sets/exercise, 3 reps/set, with 87% of estimated 1-RM loads and very long (4 min) rest between sets, beginning approximately 15 minutes after the pre-exercise jump, followed by countermovement jumps at 15 seconds, 4, 8, 12, 16, 20, and 24 minutes after the squat sets, during 1 session. All jumps were performed with the hands on the hips to eliminate arm swing (sampled at 1,000 Hz). Outcome measures included power output, jump height, and peak rate of force development (test-retest intraclass correlation coefficients: 0.979, 0.976, and 0.890). The findings demonstrated that power output and jump height (32.4 compared to 34.3 cm) at 15 seconds decreased significantly when compared to pre-exercise values, and peak rate of force development at 15 seconds exhibited a decrease that failed to reach statistical significance. Power output, jump height (36.0 cm; an increase of 4.9%), and peak rate of force development at 8 minutes increased significantly when compared to pre-exercise values and all other time points, and all outcome measures were statistically similar to pre-exercise values at 4, 12, 16, 20, and 24 minutes. Additionally, 3-RM strength correlated significantly with the change in power output at 8 minutes (r = 0.489), and 14 of 20 participants (70%) exhibited their highest power output, peak rate of force development, and jump height at 8 minutes, with 3 participants at 12 minutes and 3 participants at 4 minutes. Note that a jump-repetition control was included: 10 participants completed 7 countermovement jumps with 4 minutes between jumps and no squat sets; no significant time effect was observed, so the changes can be attributed to the squat sets. Note that this study and Kilduff et al. (2007) were performed by the same laboratory with professional rugby players from the same club, and the publications do not state whether participants overlap; overlap should be assumed possible when counting datasets.

• Kilduff, L. P., Owen, N., Bevan, H., Bennett, M., Kingsley, M. I., & Cunningham, D. (2008). Influence of recovery time on post-activation potentiation in professional rugby players. Journal of Sports Sciences, 26(8), 795-802. https://doi.org/10.1080/02640410701784517

Countermovement jump and ballistic bench throw peak power output decrease significantly 15 seconds after 1 set of back squats or bench press with 3-RM loads, increase significantly at 8 and 12 minutes (and at 16 minutes for bench throws), and the changes in power output correlate with 3-RM strength

Kilduff et al. compared 23 professional male rugby players (13 senior international players; age: 24.0 ± 3.4 years; body mass: 97.4 ± 13.4 kg; estimated 1-RM squat: 153 ± 23 kg; estimated 1-RM bench press: 124 ± 16 kg) with at least 1 year of structured weight training experience, tested during the preseason after completing a power training phase. All participants completed a lower-body session (countermovement jumps and back squats) and an upper-body session (ballistic bench throws and bench press) 48 hours apart. Following a warm-up, a pre-exercise jump or throw, and a very long (10 min) rest, all participants performed the preload exercise (back squats or bench press) for 1 set/exercise, 3 reps/set, with 3-RM loads, followed by jumps or throws at approximately 15 seconds, 4, 8, 12, 16, and 20 minutes. Countermovement jumps were performed with a bar over the shoulders attached to a cable-extension transducer (sampled at 500 Hz), and bench throws were performed on a Smith machine with 40% of estimated 1-RM loads, with the bar released at the top of the range of motion. Outcome measures included peak power output. The findings demonstrated that countermovement jump peak power output decreased significantly at 15 seconds (4,568 ± 509 compared to 4,430 ± 495 W), was statistically similar to pre-exercise values at 4 minutes, increased significantly at 8 minutes (4,862 ± 485 W) and 12 minutes (4,911 ± 444 W) (the 8 and 12 minute values were statistically similar to one another), and was statistically similar to pre-exercise values at 16 and 20 minutes. Bench throw peak power output decreased significantly at 15 seconds (856 ± 121 compared to 816 ± 121 W), was statistically similar at 4 minutes, increased significantly at 8 minutes (880 ± 130 W), 12 minutes (903 ± 145 W), and 16 minutes (863 ± 126 W), and was statistically similar at 20 minutes. Additionally, 3-RM strength correlated significantly with the change in peak power output at 12 minutes for the lower body (r = 0.631) and upper body (r = 0.590), and at 8 minutes for the lower body (r = 0.563). Note that the relative estimated 1-RM squat of this sample (approximately 1.6 times body mass) was similar to the samples in this section that did not exhibit improvement; however, the strength metric (estimated 1-RM relative to body mass) differs from the relative 3-RM values reported by Houlton et al., and these ratios should not be compared directly. Note that this study and Kilduff et al. (2008) were performed by the same laboratory with professional rugby players from the same club, and the publications do not state whether participants overlap; overlap should be assumed possible when counting datasets (??)

• Kilduff, L. P., Bevan, H. R., Kingsley, M. I. C., Owen, N. J., Bennett, M. A., Hore, A. M., Maw, J. R., & Cunningham, D. J. (2007). Postactivation potentiation in professional rugby players: Optimal recovery. Journal of Strength and Conditioning Research, 21(4), 1134-1138.

Countermovement jump height decreases significantly 15 seconds after a set of back squats with 3-RM loads when compared to a control group in professional soccer players; within-group contrasts demonstrate decreases at 15 seconds and at 8 and 12 minutes, and 6 of 11 participants exhibit individualized potentiation profiles.

Mola et al. compared 22 male professional soccer players (age: 23 ± 4.5 years) with at least 1 year of structured resistance training experience, randomly assigned to a squat group (n = 11) or a control group (n = 11). Following 3-RM testing and familiarization, a 30 minute seated washout, a warm-up (5 min of cycling and 2 min of dynamic stretching), a pre-exercise countermovement jump, and a very long (10 min) rest, the squat group performed back squats for 1 set/exercise, 3 reps/set, with 3-RM loads, followed by countermovement jumps at 15 seconds, 4, 8, 12, 16, and 20 minutes; the control group performed the identical jump protocol without the squats. All jumps were performed with the hands in the pockets to eliminate arm swing, to a self-selected depth (force platform). Outcome measures included jump height, peak power, and percentage potentiation (each jump relative to the pre-exercise jump). The findings demonstrated that jump height was significantly lower in the squat group than the control group at 15 seconds, with no other significant between-group differences for jump height or peak power at any time point. Within the squat group, the omnibus tests failed to reach statistical significance (peak power: p = 0.063; jump height: p = 0.089), and follow-up contrasts demonstrated significant decreases in peak power and jump height at 15 seconds and at 8 and 12 minutes when compared to pre-exercise values, with values statistically similar at 4, 16, and 20 minutes; no significant changes occurred in the control group. Additionally, 6 of 11 squat group participants exhibited individualized potentiation profiles, peaking at 4 minutes (n = 3), 12 minutes (n = 1), and 16 minutes (n = 2), and 5 participants exhibited no potentiation. Note that the within-group decreases follow contrasts performed after non-significant omnibus tests and should be interpreted with caution; the between-group difference at 15 seconds used corrected independent t-tests. Note that 3-RM testing preceded the experimental trial by 30 minutes, which may have contributed fatigue. Note that the protocol included a single 3-RM set, a lower training volume than the protocols in this section that resulted in significant improvements.

• Mola, J. N., Bruce-Low, S. S., & Burnet, S. J. (2014). Optimal recovery time for postactivation potentiation in professional soccer players. Journal of Strength and Conditioning Research, 28(6), 1529-1537.

