REF-935 · Mujika et al., 1995 · Canadian Journal of Applied Physiology, Vol. · Paper · PubMed

Effects of Training on Performance in Competitive Swimming

Blimited evidenceMeasured on elite athletesWritten before October 2026, being re-checked

In 18 elite swimmers the study related season-average training intensity, training volume and training frequency to changes in performance, and compared the swimmers who beat the previous year's personal record during the follow-up season (GIR, n = 8) with those who did not (GNI, n = 10).

Improvement over the season correlated significantly with mean training intensity, not with volume or frequency. It was also related to initial performance level, and the eventual improvers had lost less performance during the previous year's detraining.

The authors suggest intensity is the key factor in improving elite swimmers' performance, and that earlier detraining and initial level can jeopardize success even when adaptation to training is good.

Participants

n=18 elite swimmers; GIR (improved their personal record, n = 8) and GNI (did not, n = 10). MEDLINE indexing lists adult, male and female participants.

Protocol

One follow-up training season; mean training intensity, training volume and training frequency quantified per swimmer.
Performance variations across the season, including the decline during detraining from the previous year.
Correlations between training variables and improvement; comparison of GIR and GNI.

Key Results

Improvement in performance vs mean intensity of the training season: r = 0.69, p < 0.01.
Improvement vs training volume or frequency: not significantly correlated.
Improvement vs initial performance level: reported as negatively related (abstract prints "r = 0.90, p < 0.01").
Decline in performance during detraining from the previous year: GIR 6.21 +/- 2.30% vs GNI 9.79 +/- 2.18%, p < 0.01.

Practical Application

In 18 elite swimmers followed over a season, performance improvement correlated with the mean intensity of training (r = 0.69) and not with training volume or frequency; the swimmers who beat their previous year's best had also lost less in the off-season (6.21% vs 9.79%). To "more metres or harder metres?": among already elite swimmers, the season's intensity, not its volume, went with getting faster — as a correlation.
Protect the off-season: a larger performance loss during detraining went with not improving the following season.
Starting level matters: improvement was negatively related to initial performance level, so expect smaller gains the closer a swimmer already is to their best.
Inference, not tested here: for an age-group triathlete swimming a few times a week, raising the share of quality swimming is the more plausible lever than adding easy volume; this study did not test that population.
Not from this paper; the base's own practice: the swim framework SYNTH-74 builds sessions around a CSS-paced main set.

Limitations

n=18, observational and correlational; no control over what each swimmer did.
"Mean intensity" is the authors' own index; the abstract does not define it or give training volumes.
The abstract prints the negative relation with initial level without a minus sign (r = 0.90).
Elite pool swimmers; not triathletes, masters or novices.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

small observational cohort, n=18, correlational