ArticlePDF Available

Current Controversy: Analysis of the 2013 FINA World Swimming Championships

Authors:

Abstract and Figures

Anecdotal reports from the 2013 FINA World Swimming Championships held in Barcelona suggested that swim performances were biased, presumably due to a current. Purpose: The purpose of this study was to analyze the swimmers' performance data in order to determine the merit of these rumors. Methods: The mean lap difference (odd - even) for each lane in the 1500-meter freestyle was compared to the overall mean difference (-0.081 s) using one-sample t-tests. For each 50-meter event, a percent change in performance with advancing round was calculated and analyzed for all qualifying swimmers. Observations were categorized according to the swimmers' lane assignments for each pair of swims. Results: For the 1500-meter freestyle, the mean difference between odd and even splits was significantly different (P < .05) from the overall mean difference for each lane. This difference became decreasingly negative from lane 1 to 4 and increasingly positive from lane 5 to 8. Regarding 50-meter performances, outcomes were consistent with a significant lane effect (F(3,186) = 31.26, P < 0.001). When swimmers transitioned from lanes 1-4 for their first swim to lanes 5-8 for their second, they improved 1.11%, which was significantly greater than any other scenario. When swimmers were in lanes 5-8 for their first swim and 1-4 for their second, they were slower by 0.59%, which was significantly worse than any other lane change combination. Conclusion: The existence of a current is the only scenario that we can propose to explain these analytic findings. Given that many of the competitors' careers and incomes are dependent on a high-ranking championship performance, new policies are needed to prevent similar biases from happening in the future.
Content may be subject to copyright.
Current Controversy: Analysis of the 2013 FINA
World Swimming Championships
ANDREW CORNETT
1,2
, CHRISTOPHER BRAMMER
1
, and JOEL STAGER
1
1
Counsilman Center for the Science of Swimming, Department of Kinesiology, Indiana University, Bloomington, IN;
and
2
School of Health Promotion and Human Performance, Eastern Michigan University, Ypsilanti, MI
ABSTRACT
CORNETT, A., C. BRAMMER, and J. STAGER. Current Controversy: Analysis of the 2013 FINA World Swimming Championships.
Med. Sci. Sports Exerc., Vol. 47, No. 3, pp. 649–654, 2015. Anecdotal reports regarding the 2013 Fe
´de
´ration Internationale de Natation
World Swimming Championships held in Barcelona suggested that swimming performances were biased, presumably because of a
current. Purpose: The purpose of this study was to analyze the swimmers’ performance data to determine the merit of these rumors.
Methods: The mean time difference between odd and even 50-m lengths for each lane in the 1500-m freestyle was compared. For each
50-m event, a percent change in performance from the preliminaries to semifinals and semifinals to finals was calculated for all qualifying
swimmers. Observations were grouped according to the swimmers’ lane assignments. Results: For the 1500-m freestyle, lane assignment
significantly affected the time difference between odd and even 50-m lengths (PG0.001). The change in performance for the 50-m
events was also affected by lane assignment (PG0.001). When swimmers transitioned from lanes 1–4 for their first swim (preliminaries
or semifinals) to lanes 5–8 for their second (semifinals or finals), their performance time improved by 1.11% (95% confidence interval,
0.82%–1.41%), which was significantly greater than any other lane change scenario. When swimmers were in lanes 5–8 for their first
swim and lanes 1–4 for their second, their performance time was slower by 0.59% (95% confidence interval, 0.39%–0.80%), which was
significantly worse than any other lane change combination. Conclusions: Swimmers were advantaged or disadvantaged depending on
the direction and lane in which they swam. The existence of a current is the only cause that we can propose to explain these findings.