Countermovement jump average rate of force development decreases approximately 12% when jumps are performed 0 seconds after a 3-RM back squat set, and is significantly lower than at 2 and 5 minutes; propulsive impulse is statistically similar across rest intervals of 0 seconds to 5 minutes for countermovement and squat jumps

Houlton et al. compared 14 male recreational athletes (age: 24.86 ± 4.45 years; relative 3-RM back squat: 1.63 ± 0.24 times body mass) with at least 1 year of strength and power training experience and a back squat of at least 1.3 times body mass. Following a warm-up and pre-exercise sets of 5 countermovement and 5 squat jumps, all participants performed back squats for 1 set/exercise, 3 reps/set, with 3-RM loads, followed by a rest interval of 0 seconds, 1, 2, 3, 4, or 5 minutes, followed by countermovement or squat jumps for 1 set/exercise, 5 reps/set, with the 12 combinations (6 rest intervals by 2 jumps) assigned evenly across 4 sessions in a randomized, counterbalanced design. Outcome measures included propulsive impulse, peak force, mean force, propulsion time, and rate of force development during the propulsive phase (force platforms, sampled at 1,000 Hz), analyzed as percentage change from the pre-exercise jumps. The findings demonstrated that a significant effect occurred only for countermovement jump average rate of force development: the 0-second rest interval resulted in a decrease of 11.84 ± 5.07% (effect size: 2.52), significantly lower than the increases at the 2-minute (5.23 ± 6.04%) and 5 minute (16.07 ± 7.20%) rest intervals. All other outcome measures were statistically similar across rest intervals for both jumps, with trivial to small effect sizes for propulsive impulse. Squat jump rate of force development during the first 100 and 150 ms exhibited moderate to large decreases at all rest intervals, while propulsion time increased. Note that the significant comparisons were between rest intervals rather than against pre-exercise values, and occurred between non-consecutive intervals (0 seconds compared to 2 and 5 minutes, but not 1 minute), a pattern that may imply variance rather than a consistent potentiation time course. Note that the relative strength of this sample (1.63 times body mass) was lower than the samples in which improved jump performance was demonstrated, and back squat velocities during experimental sets were significantly faster than during the 3-RM test, which may imply a smaller neuromuscular stimulus. Note that the rate of force development variables exhibited the lowest reliability of the outcome measures, so these findings should be interpreted with caution (??)

• Houlton, L. J., Moody, J. A., Bampouras, T. M., & Esformes, J. I. (2024). Acute effects of intracontrast rest after back squats on vertical jump performance during complex training. Journal of Strength and Conditioning Research, 38(11), e645-e655. https://doi.org/10.1519/JSC.0000000000004878

Following a 3-RM back squat set with 15 seconds of rest redistributed between repetitions, countermovement and squat jump propulsive impulse exhibits small significant increases (within typical error), propulsion time lengthens, mean force decreases, and rate of force development is not enhanced at total rest periods of 1 - 5 minutes, with no differences between rest periods

[Full text verified via the publisher's version of record. The published abstract states rate of force development was lower after the squat set, citing the same statistics (g = 0.703, p = 0.009) that the published results section attaches to a significantly higher value after the squat set; the results section, descriptive tables, and discussion agree with one another, and the findings below follow them. The abstract statement is a publication error and should not be cited]

Houlton et al. compared 17 recreationally trained males (age: 25.5 ± 3.0 years; relative 3-RM back squat: 1.66 ± 0.13 times body mass) with at least 1 year of resistance training experience and a back squat of at least 1.3 times body mass. The study assessed 5 total rest periods (60, 120, 180, 240, and 300 seconds), each including 30 seconds of rest redistributed between back squat repetitions (15 seconds between repetitions 1 and 2 and between repetitions 2 and 3) and a post-set rest of the remaining duration (30 to 270 seconds). Each intervention included a control condition (a pre-rest set of 5 jumps, the assigned post-set rest, and a second set of 5 jumps) and an experimental condition (back squat warm-up sets, back squats for 1 set/exercise, 3 reps/set, with 3-RM loads and 15 seconds between repetitions, the assigned post-set rest, and a final set of 5 jumps), with the 10 combinations (5 rest periods by 2 jumps) assigned evenly across 5 sessions in a randomized, counterbalanced design and a very long (10 min) rest between conditions. Outcome measures included propulsive impulse, peak force, mean force, propulsion time, and rate of force development during the propulsive phase (force platforms, sampled at 1,000 Hz), and back squat mean velocity for comparison to the 3-RM test; the repetition with the highest propulsive impulse from each set was compared across rest periods, conditions, and time points. The findings demonstrated that, for both jumps, jumps after the squat set exhibited significantly higher propulsive impulse than jumps after the rest-only control and the pre-rest jumps (effect sizes: 0.55 - 0.70), significantly lower mean force, and significantly longer propulsion time. Countermovement jump average rate of force development was significantly higher after the squat set than after the rest-only control (effect size: 0.70), with no difference from the pre-rest jumps; squat jump rate of force development during the first 100 and 150 ms decreased significantly from pre to post in both conditions, and no other rate of force development variable changed significantly. Further, no significant main effect of rest period and no rest period by condition by time interaction occurred for any outcome measure; a rest period by time interaction occurred for countermovement jump propulsive impulse, with the post-rest jumps significantly higher than the initial jumps at the 240 second rest period, collapsed across conditions (effect size: 0.63); and back squat mean velocity was significantly faster during experimental conditions than during the 3-RM test. Note that the mean differences in propulsive impulse and average rate of force development were mostly below or similar to the typical error and smallest worthwhile change of those measures, which implies maintenance rather than enhancement of jump performance. Note that the jump-rest-jump control condition demonstrated that the rest period alone did not change propulsive impulse, so the changes after the squat set can be attributed to the squat set. Note that the relative strength of this sample (1.66 times body mass) was below the range in which improved performance has been demonstrated (1.75 to 2.10 times body mass), and the faster back squat velocities during experimental conditions may imply the redistributed-rest squat sets provided a smaller neuromuscular stimulus than continuous 3-RM repetitions.

• Houlton, L. J., Moody, J. A., Bampouras, T. M., & Esformes, J. I. (2026). Acute effect of intracontrast rest redistribution within complex-contrast training set strategies on vertical jump propulsive force. Journal of Strength and Conditioning Research, 40(8), 991-1002. https://doi.org/10.1519/JSC.0000000000005428

20-meter sprint times improve significantly 4 to 20 minutes after squat, weighted plyometric, and isometric protocols, with the squat protocol fastest at 4 to 8 minutes and the isometric protocol fastest at 12 to 16 minutes; sprint times and jump height are statistically similar to pre-exercise values at 20 seconds

Piper et al. compared 13 resistance-trained college-aged participants, 10 men (age: 21 ± 2 years; relative 1-RM squat: 1.94 ± 0.16 times body weight), and 3 women (age: 20 ± 1 year; relative 1-RM squat: 1.85 ± 0.15 times body weight), with a back squat of at least 1.7 times body weight. Following 1-RM testing and familiarization during a first session, all participants completed 4 testing sessions separated by at least 48 hours, performing a control protocol (a 4 min walk), a squat protocol (back squats for 3 sets/exercise, 5 reps/set, with 87% of 1-RM loads), a plyometric protocol (weighted jumps for 3 sets/exercise, 5 reps/set, with an additional 10% of body weight), and an isometric protocol (isometric back squats at 30° for 3 sets/exercise, 1 maximal 3 second hold/set) in random order, all with long (3 min) rest between sets. Each session included a warm-up, followed by a countermovement jump (jump mat, with arm swing) and a 20-meter sprint (timing gates at 0, 10, and 20 m) at pre-exercise, 20 seconds, 4, 8, 12, 16, and 20 minutes post-protocol. The findings demonstrated a significant time effect and condition by time interaction for 0-20 meter sprint times: times at 4 to 20 minutes were significantly faster than pre-exercise and 20 second times; the squat protocol was significantly faster than the control at 4 minutes, the squat and plyometric protocols were significantly faster than the control at 8 minutes, the isometric protocol was significantly faster than the control at 12 and 16 minutes, and the isometric protocol was significantly faster than the squat protocol at 20 minutes. For 10-20 meter split times, the squat and isometric protocols were significantly faster than the plyometric and control protocols. Jump height exhibited no significant condition effect or interaction, and a time trend (p = 0.059) with jumps at 8 minutes higher than pre-exercise. Additionally, changes in sprint times across the exercise protocols correlated with relative 1-RM strength (R² = 0.45), whereas no correlation was observed during the control protocol (R² < 0.01). Note that sprint times also improved across time points during the control protocol, so the main effects of time partly reflect repeated sprinting, and the between-protocol comparisons at each time point isolate the effects of the stimuli. Note that sprint times and jump height at 20 seconds were statistically similar to pre-exercise values, and the isometric protocol exhibited the largest (non-significant) decrement at 20 seconds (??)

