REF-074 · Berryman et al., 2018 · International Journal of Sports Physiology and Performance, Vol. · Paper
Strength Training for Middle- and Long-Distance Performance: Meta-Analysis of 28 Studies (running, cycling, cross-country skiing, swimming)
Overview
Berryman, Mujika and colleagues pooled 28 studies (from 554 candidates) in which runners, cyclists, cross-country skiers and swimmers added a block (mesocycle) of strength work on top of their usual sport training, with middle- or long-distance performance as the outcome. Across studies, performance improved moderately (net SMD 0.52 [95% CI 0.33-0.70]).
The gains went together with a better energy cost of locomotion (0.65 [0.32-0.98]), higher maximal force (0.99 [0.80-1.18]) and higher maximal power (0.50 [0.34-0.67]). Maximal-force training outperformed the other strength-training intensities, and in subgroup analyses the benefit held regardless of how good the athletes were.
Key Findings
Performance (middle- and long-distance): SMD 0.52 [0.33-0.70], moderate
Energy cost of locomotion: SMD 0.65 [0.32-0.98]
Maximal force: SMD 0.99 [0.80-1.18], large
Maximal power: SMD 0.50 [0.34-0.67]
Maximal-force (heavy) training > other strength-training intensities
Benefit consistent across athlete levels (subgroup analysis); the subgroups also examined duration, frequency, age, sex and sport, but the abstract reports no thresholds for them
Practical Application
Meta-analysis of 28 controlled studies in running, cycling, cross-country skiing and swimming (Berryman 2018): adding a strength-training mesocycle improved middle- and long-distance performance by a moderate SMD 0.52 [0.33-0.70], via better energy cost of locomotion (0.65), maximal power (0.50) and above all maximal force (0.99); heavy, maximal-force training worked better than lighter strength work, and the benefit held regardless of athlete level.
Plan for the cascade the effect sizes describe: strength (force) changes most, economy next, performance last — do not judge a strength block by an immediate time-trial jump.
Choose heavy loads: the subgroup analysis favoured maximal-force training over other intensities.
Cycling is inside the pooled sample, so the result applies to cyclists directly — this is not a running-only finding transferred by analogy.
Not from this paper; the base's own practice: two to three sessions a week for at least eight weeks, then maintained at lower volume through the aerobic blocks that follow (REF-047, REF-052); the meta-analysis examined frequency and duration as subgroups but its abstract gives no optimal values.
Limitations
Pooled from mostly small controlled trials of varied design; the energy-cost estimate in particular has a wide interval (0.32-0.98).
SMDs describe standardised effects, not seconds or watts; translating to a race result needs the sport-specific context.
Subgroup findings on frequency, duration, age and sex are not reported in the abstract, so no dose can be read off this entry.
Strength "mesocycle" length and content varied across the included studies.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B+: meta-analysis of controlled trials, mostly small studies
Content correction 2026-09-27
"28 trials specifically in middle/long-distance runners (3 km to marathon)", "a similar effect pattern in cycling is assumed, not demonstrated here" -> wrong: the pooled studies were in running, cycling, cross-country skiing and swimming.
"Optimal frequency 2-3×/week, 8-12 weeks minimum — dose from the meta-analysis" -> not in the abstract; moved to a labelled base-practice line. Added maximal power SMD 0.50 [0.34-0.67], the 95% CIs, "maximal-force training > other intensities" and "consistent irrespective of level".
Level B+ -> A- (meta-analysis of controlled trials); Sport corrected to the four pooled sports; title corrected (paper is not running-only).