Because one of Fe
´de
´ration Internationale de Natation’s primary stated objectives is to provide fairness in competition, new policies are
needed to prevent similar biases from occurring in the future. Key Words: ATHLETIC COMPETITION, SWIMMING, ANALYTICS,
PERFORMANCE, SPORT
Popular author Michael Lewis, in his book entitled
Moneyball, focused national attention on the potential
of sophisticated analyses to identify athletic promise
and predict the future competitive success of baseball
players. The concept did not originate with Lewis; rather, he
suggests that it originates from the Society of American
Baseball Research from whose acronym is derived the term
‘sabermetrics.’’ The basis of sabermetrics is to essentially
forgo traditional observation-based physical evaluations and
use statistical measures to analyze an athlete’s past perfor-
mance data as a means to predict future success. The modern
modus operandi, then, is to make important decisions (e.g.,
regarding recruiting, hiring, firing, etc.) in professional sport
on the basis of analytical outcomes rather than simple intui-
tion, opinion, or experience.
These same analytic techniques may also be useful as a
means of identifying biases in sport. For example, on the
basis of analysis of performance data, the expensive ‘‘high-
tech’’ suits introduced into swimming in late 2007 were
concluded to bias the competition outcome in favor of those
with access to them (2). The disproportionate rate of World
Record performances provided additional convincing evi-
dence of the magnitude of this bias only after the important
competitions were concluded. Nevertheless, suit design limi-
tations were subsequently imposed by the sport’s international
governing body, the Fe
´de
´ration Internationale de Natation
(FINA), to constrain competitive suit manufacturers primarily
as a means to reestablish competitive fairness (5). The point to
be made is that competition biases in sport were, and can be,
identified through similar analyses of readily available per-
formance data.
A recent scenario involving competition at the 2013 FINA
World Swimming Championships in Barcelona, Spain, pres-
ents another opportunity to use existent performance data to
determine whether the competitive outcomes were similarly
biased. Anecdotal reports regarding the competition indi-
cate that a current was present in the pool such that when
swimming toward the finishing end, swimmers were at a
Address for correspondence: Joel Stager, Ph.D., Counsilman Center for the
Science of Swimming, Indiana University, 1025 E 7th St., Bloomington, IN
47405; E-mail: stagerj@indiana.edu.
Submitted for publication February 2014.
Accepted for publication July 2014.
0195-9131/15/4703-0649/0
MEDICINE & SCIENCE IN SPORTS & EXERCISE
Ò
Copyright Ó2014 by the American College of Sports Medicine
DOI: 10.1249/MSS.0000000000000442
649
APPLIED SCIENCES
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
competitive disadvantage on one side of the pool and at an
advantage on the opposite side. If there was, in fact, a current in
the pool during competition, it would have violated FINA
regulations, which state that water flowing into and out of
the pool is ‘permissible as long as no appreciable current or
turbulence is created’’ ((4), p. 392). Flow characteristics of
the competition pool cannot be measured, though, because
the competitive venue was temporary, built specifically for the
World Championships and deconstructed immediately there-
after. However, quantitative analysis of the existing perfor-
mance data might allow appropriate conclusions to be drawn.
Thus, the purpose of this study was to analyze performance
data from the 2013 FINA World Swimming Championships
to determine whether lane assignment affected swim times.
METHODS
The performance data for this study came from swim
competition results from the 2005, 2007, 2009, 2011, and
2013 FINA World Swimming Championships and were
obtained from the Omega Timing Web site (7). Furthermore,
measurements of speed, stroke rate, and stroke length for
50-m events were obtained from the 2013 FINA World
Swimming Championships Web site (1). Values were re-
portedly captured for each 10-m segment between 15 and
45 m on each length via ‘‘the automated system installed and
programmed’’ by Centro de Alto Rendimiento de Sant Cugat
and STT Systems. Although the hypotheses only concerned
the 2013 competition, results from the 2005, 2007, 2009, and
2011 competitions were also included in an effort to determine
whether the 2013 results were atypical of FINA World
Swimming Championships. Thus, identical analyses were con-
ducted on performance data from the previous four competi-
tions. The competitions during all five of these international
championships were held in ‘‘long-course’’ pools that were
50 m long.