• Piper, A. D., Joubert, D. P., Jones, E. J., & Whitehead, M. T. (2020). Comparison of post-activation potentiating stimuli on jump and sprint performance. International Journal of Exercise Science, 13(4), 539-553. https://doi.org/10.70252/RPEZ7761

1.2 Studies Comparing Rest Intervals of 1 to 20 Minutes (Lower Body)

Countermovement jump height increases significantly following a 5-RM half squat set with a self-selected rest (approximately 6 minutes), does not change with a fixed 4-minute rest, and decreases significantly following a control protocol

do Carmo et al. compared 12 strength-trained males (age: 25.4 ± 3.6 years; 1-RM half squat: 188.7 ± 33.4 kg, approximately 2.4 times body mass). Following 6 familiarization and testing sessions (including familiarization with a 7-point perceived readiness scale), all participants performed 3 conditions in random, counterbalanced order: a fixed rest condition (a pre-exercise countermovement jump, a 4 minute rest, half back squats for 1 set/exercise, 5 reps/set, with 5-RM loads, a 4 minute rest, and a post-exercise jump), a self-selected rest condition (identical, with the post-squat rest continued until participants reported feeling fully recovered on the readiness scale), and a control condition (a pre-exercise jump, an 8 minute rest, and a post-exercise jump). Outcome measures included jump height. The findings demonstrated that jump height increased significantly from pre to post in the self-selected condition (38.2 to 40.5 cm; effect size: 0.5), did not change significantly in the fixed rest condition (38.0 to 37.7 cm), and decreased significantly in the control condition (38.4 to 37.4 cm). Post-exercise jump height was significantly higher in the self-selected condition than the fixed rest (effect size: 1.13) and control (effect size: 1.35) conditions, and the fixed rest and control conditions were statistically similar. The average self-selected rest (5:57 ± 2:44 minutes) was significantly longer than the fixed 4-minute rest. Note that the self-selected and fixed conditions differed in both rest strategy and rest duration, so the advantage cannot be attributed to self-selection alone.

• do Carmo, E. C., De Souza, E. O., Roschel, H., Kobal, R., Ramos, H., Gil, S., & Tricoli, V. (2021). Self-selected rest interval improves vertical jump postactivation potentiation. Journal of Strength and Conditioning Research, 35(1), 91-96.

Countermovement jump height increases significantly 4 to 8 minutes, but not 2 or 10 minutes, after a 5-RM back squat set in professional volleyball athletes; a single jump performed after a self-selected rest (approximately 3.5 minutes) is similarly improved

Fontanetti et al. compared 10 male professional volleyball athletes (age: 20.6 ± 1.5 years; 5-RM back squat: 146 ± 27 kg, approximately 1.6 times body mass) classified as highly trained/national level or above. Following 2 familiarization sessions and 5-RM testing (barbell back squats to 90° of knee flexion), all participants performed 3 sessions in random order: a control session (jumps at baseline and at 2, 4, 6, 8, and 10 minutes after seated rest), a fixed interval session (back squats for 1 set/exercise, 5 reps/set, with 5-RM loads, followed by jumps at 2, 4, 6, 8, and 10 minutes), and a self-selected interval session (the identical squat set, followed by a single jump once the athlete reported feeling fully recovered on a 7-point readiness scale). Outcome measures included jump height (jump mat; intraclass correlation coefficient: 0.983), with all comparisons made against the control session baseline. The findings demonstrated that jump height did not change significantly during the control session; increased significantly at 4 minutes (48.4 cm; effect size: 0.5), 6 minutes (48.3 cm; effect size: 0.4), and 8 minutes (48.2 cm; effect size: 0.4) after the squat set when compared to baseline (45.7 cm), but not at 2 or 10 minutes; and increased significantly after the self-selected interval (49.6 cm; effect size: 0.7), with no significant difference between the self-selected jump and the best fixed-interval jump (49.2 cm). The self-selected interval averaged 213 ± 63 seconds (approximately 3.5 minutes). Additionally, 8 of 10 athletes exhibited improvements exceeding the smallest worthwhile change, and 5 of those 8 self-selected an interval within their individual improvement window. Note that all comparisons used the control session's baseline jump as the reference value, a limitation identified in the source (??)

• Fontanetti, G., Barreto, R. V., Junior, R. C., Marangoni, V., Denadai, B. S., Greco, C. C., & Lima, L. C. R. (2025). The use of the self-selected rest interval method is as effective for optimizing postactivation performance enhancement in elite athletes as employing the best fixed rest interval. Journal of Strength and Conditioning Research, 39(1), 10-15. https://doi.org/10.1519/JSC.0000000000004939

Cycle sprint average power output increases significantly following 10 single back squats with 90% of 1-RM loads and a 5 minute rest, when compared to a 20 minute rest and a control session; peak power is statistically similar across sessions

Smith et al. compared 9 males (age: 25.11 ± 1.16 years; values reported as mean ± standard error) with at least 1 year of systematic weight training and a back squat of at least 1.5 times body weight. Following a familiarization and 1-RM testing session, all participants performed 3 test sessions in randomized, balanced order: a control session (a cycling warm-up followed by a maximal 10 second sprint cycle test) and 2 squat sessions (squat warm-up sets progressing from 40 to 90% of 1-RM loads, followed by parallel back squats for 10 sets/exercise, 1 rep/set, with 90% of 1-RM loads and moderate (2 min) rest between sets, followed by a 5 or 20 minute rest, followed by the sprint cycle test). Outcome measures included average power and peak power (absolute, relative to body weight, and relative to lean body weight) and the fatigue index. The findings demonstrated that average power and average power relative to body weight were significantly higher during the 5-minute session when compared to the control and 20-minute sessions; peak power and peak power relative to body weight exhibited increases during the squat sessions that failed to reach statistical significance, and all other outcome measures were statistically similar across sessions (??)

• Smith, J. C., Fry, A. C., Weiss, L. W., Li, Y., & Kinzey, S. J. (2001). The effects of high-intensity exercise on a 10-second sprint cycle test. Journal of Strength and Conditioning Research, 15(3), 344-348.

10- and 30-meter sprint times improve significantly 5 minutes, but not 3 minutes, after 10 single back squats with 90% of 1-RM loads

Chatzopoulos et al. compared 15 male amateur team-sport athletes (age: 22 ± 2 years; 1-RM back squat: 151 ± 12 kg, approximately 1.7 times body mass) with at least 5 years of sport experience, regular resistance training, and a back squat of more than 1.5 times body mass. Following familiarization and 1-RM estimation during a first session, all participants completed 2 testing sessions, each including a warm-up and 3 maximal 30-meter sprint trials (best trial analyzed; photocells at 0, 10, and 30 m), followed by back squats for 10 sets/exercise, 1 rep/set, with 90% of 1-RM loads and long (3 min) rest between sets, followed by a repeated sprint test 3 minutes after the squats (second session) or 5 minutes after the squats (third session). Outcome measures included 0-10 and 0-30 meter sprint times. The findings demonstrated that sprint times 3 minutes after the squats were statistically similar to pre-squat times, and sprint times 5 minutes after the squats were significantly faster than pre-squat times for both distances (0-10 m: 1.84 compared to 1.89 seconds; 0-30 m: 4.43 compared to 4.51 seconds). The pre-squat times of the 2 sessions were statistically similar. Note that the 3 minute and 5 minute sessions were completed in a fixed order rather than randomized.

• Chatzopoulos, D. E., Michailidis, C. J., Giannakos, A. K., Alexiou, K. C., Patikas, D. A., Antonopoulos, C. B., & Kotzamanidis, C. M. (2007). Postactivation potentiation effects after heavy resistance exercise on running speed. Journal of Strength and Conditioning Research, 21(4), 1278-1281.

Countermovement jump height, concentric peak power, and concentric peak velocity increase significantly at 4 to 20 minutes after a set of flywheel half squats at the power-maximizing inertial load, when compared to pre-exercise values and a repeated-jump control condition

Maroto-Izquierdo et al. compared 20 physically active male university students (age: 23.4 ± 2.9 years; body mass: 69.4 ± 15.4 kg) with at least 1 year of flywheel training experience. Following familiarization sessions and a session to establish the inertial load that maximized power output, all participants performed 2 testing sessions in a random, counterbalanced order, separated by 72 hours. Each session included a warm-up and a pre-exercise countermovement jump (3 min after the warm-up), followed by the experimental condition (flywheel half squats for 1 set/exercise, 6 maximal reps/set, with the individualized power-maximizing moment of inertia (0.083 ± 0.03 kg·m²)) or the control condition (maximal countermovement jumps for 1 set/exercise, 6 reps/set, in place of the flywheel set), followed by countermovement jumps at 4, 8, 12, 16, and 20 minutes. Outcome measures included jump height, concentric peak power, and concentric peak velocity. The findings demonstrated significant condition and condition by time effects for all 3 outcome measures, with significant improvements at 4 to 20 minutes occurring only after the flywheel condition (effect size range: 0.10 - 1.34), and between-condition differences significant at 4 to 20 minutes and statistically similar at pre-exercise. Note that no time point earlier than 4 minutes was assessed, so this study cannot contribute to comparisons of immediate rest intervals. Note that the conditioning set used the power-maximizing inertial load rather than heavy strength loads, the control condition included repeated maximal jumps rather than rest, and the sample included physically active students rather than trained athletes, so the sustained improvement may not generalize to heavy strength sets in athletic samples.

