Post-Activation Performance Enhancement in the Bench Press

PAP, PAPE and the fatigue balance

Classical post-activation potentiation is an increased twitch response for a given electrical stimulus after prior activation. It peaks quickly and has a short half-life, commonly linked to myosin regulatory light-chain phosphorylation. PAPE is a later change in a voluntary task—bar velocity, throw power, jump height or maximal strength—without proof that the twitch mechanism caused it. Muscle temperature, intramuscular water, activation, arousal, skill rehearsal and fatigue can all contribute [1].

The distinction matters because a heavy set may increase twitch force while voluntary performance remains depressed. Conversely, voluntary performance can improve six to ten minutes later, when classical PAP has largely dissipated. Training status, relative strength, conditioning load and rest interval have all been proposed as moderators, but their group-average influence does not identify an individual responder [2]. Calling every improvement “PAP” overstates a mechanism the study did not measure. This article uses PAPE for voluntary bench outcomes.

Coaching practice used “PAP” before the terminology split

Josh Bryant described placing a heavy bench set before speed work using the then-common PAP label [3]. Modern terminology should reserve post-activation potentiation for the short-lived physiological phenomenon measured through twitch properties and use PAPE for a voluntary performance change after a conditioning activity [1]. The practice record establishes the coaching question; controlled studies estimate the effect.

What the upper-body reviews found

The bench-press-throw meta-analysis

Krzysztofik and colleagues searched MEDLINE, PubMed and SPORTDiscus through August 2020. Eleven studies met their criteria, representing 174 resistance-trained men with average relative bench strength of 1.31 ± 0.14 times body mass. All studies tested a Smith-machine bench press throw after a conditioning activity; most were small, and nearly three quarters had moderate or serious risk of bias because of missing control conditions, small samples or a single rest interval [4].

The pooled pre-to-post effect on bench-press-throw performance was small: effect size 0.33 (95% CI 0.12 to 0.54; p < .01), with no detected heterogeneity. Moderator estimates favored one set over multiple sets (0.37 versus 0.29) and bench conditioning at 60–84% 1RM (0.43) over ballistic-plyometric activity (0.29), 85–100% loading (0.23), supramaximal loading (0.22) or concentric-only work (0.11). For conditioning below 85%, five to seven minutes produced the largest estimate (0.48). At 85% or higher, 0.15–4 minutes was negative (−0.13), while five minutes or longer was positive but uncertain.

These subgroup values are not head-to-head randomized comparisons and often pool few participants. They identify plausible starting conditions; they do not prove that 70% for exactly six minutes is an optimum.

The broader upper-body review

Finlay and colleagues screened 127 records and included 31 upper-body studies spanning bench variations, throws, swings, bodyweight activity and combined protocols. Their meta-analysis found that bench press at at least 80% 1RM significantly improved later power in a 30–40% 1RM ballistic bench throw after eight to twelve minutes (effect size 0.31, p = .03). Overweight sport implements also improved competition-weight throwing distance by roughly 1.7–8.5%, and lighter or isometric bat swings improved competition bat velocity by roughly 1.3–4.9% [5].

The useful principle is biomechanical specificity: a heavy press is most defensible before an explosive press. It does not follow that the same conditioning set improves a paused 1RM, repeated hypertrophy work or a sport skill with a different force-time profile.

The modern control-group correction

Xu and colleagues synthesized 62 studies with 1,039 participants and explicitly examined research design. They rated the pooled evidence low certainty with high risk of bias. The pre-to-post effect was trivial (0.12; 95% CI 0.06 to 0.19), while PAPE versus control appeared small (0.30; 95% CI 0.20 to 0.40). Part of that larger contrast came from a small decline in passive controls (−0.08). Prediction intervals crossed zero in both analyses, meaning future protocols can reasonably impair performance [6].

Most revealingly, PAPE after a comprehensive warm-up was essentially trivial (0.01). Larger effects appeared with incomplete warm-ups, more training experience and stronger participants. This reframes the practical question: not “can a conditioning activity beat an early baseline?” but “does it add anything to the best realistic warm-up the athlete would already perform?”

How to read a small effect at the rack

A standardized effect expresses change relative to variability; it does not tell a coach how many kilograms, meters per second or centimeters an athlete gained. The pooled 0.33 in the bench-press-throw review is therefore compatible with a meaningful change in a highly reliable power test and with a change too small to separate from noise in a less reliable field test. The confidence interval describes uncertainty around the average estimate. It does not guarantee that any individual lies inside that range, and it does not turn statistical significance into practical importance.

Prediction intervals answer another question: where effects from a future comparable protocol may fall. Xu and colleagues reported intervals that crossed zero, which fits the observable responder problem—some athlete-protocol pairs improve, others do not, and some get worse. A coach should not label an athlete a responder from the best of several post-conditioning attempts. Multiple testing, ordinary warm-up progression and day-to-day variation can manufacture an apparent win. Predefine the outcome, rest windows and smallest worthwhile change, then repeat the comparison on separate days.