In an effort to address the research questions, three sepa-
rate analyses were conducted. For all statistical analyses,
significance was set at an alpha level of 0.05. First, 50-m
splits were analyzed from the 1500-m freestyle, an event in
which the athletes typically attempt to maintain a fairly even
pace throughout most of the race. Consistent with anecdotal
reports, it was hypothesized that the time difference between
the odd 50-m lengths (recorded when swimming away from
finishing end) and the even 50-m lengths (recorded when
swimming toward the finishing end) from the 1500-m free-
style would vary by lane. To eliminate the effect of the start
and the tactical aspect of the finish, the 50-m splits from the
first or last 100 m of the races were not included. Thus, for
each preliminary and final 1500-m performance, an aver-
age of 13 odd-length 50-m splits and 13 even-length 50-m
splits was recorded. The difference between the average odd
and average even 50-m splits was then calculated for each
1500-m freestyle performance. A one-way ANOVA was used
to determine whether lane assignment affected the difference
between the odd and even 50-m splits. It is important to note
that in the 1500-m freestyle at the FINA World Swimming
Championships, all competitors swim in the preliminary round
and then the eight fastest competitors swim again in a final
heat. This means that the top eight swimmers have two per-
formances in the data set. Because the focus of this study was
to assess the effect of lane assignment on performance mea-
sures and our sample size was already limited we chose to
include all performances in the data set, even though this
meant that some swimmers would appear twice in the data
set. This is a delimitation of the study.
Logically, it would seem that for most races, any advan-
tage a swimmer received from swimming in one direction
would be counteracted by a disadvantage when swimming
in the opposite direction. However, for the eight events at
the FINA World Swimming Championship competitions that
are swum in only one direction (men’s and women’s 50-m
butterfly, backstroke, breaststroke, and freestyle), the pres-
ence of a lane bias in the pool could have a major effect on
the event results. We hypothesized that the swimmers in
lanes 5–8 were at an advantage in the 50-m events and the
swimmers in lanes 1–4 were at a disadvantage.
To test whether this apparent lane bias affected the per-
formances in the 50-m events, the percent change in per-
formance time, stroke rate, and stroke length was calculated
for all semifinalists (percent change from preliminary heats
to semifinals) and finalists (percent change from semifinal to
final). For this analysis, as was the case for the 1500-m
freestyle analysis, the swimmers that qualified for the finals
appear in the data set twice. We chose to allow this for the
same reasons noted earlier for the 1500-m freestyle analysis.
Regarding stroke rate and stroke length, three values, rep-
resentative of 15- to 25-m, 25- to 35-m, and 35- to 45-m
distance segments, were averaged and used for analysis. The
total sample was (eight 50-m events 16 semifinalists) +
(eight 50-m events eight finalists) = 192 observations. On
the basis of the swimmers’ lane assignments, each obser-
vation was categorized into one of four lane change sce-
narios (Table 1). We hypothesized that swimmers with their
first swim in lanes 1–4 and second swim in lanes 5–8 (LH)
would show the greatest improvement in performance time
and swimmers with their first swim in lanes 5–8 and second
swim in lanes 1–4 (HL) would show the least improve-
ment in performance time. For each performance variable, a
one-way ANOVA was used to determine whether the four
scenarios were different with respect to the percent change
in performance. In the event of a significant F-ratio, all
TABLE 1. Lane change scenarios for swimmers that qualified for semifinals and finals in
the men’s and women’s 50-m events at the 2013 FINA world swimming championships.
Lane Change Scenario Lane in First Swim Lane in Second Swim
LL 1–4 1–4
LH 1–4 5–8
HL 5–8 1–4
HH 5–8 5–8
For each swimmer, their first swim was either a preliminary swim or a semifinal swim and
their second swim was either th e semifinal or final swim, respectively. H, high-num bered
lanes; L, low-numbered lanes.
http://www.acsm-msse.org650 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
pairwise comparisons were conducted using the Tukey
honestly significant difference (HSD) post hoc test.