• Maroto-Izquierdo, S., Bautista, I. J., & Martín Rivera, F. (2020). Post-activation performance enhancement (PAPE) after a single bout of high-intensity flywheel resistance training. Biology of Sport, 37(4), 343-350. https://doi.org/10.5114/biolsport.2020.96318

Isometric rate of force development decreases significantly at 2 and 10 minutes and increases significantly at 15 and 20 minutes after 5 single back squats with 1-RM loads, and increases significantly at 2 minutes after 5 single back squats with the maximal power load; jump height increases significantly following both protocols, peaking at 20 minutes after the 1-RM protocol

Gilbert and Lees compared 15 male athletes (age: 24.3 ± 3.3 years; 1-RM back squat: 236.9 ± 21.2 kg, approximately 2.1 times body mass) recruited as experienced strength trainers. Following familiarization and 2 load-determination sessions (1-RM, and the load maximizing power output), all participants performed 3 conditions in counterbalanced order with at least 7 days between conditions: a strength condition (back squats for 5 sets/exercise, 1 rep/set, with 1-RM loads and very long (5 min) rest between reps), a power condition (back squats for 5 sets/exercise, 1 rep/set, with the maximal power load), and a control condition (light activity for the matched duration, approximately 25 minutes, with no lifting). Outcome measures included maximum isometric force and isometric rate of force development of the dominant leg (seated, hip and knee at 90°; sampled at 1,000 Hz) and countermovement jump height, assessed before and at 2, 10, 15, 20, and 60 minutes after each condition. The findings demonstrated no significant changes in isometric force or rate of force development during the control condition, and no differences between conditions at pre-exercise. Following the strength condition, isometric rate of force development decreased significantly at 2 minutes (p = 0.0023) and 10 minutes (p = 0.035) when compared to pre-exercise values (4,867 ± 881 N/s), increased significantly at 15 minutes (p = 0.021) and 20 minutes (p = 0.006; a peak increase of 11.8%), and was statistically similar to pre-exercise values at 60 minutes. Following the power condition, isometric rate of force development increased significantly at 2 minutes (an increase of 6.7%), and all later time points were statistically similar to pre-exercise values. Jump height increased significantly following both conditions, peaking at 20 minutes following the strength condition. Note that the results text contains an internal inconsistency at the 20-minute time point (reported as both a significant increase and not significantly different in consecutive sentences); the temporal profile and the reported statistics support the significant increase.

• Gilbert, G., & Lees, A. (2005). Changes in the force development characteristics of muscle following repeated maximum force and power exercise. Ergonomics, 48(11-14), 1576-1584. https://doi.org/10.1080/00140130500101163

Countermovement jump height and mean power output do not change at any time point following a velocity-loss-limited back squat set with 80% of 1-RM loads; jump height is significantly lower than the control condition at 4 to 8 minutes, and the control condition decreases significantly at 10 minutes

Krzysztofik et al. compared 16 resistance-trained female volleyball players (age: 24 ± 5 years; relative 1-RM back squat: 1.45 ± 0.19 times body mass). Following a familiarization and 1-RM testing session, all participants performed 2 conditions in random order within 3 weeks: a squat condition (back squats for 1 set/exercise, reps until a 10% loss of mean velocity (4.1 ± 1.3 reps completed), with 80% of 1-RM loads) and a control condition (no conditioning activity), with countermovement jumps at baseline (5 min before) and at 2, 4, 6, 8, and 10 minutes after. Outcome measures included jump height and relative mean power output. The findings demonstrated that jump height and mean power did not change significantly at any time point during the squat condition; jump height and mean power decreased significantly at 10 minutes during the control condition; and jump height during the control condition was significantly higher than the squat condition at 4, 6, and 8 minutes. Additionally, 9 of 16 participants (56%) exhibited a positive individual response to the squat condition (best post-squat jump: an increase of 6.1%; effect size: 0.43), and responders and non-responders were statistically similar in relative strength, repetitions completed, and training experience. Note that the squat set was limited to approximately 4 reps by the velocity-loss cutoff, a lower training volume than the protocols in this section that resulted in significant improvements.

• Krzysztofik, M., Kalinowski, R., Trybulski, R., Filip-Stachnik, A., & Stastny, P. (2021). Enhancement of countermovement jump performance using a heavy load with velocity-loss repetition control in female volleyball players. International Journal of Environmental Research and Public Health, 18(21), 11530. https://doi.org/10.3390/ijerph182111530

Cycle ergometer power output is statistically similar following a single back squat set with 85% of 1-RM loads and 5 to 20 minutes of rest when compared to a control session; each participant's best rest interval results in significantly higher peak power than control, and stronger participants exhibit their best performance at shorter rest intervals

Jo et al. compared 12 recreationally trained males (age: 23 ± 1 year; relative 1-RM back squat: 1.4 ± 0.1 times body mass) with at least 1 year of back squat experience. All participants performed a control session (a 10 minute cycling warm-up, a moderate (2 min) rest, back squats for 1 set/exercise, 5 reps/set, with 50% of 1-RM loads, a moderate (2 min) rest, and a 30 second Wingate test) and 4 experimental sessions (back squats for 1 set/exercise, 5 reps/set, with 85% of 1-RM loads, followed by 5, 10, 15, or 20 minutes of rest, followed by the Wingate test) in random order, separated by 48 to 72 hours. Outcome measures included peak power, relative peak power, mean power, and the fatigue index. The findings demonstrated that no outcome measure differed significantly between the control and any experimental session. However, each participant's highest values across the 4 rest intervals were significantly higher than control values (peak power: an increase of 7.1%; relative peak power: 7.1%; fatigue index: 8.9%), and the rest interval that produced each participant's highest values correlated negatively with relative 1-RM strength (r = -0.771), with stronger participants exhibiting their best performance after shorter rest intervals. Note that the protocol included a single set of 5 reps, a lower training volume than the protocols in this section that resulted in significant improvements at the group level.

• Jo, E., Judelson, D. A., Brown, L. E., Coburn, J. W., & Dabbs, N. C. (2010). Influence of recovery duration after a potentiating stimulus on muscular power in recreationally trained individuals. Journal of Strength and Conditioning Research, 24(2), 343-347.

Total kicks increase significantly following a complex half squat and jump protocol with a 10-minute rest when compared to a control condition; countermovement jump height and kick impact are statistically similar across strength, plyometric, and complex protocols at rest intervals of 5 minutes, 10 minutes, and self-selected (approximately 5.5 minutes)

da Silva Santos et al. compared 11 black-belt taekwondo athletes (age: 20.3 ± 5.2 years). All participants performed 1 control condition and 9 experimental conditions (3 conditioning activities by 3 rest intervals) in random order, 2 conditions per day across 5 days with a 90-minute rest between conditions. The conditioning activities included a strength condition (half squats for 3 sets/exercise, 1 rep/set, with 95% of 1-RM loads), a plyometric condition (vertical jumps over a 40-cm barrier for 3 sets/exercise, 10 reps/set, with short (30 sec) rest between sets), and a complex condition (3 sets/exercise of half squats for 2 reps with 95% of 1-RM loads immediately followed by 4 vertical jumps over the barrier), each followed by a 5 minute, 10 minute, or self-selected (328 ± 139 seconds) rest before a countermovement jump and the Frequency Speed Kick Test. Outcome measures included jump height, average kick impact, and total kicks. The findings demonstrated that total kicks following the complex condition with the 10 minute rest (23 ± 5 kicks) were significantly higher than the control condition (19 ± 3), the strength condition with the self-selected rest (18 ± 2), and the plyometric condition with the 5 minute rest (18 ± 3), and jump height and kick impact were statistically similar across all conditions. Note that the omnibus test for total kicks failed to reach statistical significance (p = 0.239), and the significant post hoc comparisons should be interpreted with caution (??)

