REF-1212 · Mujika, Padilla & Pyne, 2002 · International Journal of Sports Medicine · Paper · PubMed

How much faster should a taper make me on the day?

At the Sydney 2000 Olympics swimmers were 2.2% faster than 3-4 weeks earlier, 91 of 99 swims faster (range −1.1% to +6.0%). World-class swimmers; taper content not described.

Blimited evidenceMeasured on elite to world-class athletes

Mujika, Padilla and Pyne measured how much Olympic swimmers improved over the final 3 weeks of training before the Sydney 2000 Games. They took swimmers who raced the same event or events at the Telstra 2000 Grand Prix Series in Melbourne (26-27 August 2000) and again 21-28 days later at the Olympic Games (16-23 September 2000): 99 performances, 50 by men and 49 by women.

Times improved by 2.18 ± 1.50% overall (p < 0.0001), ranging from 1.14% slower to 6.02% faster; 91 of the 99 swims were faster. Men improved more than women (2.57 ± 1.45% vs 1.78 ± 1.45%, P < 0.01).

The size of the gain was similar for all events and was achieved by swimmers from different countries and performance levels (no comparison across countries or levels is reported in the abstract). The authors offer the figures as a quantitative framework for setting realistic goals from a swimmer's performance before the final training phase.

Our reading: a clean, large-sample estimate of what the final 3 weeks before a championship yield at the top of the sport, about 2%; it matches the 2-3% range of other taper data (REF-536, REF-821, REF-1196) but cannot separate the taper from the Olympic occasion, and the men-women difference is a single observation in one sport.

Protocol

Who: Olympic swimmers who swam the same event at both meets; n = 99 performances (50 male, 49 female). The number of individual swimmers, nations and events is not given in the abstract.
Comparison: official times at the Telstra 2000 Grand Prix Series (Melbourne, 26-27 August 2000) against the Sydney 2000 Olympic Games (16-23 September 2000), 21-28 days later (the abstract's figure).
Exposure: the final 3 weeks of training (F3T), whose content (volume, intensity, taper shape) is not described in the abstract.
Analysis: percentage change per performance; men vs women compared; consistency across events, countries and performance levels examined.
Not given in the abstract: whether heats, semi-finals or finals were used, how multiple swims per swimmer were handled, and funding.

Key Results

GroupPerformancesImprovement over the final 3 weeks (mean ± SD)P
All992.18 ± 1.50% (range −1.14% to 6.02%)< 0.0001
Men502.57 ± 1.45%men vs women P < 0.01
Women491.78 ± 1.45%
Direction: 91 of 99 performances were faster after the final 3 weeks, 8 slower.
Consistency: the size of the gain was similar for all competition events and was achieved by swimmers from different countries and performance levels (abstract; statistics not given).
Checks: the weighted mean of the men's and women's gains is 2.18% (our arithmetic), consistent with the overall figure. The abstract's 21-28 days does not quite match its own dates: a swim on 27 August and again on 16 September is 20 days apart (our arithmetic; minor).

Mechanisms

Authors' interpretation: the improvement reflects the final phase of training leading to the Games; the abstract does not attribute it to any particular taper variable.
Inference: the Grand Prix was a pre-Olympic meet raced in heavier training and with less at stake; part of the 2.18% is the Olympic occasion (motivation, full rest, competition) rather than the taper alone, and the abstract cannot apportion it.
Inference: why men gained more is not explained in the abstract; differences in events, in how hard each sex trained before the Grand Prix, or in taper practice are all possible.

Practical Application

How much faster should a taper make me on the day? At the Sydney 2000 Olympics swimmers were 2.2% faster than 3-4 weeks earlier, 91 of 99 swims faster (range −1.1% to +6.0%). World-class swimmers; taper content not described.
Set the goal from the last pre-taper race: the authors' own use of the data is to set realistic targets from where the athlete stands before the final phase; about 2% is the central expectation at elite level, 0.5-6% the usual range across studies (REF-1196, B+).
Do not judge the taper by a lab test: the gain here is in competition; pooled taper data show race and time-trial gains with no change shown in VO2max (REF-3492, A-).
Women and men: in these Olympic swimmers women gained less (1.78% vs 2.57%); in recreational marathoners women gained more from strict tapers (REF-821, B+). There is no consistent sex rule; taper women the same way and judge by the race.
A slower swim after the taper happens: 8 of 99 were slower. One flat race is not proof that the taper failed; arriving overreached (REF-1203, REF-1056) or unwell (REF-1209) are the named risks.

Limitations

Abstract only. The full text (Thieme) is not open access; the number of swimmers, the rounds used and the handling of repeated swims could not be read.
No comparison group. Every swimmer was in the final phase before the Olympics; the gain cannot be attributed to the taper as such, nor its size to any taper variable.
Two different meets. A Grand Prix and the Olympic Games differ in motivation, competition and conditions.
Repeated performances. 99 performances, not 99 independent swimmers; swimmers in several events count more than once.
Selection. Only swimmers who raced the same event at both meets were included.
Funding and conflicts: not given in the abstract.
How this entry was checked and changed
Why this level

from the abstract; n = 99 performances (the number of swimmers is not given in the abstract), no comparison group, and the content of the final 3 weeks of training is not described