Finally, we wanted to test whether evidence of a lane bias
that may have affected the medal winners in the 50-m events
exists. Before describing the analyses used to do so, it is
important to note that, in the finals of each event, the fastest
swimmer from the semifinal heats was seeded in lane 4 for
the finals, the next fastest swimmer was seeded to the right
in lane 5, and the remaining swimmers were seeded to the
left and right in an alternating fashion on the basis of rank.
The number of medals won by swimmers competing in lanes
1–3 and in lanes 6–8 at the 2013 World Championships was
compared with that in the previous four World Champion-
ships. We grouped the lanes in this manner because the
swimmers competing in lanes 4 and 5 are the top two seeds in
the race and, typically, both win medals. As such, the swim-
mers in lanes 4 and 5 serve to cancel each other out, which
could potentially mask any effect. If the rumored current
did exist, the swimmers in lanes 1–3 would have been at a
competitive disadvantage in the 50-m events whereas the
swimmers in lanes 6–8 would have been at a competi-
tive advantage. Thus, we hypothesized that more swimmers
competing in lanes 6–8 and fewer swimmers competing in
lanes 1–3 won medals in 2013 as compared with those in
previous FINA World Swimming Championship competi-
tions. A 2 5 (lane group by competition year) contingency
table was constructed and analyzed using a Pearson chi-square
test of independence.
RESULTS
The mean difference between the odd and even 50-m
splits for the 1500-m freestyle for each lane is displayed in
Figure 1. Lane assignment had a significant effect on the
difference between the odd and even 50-m splits (PG0.001).
Tukey HSD was used to identify which lanes were signifi-
cantly different with respect to the difference between odd
and even 50-m splits (Fig. 1).
One-way ANOVA indicated that lane change scenario
had a significant effect on the percent change in performance
with advancing round (PG0.001). Tukey HSD showed that
the swimmers in LH improved significantly more (1.11%;
95% confidence interval (CI), 0.82%–1.41%) than when
swimmers were in HL (j0.59%; 95% CI, j0.80% to
j0.39%), LL (0.19%; 95% CI, j0.20% to 0.58%), and HH
(0.43%; 95% CI, 0.20%–0.67%). In addition, HL showed a
decline in performance, and this was a significantly a worse
result than the three other lane change scenarios (PG0.05).
HL from the 2013 competition was the only scenario from
the past five FINA World Swimming Championships that
showed a significant decline in 50-m performance (PG0.05)
with advancing round (Fig. 2).
The percent change values from semifinals to finals for
stroke length, stroke frequency, and swim speed were com-
pared across the lane change scenarios. Results showed that
stroke length (P= 0.002) and speed (PG0.001) changed as
a function of lane change scenario but stroke frequency
did not (P= 0.20). Tukey HSD showed that the swimmers
in LH increased stroke length significantly more (0.43%;
FIGURE 1—The difference between the odd and even 50-m splits from
the 1500-m freestyle at the 2013 FINA World Swimming Champion-
ships as a function of lane. Error bars represent the 95% CI around the
mean. Significant differences between pairs of lanes are indicated by
the numbers above each lane (PG0.05).
FIGURE 2—The percent change in performance from preliminaries to
semifinals or from semifinals to finals for the four lane change sce-
narios, which represent the lane assignments for the pair of swims, at
the 2005, 2007, 2009, 2011, and 2013 FINA World Swimming Cham-
pionships. Error bars represent the 95% CI around the mean. Swim-
mers that competed in lanes 1–4 or 5–8 for both swims were assigned to
LL and HH, respectively; swimmers that competed in lanes 1–4 for
their first swim and lanes 5–8 for their second swim were assigned to
LH; and swimmers that competed in lanes 5–8 for their first swim and
lanes 1–4 for their second swim were assigned to HL. There was one
occasion at the past five World Championships where a scenario
showed significant decline in performance (PG0.05), and that was HL
in 2013. All other groups showed significant improvement or no change
in performance.