• Santos, J. F. da S., Valenzuela, T. H., & Franchini, E. (2015). Can different conditioning activities and rest intervals affect the acute performance of taekwondo turning kick? Journal of Strength and Conditioning Research, 29(6), 1640-1647. https://doi.org/10.1519/JSC.0000000000000808

Loaded half squat jump average power output increases significantly at 10 minutes, but not 5 minutes, after 2 sets of back half squats with 4-RM loads or 2 sets of half squat jumps, and decreases significantly at 10 minutes during a control protocol

Krčmár et al. compared 11 strength-trained athletes (6 male ice hockey players, 3 female volleyball players, 1 rower, and 1 karate athlete; age: 22 ± 1.8 years) with at least 4 years of training experience. All participants performed 3 trials in random order, separated by at least 48 hours: a jump condition (loaded half squat jumps for 2 sets/exercise, 4 reps/set, with the load maximizing average power), a squat condition (back half squats (to 90° of knee flexion) for 2 sets/exercise, 4 reps/set, with 4-RM loads), and a control condition (no conditioning activity). Outcome measures included maximum average power output during loaded half squat jumps at the power-maximizing load, measured before and at 5 and 10 minutes after each condition. The findings demonstrated that average power did not change significantly at 5 minutes in any condition, increased significantly at 10 minutes following the jump condition (1,496 to 1,552 W; an increase of approximately 3.7%) and the squat condition (1,501 to 1,556 W; approximately 3.7%), and decreased significantly at 10 minutes during the control condition (1,517 to 1,507 W). At 10 minutes, both conditioning conditions were significantly higher than the control condition (??)

• Krčmár, M., Šimonek, J., & Vasiľovský, I. (2015). The acute effect of lower-body training on average power output measured by loaded half-squat jump exercise. Acta Gymnica, 45(3), 103-111. https://doi.org/10.5507/ag.2015.014

1.3 Upper Body Studies (Bench Press)

Bench press throw mean power output increases significantly following a bench press set with 65% of 1-RM loads and a 4-minute rest, and following a set with 85% of 1-RM loads and an 8-minute rest; power decreases significantly (approximately 1%) following the 85% set with a 4-minute rest, and is statistically similar following the 65% set with an 8-minute rest

Liossis et al. compared 9 amateur combat and martial arts athletes (age: 26.1 ± 3.4 years; bench press 1-RM: 83.9 ± 8.4 kg). Following 1-RM testing and familiarization with Smith machine bench press throws, all participants performed 4 protocols in random order on non-consecutive days, each including a warm-up, pre-conditioning bench press throws for 1 set/exercise, 3 reps/set, with 30% of 1-RM loads, a conditioning set of bench press for 1 set/exercise, 5 reps/set, with 65% or 85% of 1-RM loads, a 4 or 8 minute rest, and an identical set of post-conditioning throws. Outcome measures included mean power output during the throws. The findings demonstrated that mean power increased significantly following the 65% of 1-RM set with the 4 minute rest (530.3 to 556.8 W; an increase of 26.4 ± 13.8 W) and the 85% of 1-RM set with the 8 minute rest (534.2 to 554.7 W; an increase of 20.9 W), decreased significantly following the 85% of 1-RM set with the 4 minute rest (541.7 to 537.1 W; a decrease of 4.6 ± 5.0 W, approximately 1%), and did not change significantly following the 65% of 1-RM set with the 8 minute rest (538.0 to 530.8 W). Additionally, the differences in power output between the 4 protocols were statistically significant, with the largest increase following the 65% of 1-RM set with the 4-minute rest (??)

• Liossis, L. D., Forsyth, J., Liossis, C., & Tsolakis, C. (2013). The acute effect of upper-body complex training on power output of martial art athletes as measured by the bench press throw exercise. Journal of Human Kinetics, 39, 167-175. https://doi.org/10.2478/hukin-2013-0079

Bench press concentric mean power output increases significantly following a 1-RM bench press and a 7-minute rest when compared to a 1-minute rest and a control condition; concentric peak power and eccentric power are statistically similar across rest intervals of 1 to 7 minutes

Ferreira et al. compared 11 healthy males (age: 25 ± 4 years; bench press 1-RM: 76 ± 19 kg). Following 2 control sessions (1-RM testing, and a baseline power session of bench press for 1 set/exercise, 6 reps/set, with 50% of 1-RM loads, the load producing the highest power output for this sample), all participants performed 4 experimental sessions in random order on different days, each including a 1-RM bench press, followed by a 1, 3, 5, or 7 minute rest, followed by the identical power set. Outcome measures included concentric and eccentric mean and peak power (linear position transducer; intraclass correlation coefficient: 0.95). The findings demonstrated that a significant difference occurred only for concentric mean power: the 7 minute condition was significantly higher than the control and 1 minute conditions, and the 3 and 5 minute conditions were statistically similar to all conditions. Concentric peak power and eccentric mean and peak power were statistically similar across all conditions (??)

• Ferreira, S. L., Panissa, V. L., Miarka, B., & Franchini, E. (2012). Postactivation potentiation: Effect of various recovery intervals on bench press power performance. Journal of Strength and Conditioning Research, 26(3), 739-744. https://doi.org/10.1519/JSC.0b013e318225f371

Section 2: Session Structure - Exercise Order, Alternating Pairs, and Circuits

Session jump squat height, peak power, and maximal force are statistically similar following power-first, strength-first, and alternating orders of jump squats and back squats; peak power during the 1st jump squat set is significantly lower following the strength-first order, and participants with higher relative strength exhibited improved performance with the alternating order

Duthie et al. compared 11 national- and international-level female field hockey and softball athletes (age: 23.7 ± 3.2 years; predicted 1-RM half squat: 1.98 ± 0.18 times body weight) with more than 2 years of high-intensity resistance training experience. All participants performed a power-first session (all jump squat sets before back squat sets), a strength-first session (all back squat sets before jump squat sets), and an alternating session (alternating sets of the 2 exercises) in random order, 1 session/protocol, with 3 to 5 days between sessions. The exercise protocols included Smith machine jump squats for 3 sets/exercise, 4 reps/set, with 30% of predicted 1-RM loads, and Smith machine half squats (to 90° of knee flexion) for 3 sets/exercise with 3-RM loads. Outcome measures included jump height, peak power, and maximal force during each jump squat set, performed from a stationary 90° knee angle without a countermovement (force platform and position transducer). The findings demonstrated that session means for jump height, peak power, and maximal force were statistically similar following all 3 orders; peak power during the 1st set was significantly lower following the strength-first order than the power-first order (2,768 compared to 2,842 W); and the change in peak power (r = 0.66) and maximal force (r = 0.76) with the alternating order correlated significantly with predicted 1-RM strength, with the stronger half of the sample (predicted 1-RM: 139 kg) exhibiting increased maximal force with the alternating order and the weaker half (116 kg) exhibiting a decrease.

• Duthie, G. M., Young, W. B., & Aitken, D. A. (2002). The acute effects of heavy loads on jump squat performance: An evaluation of the complex and contrast methods of power development. Journal of Strength and Conditioning Research, 16(4), 530-538.

Countermovement jump height, peak force, rate of force development, peak power, and squat peak force are statistically similar whether all squat repetitions precede or follow all countermovement jumps in a session; 5- and 20-meter sprint times are significantly slower after 3 countermovement jumps and statistically similar after 3 squats with 3-RM loads

Deutsch and Lloyd compared 8 male university first-team rugby union players (age: 20.4 ± 1.7 years) with 3.3 ± 0.9 years of resistance training experience, tested during the competitive season. All participants performed a control session for each exercise, then a squat-first session and a jump-first session in a randomized crossover design across 5 test days. The exercise protocols included barbell parallel back squats (to 90° of knee flexion) for 3 single repetitions with 3-RM loads and long (3 min) rest between repetitions, and countermovement jumps for 3 single repetitions with short (1 min) rest between repetitions, with a very long (10 min) recovery between the 2 exercises; 20-meter sprints (3 trials, long (3 min) rest) were performed 10 minutes after each single-exercise control session. Outcome measures included 5-, 10-, and 20-meter sprint times, countermovement jump height, peak force, peak rate of force development, amortization phase duration, and peak power, and squat peak force and peak rate of force development (force platform). The findings demonstrated that all countermovement jump and squat measures were statistically similar between the squat-first and jump-first orders; squat peak force was significantly higher during both order sessions than the control session; 5-meter (1.23 compared to 1.13 s) and total 20-meter (3.29 compared to 3.18 s) sprint times were significantly slower after the countermovement jumps than the control; and sprint times after the squats were statistically similar to the control. Note that the sprint comparisons followed the single-exercise sessions rather than the 2-exercise order sessions. Note that the sample included 8 participants, and non-significant comparisons should be interpreted with caution.

