REF-1298 · Marcora & Bosio, 2007 · Scandinavian Journal of Medicine & Science in Sports · Paper · PubMed

Do drop jumps two days before a hard run slow me down?

In a 30-runner RCT, 100 drop jumps cut a 30-min treadmill time trial by 4% at 48 h; economy did not change significantly, they ran slower at the same effort. A big dose (B+).

Main caveat: Read from the abstract only; group sizes, the control group's change, the between-group test and individual responses could not be checked.
moderate evidenceMeasured on recreational to trained athletes

Marcora and Bosio asked whether exercise-induced muscle damage, known to reduce strength and power, also reduces endurance running performance. Thirty moderately trained runners (24 men, six women) were randomly assigned to a damage group, which jumped 100 times from a 35-cm bench, or to a control group that did no damaging exercise.

Before and 48 h after, they were tested on markers of damage, on steady-state cardiorespiratory, metabolic and perceptual responses during a constant-speed submaximal run, and on the distance run in a 30-min self-paced treadmill time trial. Soreness, creatine kinase and knee-extensor strength changed significantly (P < 0.01).

The time trial fell by 4% (P < 0.01) because the runners chose a slower speed (P = 0.02), with no change in perceived exertion during the trial (P = 0.31).

Running economy and the other responses to the submaximal run did not change significantly; there was a trend (P = 0.08) for higher perceived exertion at the submaximal speed, and that rise correlated with the loss in time-trial performance (P < 0.01).

The authors conclude that muscle damage impairs endurance running and that the effect seems to be mediated by the sense of effort. Our reading: the clearest randomised number in the literature for a jump session two days before a hard run — and it shows that 48 h is not enough after a new, high-dose plyometric session, while a normal economy test would have missed the loss.

Protocol

Participants: n = 30 moderately trained adult runners, 24 men and 6 women; group sizes, age, VO2max and training history are not in the abstract.
Familiarity: not stated in the abstract whether the runners did plyometric or drop-jump training (probably unaccustomed, our inference).
Allocation: random assignment to muscle damage (EIMD) or control.
Damaging session: 100 jumps from a 35-cm bench (drop jumps); controls did no muscle-damaging exercise.
Tests: before and 48 h after the intervention — markers of damage (muscle soreness, creatine kinase, knee-extensor strength); a constant-speed submaximal run (steady-state cardiorespiratory, metabolic and perceptual responses, including running economy); a 30-min self-paced treadmill time trial (distance covered).
Time point: 48 h only; no 24-h or 72-h test.
Funding and interests: the printed statement could not be read; PubMed indexes the paper as "Research Support, Non-U.S. Gov't" and OpenAlex lists Bangor University as funder (the author's institution).

Key Results

48 h after 100 drop jumps vs before (abstract; the abstract reports these as effects of the damage, without group-by-time statistics):

OutcomeResultp
Performance — 30-min self-paced treadmill time trial (distance)−4%< 0.01
Running speed chosen in the time trialreduced0.02
Perceived exertion during the time trialno change0.31
Running economy and other responses at a constant submaximal speedno significant change—
Perceived exertion at the submaximal speedtrend to an increase0.08
Correlation: rise in submaximal RPE vs loss in time-trial performancesignificant< 0.01
Markers — soreness, creatine kinase, knee-extensor strengthsignificant changes< 0.01
The abstract does not give the control group's change, confidence intervals, the size of the strength loss or individual responses.

Mechanisms

Measured: time-trial distance fell while running economy, heart rate and metabolic responses at a fixed submaximal speed did not change significantly; the runners chose a lower speed at an unchanged rating of perceived exertion.
Authors' interpretation: muscle damage reduces endurance performance through an increased sense of effort (a psychobiological account, compare REF-370), not through a worse oxygen cost of running.
Inference: a damaged runner may meet a planned pace at an unchanged heart rate but a higher effort, or hold the usual effort at a slower pace; a submaximal economy check alone can miss the damage.

Practical Application

Do drop jumps two days before a hard run slow me down? In a 30-runner RCT, 100 drop jumps cut a 30-min treadmill time trial by 4% at 48 h; economy did not change significantly, they ran slower at the same effort. A big dose (B+).
After a new or high-dose jump session, 48 h is not clean before a key run (this trial).Not from this paper; the base's own practice is a small progressive jump dose placed not within about 48 h before a key session and, when the jump session is new to the athlete, not within 72 h before maximal running — a race, a time trial or a test (SYNTH-45; plyometrics count as damaging in SYNTH-31); this trial found a high-dose session, probably unaccustomed (our inference), still cost 4% at 48 h, the only time point tested.
Judge the day by pace at effort, not by heart rate (inference): the runners' physiology at a set speed looked normal; the loss showed up as a slower self-chosen pace at the same rating of effort.
Familiar, low-dose jump work is a different case — this trial says nothing about it; REF-1297 and REF-1301 point to faster recovery once the session is repeated or lighter.

Limitations

Read from the abstract only; group sizes, the control group's change, the between-group test and individual responses could not be checked.
A single, large, probably unaccustomed dose (100 drop jumps from 35 cm); no repeated bout, so the result is the upper end of the cost.
One time point (48 h); when performance came back is unknown.
30-min treadmill time trial; outdoor or race transfer and other distances not tested; 6 women only.
Funding: Bangor University per OpenAlex; PubMed lists non-US government support; no conflict-of-interest statement could be read.
How this entry was checked and changed
Why this level

parallel RCT with n = 30 (the rubric's 20-39 band), runners, a performance outcome; read from the abstract (the full text is not open; a University of Bologna repository record listed by OpenAlex is not marked open access and sits behind a bot-check page), so the between-group statistics, group sizes and individual responses could not be checked