LANE BIAS IN SWIMMING Medicine & Science in Sports & Exercise
d
651
APPLIED SCIENCES
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
95% CI, j1.25% to 2.10%) than when swimmers were in HL
(j2.97%; 95% CI, j3.85% to j2.09%), LL (j2.23%;
95% CI, j3.80% to j0.66%), and HH (j2.06%; 95% CI,
j3.40% to j0.73%). Furthermore, Tukey HSD showed that
the swimmers decreased swim speed significantly more in HL
(j1.42%; 95% CI, j2.01% to j0.83%) than when swim-
mers were in LH (1.08%; 95% CI, 0.61% to 1.56%),
LL (0.13%; 95% CI, j0.28% to 0.53%), and HH (0.46%;
95% CI, j0.08% to 1.01%).
The number of medal winners from the 50-m events for
lanes 1–3 and 6–8 from the past five FINA World Swim-
ming Championships is shown in Table 2. The Pearson chi-
square test of independence was statistically significant
(P= 0.039), which indicates that the variables were not
independent.
DISCUSSION
Competitive sport operates under the assumption that each
competitor’s success is determined solely by hard work, com-
mitment, discipline, and talent and is not influenced by veiled
external variables or artificially contrived biases. However,
there are documented circumstances whereby external biases
have been shown to unfairly influence sporting outcomes and
other cases in which only rumors of a bias persist. Obviously,
a valid mechanism for identifying performance biases is par-
amount to the maintenance of fairness in competition. The
current investigation illustrates that quantitative analysis of
existing performance data can be used as a tool to identify the
circumstances when and where performances have been sub-
ject to some bias. Through careful analysis of the meet results
derived from the 2013 FINA World Swimming Champion-
ships, our intent was to determine whether credible evidence
exists in support of the anecdotal reports that a current in the
competition venue affected the race outcomes. Our analysis of
the race outcomes included multiple inferential procedures
performed on the results of the 1500-m freestyle and the 50-m
events for all four competitive strokes, and all of our findings
regarding these events were consistent with the existence of
the rumored current.
The analyses regarding the 1500-m freestyle allowed de-
scription of the ‘‘direction’’ of the external bias. As Figure 1
demonstrates, the external bias was not limited to the outer
lanes as the original rumor suggested. Rather, all lanes
seemed to be affected to some extent and in a very particular
manner moving across the pool. The difference between odd
and even 50-m splits was negative for lanes 1–4 and positive
for lanes 5–8. The magnitude of this difference increased
when moving from the center lanes, lanes 4 and 5, to the
outermostlanes, lanes 1 and 8 (Fig. 1). This provides the initial
documentation that swimmers in the 1500-m freestyle were
faster when swimming away from the finishing end than
when swimming back toward it on one side of the pool (i.e.,
lanes 1–4) and faster when swimming toward the finishing
end than when swimming away from it on the other side of
the pool (i.e., lanes 5–8).
In addition, our finding concerning the 1500-m freestyle
illustrated the strength of the external bias. The relation be-
tween lane assignment and the difference between odd and
even splits from the 1500-m freestyle is statistically signifi-
cant (r= 0.87, PG0.001). Thus, the lane assignment in the
1500-m freestyle accounted for 75% of the variance in the
difference between the odd and even 50-m splits. We can
put the strength of this relation into context by calculating
the same correlation coefficient for the four FINA World
Swimming Championships before 2013. In doing so, we find
that the championship held in 2007 was the only other com-
petition in which there was significant relation between lane
assignment and difference between odd and even 50-m splits
for the 1500-m freestyle (r=j0.25, P=0.021).Therelation
was such that the difference between odd and even splits was
negative for all lanes and became increasingly negative mov-
ing across the pool from lane 1 to 8. At the 2007 FINA World
Swimming Championships, then, the lane in which the swim-
mer competed accounted for about 6% of the variance in the
difference between odd and even splits. There seems to be
relatively little commonality between the two events other
that the fact that both the 2007 and 2013 FINA World
Swimming Championships were held in ‘‘temporary’’ pools.