• Deutsch, M., & Lloyd, R. (2008). Effect of order of exercise on performance during a complex training session in rugby players. Journal of Sports Sciences, 26(8), 803-809. https://doi.org/10.1080/02640410801942130

Countermovement jump peak power during a session is significantly lower when sets of squats with 5-RM loads are alternated with jump sets, or when isometric squats precede all jump sets, than when jumps are performed alone; sessions performing all squat sets with 5-RM loads before jump sets are statistically similar to jumps alone

Talpey et al. compared 18 recreationally trained male team-sport athletes (age: 21.1 ± 3.3 years; relative 5-RM back half-squats: 1.5 ± 0.2 times body mass) with at least 1 year of resistance training experience, who were tested during the off-season. Following a 4-week familiarization period, all participants performed 5 sessions in random order with at least 72 hours between sessions: jumps alone, all squat sets before all jump sets, alternating squat and jump sets, all isometric squat sets before all jump sets, and alternating isometric squat and jump sets. The exercise protocols included countermovement jumps for 3 sets/exercise, 4 reps/set, with moderate (2 min) rest after each jump set, Smith machine half back squats for 4 reps/set with 5-RM loads, and single 5-second near-maximal isometric back squats (110° knee angle), with very long (4 min) rest after each squat or isometric set. Outcome measures included peak power during each jump set (force platform and position transducer). The findings demonstrated that session mean peak power was significantly lower during the alternating squat session and the isometric-first session than during jumps alone, and statistically similar during the squat-first and alternating isometric sessions; peak power during the 1st set was significantly lower only during the isometric-first session; peak power during the 2nd set was significantly lower during the alternating squat, alternating isometric, and isometric-first sessions; peak power during the 3rd set was statistically similar across all sessions; and changes relative to jumps alone did not correlate with absolute or relative strength. Note that jump power was never significantly lower during the squat-first session, in which every squat set was followed by 4 minutes of rest.

• Talpey, S. W., Young, W. B., & Saunders, N. (2014). The acute effects of conventional, complex, and contrast protocols on lower-body power. Journal of Strength and Conditioning Research, 28(2), 361-366.

Vertical jump height and power output increase significantly and similarly following 4 weeks of complex training (resistance and plyometric exercises alternated within each session) and compound training (resistance and plyometric exercises in separate weekly sessions), with significant improvements beginning after the 3rd week of training

Mihalik et al. compared 31 collegiate club volleyball players (11 males and 20 females), assigned by matching gender and pre-training jump height to a complex training group (n = 15; age: 20.3 ± 2.2 years) or a compound training group (n = 16; age: 20.9 ± 2.4 years) for 4 weeks, 2 sessions/week. The complex training group alternated resistance and plyometric exercises within each session, in order: back squats, depth jumps (30 cm), single-leg lunges, split squat jumps, deadlifts, and double-leg bounds, for 3 sets/exercise, 6 reps/set; the compound training group performed the 3 resistance exercises in 1 weekly session and the 3 plyometric exercises in the other, for 6 sets/exercise, 6 reps/set, matching total weekly work. The exercise protocols included free-weight resistance exercises with 60% of 1-RM loads performed as explosively as possible, with short (1 min) rest between sets and moderate (2 min) rest between exercises. Outcome measures included countermovement vertical jump height with an arm swing (jump-and-reach device) and power output estimated from jump height and body mass, assessed at pre-training and after each week of training. The findings demonstrated that jump height and power output increased significantly in both groups after the 3rd week (complex: approximately 5.4 and 4.8%; compound: approximately 9.1 and 7.5%) and increased further after the 4th week, with no significant differences between groups and no significant difference in the rate of improvement between groups or between sexes. Note that power output was estimated from jump height and body mass rather than measured directly, so the 2 outcomes are not independent. Note that the compound format separated resistance and plyometric exercises into different sessions, so this study compares within-session alternation to separate sessions rather than to within-session strength and power blocks.

• Mihalik, J. P., Libby, J. J., Battaglini, C. L., & McMurray, R. G. (2008). Comparing short-term complex and compound training programs on vertical jump height and power output. Journal of Strength and Conditioning Research, 22(1), 47-53. https://doi.org/10.1519/JSC.0b013e31815eee9e

Following 6 weeks, countermovement jump height and overhead throw distance increase significantly only with strength and power exercises on alternating days; 1-RM strength increases significantly with both alternating-day and paired-session training, with greater percentage increases in bench press and box squat strength and increases in muscle fiber cross-sectional area only following paired sessions

Stasinaki et al. compared 23 moderately trained male physical education students (age: 21.9 ± 1.9 years) with no systematic resistance training during the previous year, randomly assigned to a control group (n = 7), an alternating-day group (n = 7 of 9 enrolled), or a paired-session group (n = 9) for 6 weeks, 3 sessions/week. Both training groups performed incline leg press, bench press, and Smith machine box squats (to 90° of knee flexion) with 6-RM loads (approximately 85% of 1-RM), leg press throws, bench press throws, and squat jumps with 30% of 1-RM loads, and depth jumps from 40 to 45 cm; the alternating group performed the strength exercises for 4 sets/exercise, 6 reps/set, on 1 training day and the power exercises for 4 sets/exercise, 8 reps/set, on the next; the paired group performed 2 sets/exercise every session, with each strength set followed after a long (3 min) rest by a set of the matched power exercise; total weekly repetitions were equal between groups (1,512 each). Outcome measures included countermovement jump height, backward overhead throw distance, 1-RM strength (leg press, bench press, and box squat), vastus lateralis and gastrocnemius thickness, fascicle angle, and fascicle length, and vastus lateralis fiber type composition and cross-sectional area. The findings demonstrated that fiber type composition and body mass did not change in either group; jump height (4.0%) and throw distance (9.2%) increased significantly only in the alternating group, and the between-group difference in the jump height change did not reach statistical significance; 1-RM strength increased significantly in both groups for all 3 exercises, with significantly greater percentage increases in box squat (35.6 compared to 27.2%) and bench press (18.4 compared to 5.0%) strength in the paired group; vastus lateralis thickness and fascicle angle increased significantly in both groups, with a greater percentage increase in thickness in the alternating group; gastrocnemius fascicle angle increased significantly in both groups (greater in the paired group), and gastrocnemius fascicle length decreased significantly only in the paired group; and type I, IIA, and IIX fiber cross-sectional areas increased significantly only in the paired group. Note that the alternating format separated strength and power exercises onto different training days, so this study compares paired sessions to separate days rather than to within-session blocks. Note that the paired group was significantly stronger at pre-training and trained with higher absolute loads, and sessions contained 4 compared to 2 sets/exercise, so format is confounded with baseline strength and session structure (??)

• Stasinaki, A. N., Gloumis, G., Spengos, K., Blazevich, A. J., Zaras, N., Georgiadis, G., Karampatsos, G., & Terzis, G. (2015). Muscle strength, power, and morphologic adaptations after 6 weeks of compound vs. complex training in healthy men. Journal of Strength and Conditioning Research, 29(9), 2559-2569. https://doi.org/10.1519/JSC.0000000000000917

Back squat, Romanian deadlift, and calf raise 1-RM strength and thigh and calf girths increase significantly and similarly following 6 weeks of resistance training, plyometric training, or complex training pairing the 2 within the same set

MacDonald et al. compared 30 recreationally trained males (age: 21.73 ± 3.40 years) with at least 6 months of regular resistance training, randomly assigned to a resistance training group (n = 11), a plyometric training group (n = 9), or a complex training group (n = 10) for 6 weeks, 2 sessions/week, with 2 to 3 days between sessions. The resistance group performed machine-guided high bar back squats, Romanian deadlifts, and standing calf raises for 3 sets/exercise, 3 to 6 reps/set, with 75 to 90% of 1-RM loads in the 1st weekly session and 45 to 67% of 1-RM loads (including speed squats) in the 2nd; the plyometric group performed lateral jumps, depth jumps (30.5 progressing to 45.7 cm), and box jumps for 3 sets/exercise, 3 to 7 reps/set; and the complex group performed both programs, with each resistance exercise paired in the same set with its biomechanically similar jump, rest within pairs of up to 30 seconds in weeks 2 to 4 and 3 minutes in weeks 6 to 8, long (3 min) rest between sets, and 4 minutes between exercises. Outcome measures included 1-RM strength for the 3 resistance exercises, quadriceps and triceps surae girths, body mass, and body fat percentage at pre-training, 5 weeks, and 9 weeks. The findings demonstrated that 1-RM strength for all 3 exercises and both girth measures increased significantly across time points in all groups, with no significant differences between groups at any time point; body mass increased significantly only in the resistance group at 5 weeks (approximately 1.4 kg) and was statistically similar to pre-training values at 9 weeks; and body fat percentage increased significantly in the resistance and plyometric groups by 9 weeks and did not change significantly in the complex group. Note that girth increases were 0.1 to 1.78 cm, and diet was not controlled. Note that the complex group's training volume was the sum of the other 2 programs, so format and volume are confounded.