Absolute differences supply the practical scale when authors report them. Wilcox and colleagues observed roughly 2.9–3.1 kg higher mean 1RM after the plyometric conditions than in the first session. That difference could matter in competition, but session order weakens the causal claim because the unconditioned trial always came first. Finlay and colleagues’ 0.31 pooled effect for a light bench throw after heavy bench is stronger evidence for that particular sequence, not for a maximal paused bench. The target task, not the conditioning activity’s intensity, determines whether the result transfers.

This is why a complete warm-up is the appropriate comparator. Passive waiting tests whether the conditioning activity beats doing nothing; athletes rarely choose that alternative. A warm-up-only control tests incremental value. If the conditioning set merely recreates the temperature, rehearsal or arousal already achieved, it adds handling and fatigue without improving readiness. The most useful PAPE protocol is not the most dramatic one—it is the smallest repeatable addition that clears test noise and leaves the rest of the session intact.

Comparator quality changes the answer

A 2025 synthesis of velocity-based conditioning activities included nine studies and 152 participants. Bench-press conditioning activity, loads at or above 80% 1RM, one set and stopping near 10% velocity loss showed favorable trends; the modeled window was roughly 5.8–12.8 minutes with an estimated optimum near 9 minutes. Subgroup differences were not statistically resolved, so these are testable starting points rather than prescriptions [7].

Bench-Specific Performance Examples

Wilcox and colleagues tested 12 male college athletes across three sessions. After a general and bench-specific warm-up, two plyometric push-ups or two 3–5 kg medicine-ball chest passes were performed 30 seconds before each 1RM attempt in counterbalanced second and third sessions. Bench 1RM averaged 120.9 ± 23.2 kg in the first session, 123.8 ± 23.5 after push-ups and 124.0 ± 24.1 after chest passes; both later conditions were significantly higher than the first [8]. The magnitude—about 2.9–3.1 kg—is practical, but the non-conditioning baseline always occurred first, so familiarization or order cannot be fully excluded.

Krzysztofik and colleagues later studied whether conditioning could increase subsequent bench-training volume. A heavy bench protocol changed time under tension without increasing repetition count, underscoring that “more work” depends on the metric selected [9]. PAPE for one explosive repetition is not automatically PAPE across several fatiguing sets.

AEL and Subsequent Concentric Performance

AEL has a superficially appealing PAPE rationale: a heavier descent might increase preactivation or residual force enhancement before the press. Direct bench evidence is inconsistent. In 10 trained men, lowering 110% and pressing 85% produced a slower second repetition and an earlier, lower sticking region than 85/85 despite higher activation in some muscles [10].

Yang and colleagues pooled eight AEL studies. Mean concentric velocity was significantly lower on the repetition exposed to AEL (effect size −0.25), then similar on repetitions two and three [11]. Zhang and colleagues likewise found similar acute submaximal concentric performance overall and higher perceived effort with AEL; chronic strength was not superior [12]. AEL can be a training exposure, but “heavier down” should not be assumed to potentiate “faster up.”

An N-of-1 protocol worth keeping

1. Choose one target. Use the actual competition or training outcome: 30% bench-throw velocity, medicine-ball distance, valid 1RM, or another repeatable measure.
2. Complete the real warm-up. Include the general, specific and high-intensity work the athlete would ordinarily use.
3. Establish reliability. Measure baseline on several days before interpreting a one-day personal best, using the setup controls in Stability, Traction & Setup Repeatability.
4. Use the smallest conditioning dose. Start with one to three heavy reps, two to five ballistic contacts, or a brief isometric—not a fatiguing workout.
5. Sample time. For heavy bench before a light throw, test roughly 4, 8 and 12 minutes across separate exposures. Ballistic activity may use shorter windows.
6. Compare with a warm-up-only control. Randomize order across sessions and keep motivation, feedback and equipment constant.
7. Keep a responder rule. Retain the protocol only if improvement exceeds normal test noise on more than one day without harming the next task.

These windows summarize published groups, not a universal response curve. Strong athletes may tolerate more conditioning, but the 2025 synthesis also found wide prediction intervals. The athlete’s repeated data outrank the group-average clock.

Where the AMM Findings Fit in the PAPE Evidence

The acute Launch Pad crossover compared five bench repetitions at 70% 1RM under two simultaneous equipment conditions in 10 resistance-trained men. Mean and peak concentric velocity, pectoralis surface EMG and vertical displacement were higher in the Launch Pad condition [13]. PAPE requires a conditioning activity followed by a recovery interval and a subsequent standardized performance test. The AMM acute study instead tested performance under two simultaneous equipment conditions.

The four-week study in 42 intermediate-trained men and the eight-week study in 30 collegiate football players reported larger strength and performance gains under the Launch Pad condition [14][15]. Training adaptation and acute PAPE are separate research questions.

Within the PAPE evidence, the AMM studies supply adjacent equipment-condition and training-adaptation findings rather than a conditioning-activity experiment. Joint Ops has no published PAPE outcome study and remains outside that classification. This distinction identifies the next direct design: a conditioning activity, recovery interval, standardized bench-performance test and complete-warm-up control.