Whether this common feature played a role in the lane biases
is unclear. We were unable to find any research comparing
performances in temporary swimming pools with those in
permanent ones.
Analysis of the outcomes of the 50-m events provided an
alternative approach for the estimation of the magnitude of
the lane bias. Comparisons were possible between the results
of the preliminary heats, semifinal heats, and final heats for
the four competitive strokes and for both sexes. And like the
1500-m results, the findings regarding the 50-m events are
pronounced, especially when compared with those of previous
competitions (Fig. 2). However, to interpret these findings, it
was necessary to first calculate the typical change in perfor-
mance with an advance in competitive round. To do so, we
computed the change in performance from preliminaries to
semifinals and from semifinals to finals for each swimmer
competing in the 50-m events at the past five FINA World
Swimming Championships. We determined that, on average,
swimmers improve by 0.26% (95% CI, 0.20%–0.32%;
N= 955). Because of this, we expected the swimmers to have
improved at the 2013 competition when both swims were on
the same side of the pool (the LL and HH lane change sce-
narios). And that is what happened. When swimmers were
in LL, they improved by 0.19%, and when swimmers were
in HH, they improved by 0.43%. But importantly, although
TABLE 2. The number of medals won by swimmers in lanes 1–3 and 6–8 at the past five
FINA world swimming championships.
Lanes 2005 2007 2009 2011 2013
1–3 9596 1
6–8 357511
Pearson chi-square test of independence was significant, W
2
(4) = 10.11, P= 0.039.
http://www.acsm-msse.org652 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
these improvements were significantly greater than zero,
they were not statistically different from the mean improve-
ment observed over the course of the five FINA World
Swimming Championships from 2005 to 2013 (i.e., 0.26%
improvement).
On the other hand, when a swimmer’s lane assignment
switched from one side of the pool to the other with an ad-
vance in round (the LH and HL lane change scenarios), the
results differed. When swimmers were in LH, their perfor-
mances improved significantly more (1.11% improvement)
than what we determined as being typical (0.26% improve-
ment; PG0.05). This improvement was more than double
that of any lane change scenario from any FINA World
Swimming Championship competition held from 2005 to
2013 (Fig. 2). One explanation for this finding is simply that
swimmers improved to such an extent because they swam
against a current during their first swim and swam with the
current during their second swim. We also confirmed the
other extreme, that being when swimmers were in HL. Our
comparison showed that these swimmers were, on average,
0.59% slower than they were during the previous competi-
tive round. This was the only lane change scenario docu-
mented to swim slower with advancing rounds at the FINA
World Swimming Championships between 2005 and 2013
(Fig. 2). All other groups analyzed showed a significant im-
provement or no change in performance. Once again, this
result is consistent with the explanation that a current existed
in the 2013 competition pool such that swimmers’ perfor-
mances were worse because they swam with a current dur-
ing their first swim and swam against a current during their
second swim.
If the presumed pool current caused the performance bias,
then logically, one or more of the determinants of swim
speed must be affected as well. Although the mechanics of
human propulsion in the water are only partially understood,
it is accepted that swim speed can be viewed as the product
of stroke rate and distance per stroke (3). It stands to reason,
then, that differences in stroke length and/or stroke rate would
be apparent. Consistent with the analyses of performance
times, swimmers in LH had between 2.5% and 3.4% greater
stroke length than the three other lane change scenarios.