• MacDonald, C. J., Lamont, H. S., & Garner, J. C. (2012). A comparison of the effects of 6 weeks of traditional resistance training, plyometric training, and complex training on measures of strength and anthropometrics. Journal of Strength and Conditioning Research, 26(2), 422-431. https://doi.org/10.1519/JSC.0b013e318220df79

Countermovement jump height, peak power, and relative peak power do not change following 6 weeks of resistance, plyometric, or complex training; jump peak ground reaction force increases significantly following plyometric and complex training, and broad jump distance increases significantly following plyometric training only

MacDonald et al. compared 34 recreationally trained males (resistance: n = 13; plyometric: n = 11; complex: n = 10; age: 20.3 to 22.5 years across groups) with at least 6 months of regular resistance training, using the same 6-week, 2 sessions/week protocols as the companion strength report (complex pairs within the same set, up to 30 seconds between paired exercises in weeks 2 to 4 and 3 minutes in weeks 6 to 8). Outcome measures included countermovement jump height, peak ground reaction force, peak power, peak power relative to body mass and to fat-free mass, and broad jump distance and peak ground reaction force at pre-training, 5 weeks, and 9 weeks. The findings demonstrated that no measure differed significantly between groups at any time point; jump height, peak power, and peak power relative to body mass did not change significantly in any group; countermovement jump peak ground reaction force increased significantly in the plyometric and complex groups at 5 weeks; peak power relative to fat-free mass increased significantly in the plyometric group from 5 to 9 weeks; and broad jump distance increased significantly in the plyometric group at 9 weeks. Note that this report and the companion strength report describe the same protocols from the same laboratory with different group sizes (34 compared to 30 participants); sample overlap is unstated, so count these samples cautiously.

• MacDonald, C. J., Lamont, H. S., Garner, J. C., & Jackson, K. (2013). A comparison of the effects of six weeks of traditional resistance training, plyometric training, and complex training on measures of power. Journal of Trainology, 2, 13-18.

Back squat 1-RM strength, standing vertical jump height, and countermovement jump height improve significantly and similarly following 9 weeks of block training with all squat sets performed before jump squat sets or the reverse; running vertical jump height improves significantly more when squat sets are performed first, and 20-meter sprint times do not improve significantly in either group

Talpey et al. compared 20 recreationally trained male team-sport athletes (age: approximately 21 years; relative 1-RM half back squat: approximately 1.7 times body weight) with at least 1 year of resistance training experience, randomly assigned to a squat-first group (n = 9) or a jump-squat-first group (n = 11) for 9 weeks, 2 sessions/week, with a 2-week break after week 4. Both groups performed Smith machine half back squats (to 90° of knee flexion) for 3 to 4 sets/exercise, 3 to 6 reps/set, with 3-RM to 8-RM loads and very long (4 min) rest between sets, and jump squats for 3 to 4 sets/exercise, 4 reps/set, with body weight (the load that maximized power output at pre-testing) and long (3 min) rest between sets, with all sets of the assigned 1st exercise completed before the 2nd. Outcome measures included 1-RM half squat, 20-meter sprint times (with 0-10 and 15-20 m splits), standing vertical jump height, running vertical jump height, and countermovement jump height, velocity, force, and power at body weight and with 10 and 20% of 1-RM loads. The findings demonstrated that 20-meter sprint times did not improve significantly in either group; 1-RM strength (24.4 compared to 23.3%) and standing vertical jump height (9.6 compared to 6.7%) improved significantly and similarly in both groups; unloaded countermovement jump height improved significantly in both groups; and running vertical jump height improved significantly more in the squat-first group (5.6%), with no significant change in the jump-squat-first group. Note that the 15- to 20-meter split improved significantly within the squat-first group only, without a significant between-group interaction. Note that both orders separated all sets with 3 to 4 minutes of rest, so this study compares block orders rather than paired supersets.

• Talpey, S. W., Young, W. B., & Saunders, N. (2016). Is nine weeks of complex training effective for improving lower body strength, explosive muscle function, sprint and jumping performance? International Journal of Sports Science & Coaching, 11(5), 736-745.

Half squat 1-RM strength and countermovement jump height increase significantly and similarly following 8 weeks of strength-first, plyometric-first, or alternating set sequences of half squats and drop jumps; 10- and 20-meter sprint speed decreases significantly only with the plyometric-first sequence

Kobal et al. compared 27 elite male under-20 soccer players (age: 18.9 ± 0.6 years) from the same professional club during the competitive season, with concurrent technical-tactical soccer training and 1 official match per week, and no previous participation in a systematic resistance training program. Participants were stratified into quartiles by countermovement jump height and randomly assigned within quartiles to a strength-first group (all half squat sets before drop jump sets), a plyometric-first group (all drop jump sets before half squat sets), or an alternating group (half squat and drop jump sets alternated set by set) (n = 9/group) for 8 weeks, 2 sessions/week, with 48 to 72 hours between sessions. All groups performed back half squats and drop jumps with equal total volume and relative intensity, periodized from 3 sets/exercise, 10 reps/set, with 60% of 1-RM loads and 3 sets/exercise, 12 reps/set, from 30 cm in weeks 1 to 2, to 5 sets/exercise, 6 reps/set, with 80% of 1-RM loads and 5 sets/exercise, 10 reps/set, from 45 cm in weeks 5 to 6, with a reduced-volume taper in weeks 7 to 8 and long (3 min) rest between all sets and exercises. Outcome measures included Smith machine half squat 1-RM strength, countermovement jump height (contact platform), 10- and 20-meter sprint speed, and 505 agility test time at pre-training, 4 weeks, and 9 weeks (1 week after the final session). The findings demonstrated that half squat 1-RM strength (48.6, 46.3, and 53%) and countermovement jump height (13, 14.2, and 14.7%) increased significantly in all 3 groups, with no significant differences between groups at any time point; jump height gains occurred from pre-training to 4 weeks without further significant change; 10-meter (7%) and 20-meter (6%) sprint speed decreased significantly only in the plyometric-first group, at both 4 and 9 weeks, with a 3% decrease in the strength-first group that did not reach statistical significance despite large effect sizes (-1.86 and -1.43) and trivial changes in the alternating group; and agility was statistically similar to pre-training values in all groups. Note that 3 minutes of rest separated all sets in every group, including between each half squat and drop jump set in the alternating sequence. Note that participants trained concurrently for soccer with weekly matches and had no systematic resistance training background, which may explain the large strength gains, and the study did not include a non-training control group (??)