Furthermore, swimmers in HL were between 1.5% and 2.5%
slower than the three other lane change scenarios. Although
parallel with our previous analyses, we readily recognize that
this analysis of stroke length, stroke rate, and swim speed
can only be considered circumstantial evidence because little
is known about the data collection procedures. We cannot,
therefore, defend the accuracy, precision, and validity of
these data. Nevertheless, because the comparisons were
statistically significant and theoretically consistent with our
other analyses, they lend further confirmation of the rumored
pool current.
The 2013 FINA World Swimming Championships is not
the first instance in which an external bias was shown to
affect the results of a major international competition (2,6).
The most pertinent example comes from the sport of track
and field and is especially relevant because it involved the
recognition of a lane bias that was subsequently corrected
through an analysis of available performance data (6). Briefly,
Julin and Dapena (6) explained that at the 1996 Olympic
Games start times differed by lane assignment despite a
loudspeaker that transmitted the sound of the starter’s gun
being placed next to each starting block. Further, the differ-
ence in start time corresponded with the distance each lane
was from the starting gun. Because sound travels through the
air at a speed of approximately 350 mIs
j1
, it takes an addi-
tional 0.03 s for the sound from the starter’s gun to reach an
athlete for every 10 m of distance that an athlete is from the
starter’s gun. It seemed that the athletes were still responding
to the sound of the gun transmitted through the air and not the
sound transmitted through the loudspeaker system. Thus, the
athletes who were positioned farther from the starting gun
were at a competitive disadvantage.
This example from the sport of track and field demonstrates
an important and relevant point. Analysis of performance data
can and should be used to identify biases influencing the
outcomes of athletic competitions. In response to the unfair
competitive venue, the International Association of Athletics
Federations, the world governing body for track and field,
mandated new regulations that required different start tech-
nology to eliminate the bias. The analysis of performance data
from the 2013 FINA World Swimming Championships re-
vealed a similar lane bias, but it remains to be seen whether
changes will be mandated to eliminate this bias at future
swimming competitions.
CONCLUSIONS AND RECOMMENDATIONS
The 2013 FINA World Swimming Championships was
held in a temporary swim pool, built specifically for the
event and taken apart immediately afterwards. As such, it
is impossible to conduct any direct testing of the physical
properties of the competition facility. Instead, analyses of
the performance data allowed confirmation of the rumors
that a current existed in the pool during competition. And
although it cannot be definitively concluded that there was a
current in the pool, the results of all of the analyses suggest
that this was, indeed, the case. Swimmers were at a com-
petitive advantage when swimming toward the finishing end
on one side of the pool (lanes 5–8) and at a competitive
disadvantage when swimming in the same direction on the
other side of the pool (lanes 1–4). This advantage was evi-
dent in the final meet results of the 50-m events where a
disproportionate number of swimmers won medals when
swimming in lanes 6–8.
However, the purpose of this paper was not necessarily to
dwell on medals that ‘‘might have been.’’ Rather, the focus
is centered on future competitions. It is hoped that the evi-
dence provided herein will force the swimming community
to consider how similar lane biases can be prevented at sub-
sequent competitions. It is recommended, as a result of these
findings, that the FINA recruit the assistance of engineers and
LANE BIAS IN SWIMMING Medicine & Science in Sports & Exercise
d
653
APPLIED SCIENCES
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
other pool design experts to assess, advise, and limit the
potential for currents to influence race outcomes. From con-
tinued informed dialog among these experts, new regulations
can be put forth that fulfill one of FINA’s primary stated
objectivesto provide fairness in competition.
The authors would like to thank Magdalena Kuras and Jason
Marsteller for making us aware of the rumored bias and Neal Cassady
for driving the bus.
No external funding was received for this project.
The authors declare no conflict of interest.
The results of the present study do not constitute endorsement by
the American College of Sports Medicine.