• Kobal, R., Loturco, I., Barroso, R., Gil, S., Cuniyochi, R., Ugrinowitsch, C., Roschel, H., & Tricoli, V. (2017). Effects of different combinations of strength, power, and plyometric training on the physical performance of elite young soccer players. Journal of Strength and Conditioning Research, 31(6), 1468-1476. https://doi.org/10.1519/JSC.0000000000001609

Oxygen consumption, heart rate, blood lactate, energy cost, and excess post-exercise oxygen consumption are significantly higher during and after a high-intensity resistance circuit performing different-muscle-group exercises between sets than during a traditional set structure with the same exercises, loads, and volume, in approximately one third of the session time

Marín-Pagán et al. compared 10 amateur male field soccer players (age: 23.1 ± 3.8 years; maximal oxygen consumption: 58.2 ± 1.9 mL/kg/min) with at least 3 years of competition experience. Following a familiarization and 6-RM load determination session and an incremental treadmill test, all participants performed a traditional strength session and a high-intensity resistance circuit session in random, counterbalanced, crossover order, with 72 hours between sessions. Both sessions included the same 6 exercises (pec deck, knee extension, elbow flexion, knee flexion, lat pulldown, and ankle extension) for 3 sets/exercise, reps-to-failure/set, with 6-RM loads, organized as 2 blocks of 3 exercises with a 5-minute rest between blocks; the traditional session used long (3 min) rest between all sets and exercises, and the circuit session sequenced the 3 different-muscle-group exercises with 35 seconds between consecutive exercises, preserving approximately 3 minutes of recovery for each muscle group while shortening the session by approximately two thirds. Outcome measures included oxygen consumption, heart rate, respiratory exchange ratio, energy cost, post-exercise blood lactate, and excess post-exercise oxygen consumption over 20 minutes. The findings demonstrated that oxygen consumption (75% higher), heart rate (39% higher), respiratory exchange ratio, and energy cost (66% higher) were significantly higher during the circuit session; blood lactate was significantly higher at 1.5, 5, and 7 minutes post-exercise (9.4, 8.7, and 8.4 compared to 4.4, 3.9, and 3.2 mmol/L) and excess post-exercise oxygen consumption was significantly higher (126%) following the circuit session; and the respiratory exchange ratio was statistically similar between sessions post-exercise. Note that the circuit maintained full local recovery by performing different-muscle-group exercises in place of passive rest, the structural feature underlying the density argument .

• Marín-Pagán, C., Blazevich, A. J., Chung, L. H., Romero-Arenas, S., Freitas, T. T., & Alcaraz, P. E. (2020). Acute physiological responses to high-intensity resistance circuit training vs. traditional strength training in soccer players. Biology, 9(11), 383. https://doi.org/10.3390/biology9110383

Section 3: Comparing Circuit Training and Horizontal Loading

Bench press repetitions, bar velocity, and power output are statistically similar when 6-RM leg extension and ankle extension sets are performed during the 3 minutes between heavy bench press sets or when the rest is passive; average heart rate is significantly higher during the circuit condition

Alcaraz et al. compared 10 healthy men (age: 26 ± 1.6 years; bench press 1-RM: 78 ± 12 kg) with 3.1 ± 2.1 years of resistance training experience and at least 1 year of dynamic free-weight training. All participants performed a traditional session and a heavy resistance circuit session in random, counterbalanced, crossover order, 1 session/week following a familiarization week, at the same time of day. The traditional session included modified Smith machine bench presses for 5 sets, reps-to-volitional-fatigue/set, with 6-RM loads, long (3 min) passive rest between sets, and a 3-second eccentric with a maximal velocity concentric, and lasted approximately 13 minutes; the circuit session included the identical bench press sets, with 1 set each of leg extensions and ankle extensions with 6-RM loads performed between bench press sets and 35 seconds between exercises, maintaining approximately 3 minutes between bench press sets. Outcome measures included repetitions per set, average and peak bar velocity, average and peak power (rotary encoder), and heart rate. The findings demonstrated that total bench press repetitions (20.8 compared to 21.7), average and peak bar velocity, and average and peak power were statistically similar between conditions in every set; average heart rate was significantly higher during the circuit session (129 compared to 113 beats/minute, approximately 71 compared to 62% of age-predicted maximum); and the difference was attributable to significantly higher heart rates during the rest periods, with statistically similar heart rates during the lifting itself. Note that outcomes were assessed for the bench press only, during a single session.

• Alcaraz, P. E., Sánchez-Lorente, J., & Blazevich, A. J. (2008). Physical performance and cardiovascular responses to an acute bout of heavy resistance circuit training versus traditional strength training. Journal of Strength and Conditioning Research, 22(3), 667-671. https://doi.org/10.1519/JSC.0b013e31816a588f

Bench press and half squat 1-RM strength, bench press peak power, lean body mass, and 20-meter shuttle-run performance improve significantly and similarly following 8 weeks of training with 6-RM loads performed as circuits with 35 seconds between different-muscle-group exercises or as a traditional structure with 3 minutes between sets; circuit sessions required approximately half the time

Alcaraz et al. compared 33 resistance-trained men (age: 22.7 ± 3.3 years) with at least 12 months of resistance training experience and the ability to produce isometric squat force equal to twice body mass, randomly assigned to a circuit group (n = 15), a traditional group (n = 11), or a control group (n = 7); withdrawals resulted primarily from injuries sustained outside the study. Following a 1-week familiarization, both training groups performed 6 exercises in 2 blocks of 3 different-muscle-group exercises (leg curl, bench press, and standing calf raise; lat pulldown, half squat, and preacher curl) for 8 weeks, 3 sessions/week, 3 progressing to 6 sets/exercise, 6-RM/set, with 6-RM loads (approximately 85 to 90% of 1-RM), a 3-second eccentric with a maximal velocity concentric, and 5 minutes between blocks; the traditional group used long (3 min) rest between sets, with sessions lasting 105 to 125 minutes, and the circuit group performed each block as a circuit with 35 seconds between exercises, with sessions lasting 55 to 78 minutes. Outcome measures included bench press and half squat 1-RM strength, bench press peak power with 30 to 80% of 1-RM loads, body composition (dual x-ray absorptiometry), Wingate peak cycling power and maximum lactate, and 20-meter shuttle-run time. The findings demonstrated that bench press (19.5 compared to 17.7 kg) and half squat (44.2 compared to 45.0 kg) 1-RM strength increased significantly and similarly in both training groups and significantly more than the control group; bench press peak power increased significantly at most loads in both training groups with no significant between-group differences; shuttle-run performance improved significantly in both training groups when compared to the control group; peak cycling power increased significantly only in the traditional group; body fat percentage decreased significantly only in the circuit group (1.5%); lean mass increased significantly in both training groups (1.5 and 1.2 kg); and bone mineral density and content did not change in any group. Note that power outcomes were measured in the bench press and cycling, and no jump outcomes were included. Note that the circuit sequenced different muscle-group exercises, preserving approximately 3 minutes or more between sets for the same muscle group (??).

• Alcaraz, P. E., Perez-Gomez, J., Chavarrias, M., & Blazevich, A. J. (2011). Similarity in adaptations to high-resistance circuit vs. traditional strength training in resistance-trained men. Journal of Strength and Conditioning Research, 25(9), 2519-2527. https://doi.org/10.1519/JSC.0b013e3182023a51

Hang clean, bench press, and back squat 1-RM strength increase significantly following 6 weeks of a mixed Olympic lift and strength exercise program performed in blocked or circuit order by high school football players; relative strength gains are statistically similar between orders, and sprint and agility times do not improve with either

Johnson et al. compared 39 high school football players from 2 separate teams, with each team assigned to a training order rather than randomly assigning individuals: a circuit-order team (n = 23; age: 16 ± 1 years) and a blocked-order team (n = 16; age: 16 ± 2 years); the circuit-order team was significantly stronger and more agile at pre-training. Both teams performed the same 6-week summer program, 3 sessions/week, of total-body multi-joint exercises mixing Olympic lift variations and strength exercises (e.g., hang cleans, power jerks, push presses, dumbbell snatches, bench and incline presses, front and back squats, lunges, weighted step-ups, inverted rows, and pull-ups) for 3 sets/exercise with undulating volume; hang cleans were performed first in every session by both teams in blocked order for 5 sets/exercise, 1 to 3 reps/set; the blocked-order team completed all sets of each remaining exercise before the next exercise, and the circuit-order team completed 1 set of each exercise before repeating the sequence; rest between sets and activity outside of football practice were not controlled. Outcome measures included hang clean, bench press, and back squat 1-RM strength, a strength index (sum of the 3 lifts divided by body mass), 40-yard sprint time, and pro agility time. The findings demonstrated that all 3 lifts increased significantly for both teams; post-training strength index was statistically similar between orders after controlling for pre-training strength; hang clean gains were significantly greater in the blocked-order team, although both teams performed hang cleans in blocked order, which the authors attributed to that team's weaker pre-training status; and sprint and agility times did not improve significantly in either team. Note that the team-based allocation, unequal baseline strength, and uncontrolled rest intervals limit between-order comparisons, and findings are best interpreted as both orders producing significant strength gains. Note that this study demonstrates that a program including Olympic lift variations can be performed in circuit order without an apparent loss of strength adaptation.

• Johnson, S., Burns, S., & Azevedo, K. (2013). Effects of exercise sequence in resistance-training on strength, speed, and agility in high school football players. International Journal of Exercise Science, 6(2), 126-133.