REFERENCES
1. 2013 FINA World Championships Web site [Internet]. Barcelona:
FINA; [cited 2013 Sept 20]. Available from: http://www.
bcn2013.com.
2. Brammer CL, Stager JM, Tanner DA. Beyond the ‘‘high-tech’’ suits:
predicting 2012 Olympic swim performances. Meas Phys Edu
Exerc Sci. 2012;16:183–93.
3. Craig AB, Pendergast DR. Relationships of stroke rate, distance per
stroke, and velocity in competitive swimming. Med Sci Sports.
1979;11(3):278–83.
4. Fe
´de
´ration Internationale de Natation. FINA Handbook 2013-17.
Lausanne (Switzerland): FINA; 2013. p. 392.
5. Fe
´de
´ration Internationale de Natation Web site [Internet]. Lausanne:
FINA. [cited 2013 Dec 2]. Available from: http://www.fina.org/H2O.
6. Julin L, Dapena J. Sprinters at the 1996 Atlanta Olympic Games did
not hear the starter’s gun through the loudspeakers in the starting
blocks. New Stud Athl. 2003;18(1):23–7.
7. Omega Timing Web site [Internet]. Bienne: Omega SA; [cited 2013
Aug 5]. Available from: http://www.omegatiming.com.
http://www.acsm-msse.org654 Official Journal of the American College of Sports Medicine
APPLIED SCIENCES
Copyright © 2015 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
... When a competitor achieves a time better than the world record, the score exceeds 1000 points, and when the time is slower, which happens much more often, the result is placed in the table between the nearest times in the FINA table and the number of points he has achieved is determined. The FINA table containing the times includes a summary of the times and a corresponding number of points ( [49]). For example, at the same distance mentioned above, time 21.60 gives 825 points, while time 22.70 gives 711 points. ...
Article
Full-text available
The potential of a swimmer can be exploited to the full if an individual approach is applied to training early enough. In the initial stage of a competitor’s career, it’s hard to say whether the person will reach the world level. It is influenced by factors related to physical and mental characteristics, swimming technique or the current skills level of the athlete. Some of them cannot be improved by swimmers; however, a few of these factors can be affected and upgraded by working on them. The data from 30 swimmers were collected in the research. Then, adopting the COMET method, the values of attributes characteristic for the athletes were entered into the proposed model to determine the predispositions of each of them. The applied method proved to be effective in assessing the parameters of competitors, and at the same time, it is a modern approach free of the rank reversal phenomenon. In addition, the model obtained allows for manipulation of input data to check how the change of a given criterion will affect the overall rating. The model developed is forward-looking. The results are satisfactory but can be improved to analyse somatic traits better and achieve more accurate predictions.
... They used descriptive statistics, analysis of variance (ANOVA), and correlation analysis to find out that the performance enhancement was approximately 3-4% for the time-frame. Cornett, Brammer, and Stager (2015) investigated a controversy of the 2013 FINA World Swimming Championships. Through statistical analysis, they identified that the swimming performances were biased depending on the direction and lane in which the swimmers swam. ...
Article
Full-text available
The recently concluded 2019 World Swimming Championships was another major swimming competition that witnessed some great progresses achieved by human athletes in many events. However, some world records created 10 years ago back in the era of high-tech swimsuits remained untouched. With the advancements in technical skills and training methods in the past decade, the inability to break those world records is a strong indication that records with the swimsuit bonus cannot reflect the real progressions achieved by human athletes in history. Many swimming professionals and enthusiasts are eager to know a measure of the real world records had the high-tech swimsuits never been allowed. This paper attempts to restore the real world records in Men’s swimming without high-tech swimsuits by integrating various advanced methods in probabilistic modeling and optimization. Through the modeling and separation of swimsuit bias, natural improvement, and athletes’ intrinsic performance, the result of this paper provides the optimal estimates and the 95% confidence intervals for the real world records. The proposed methodology can also be applied to a variety of similar studies with multi-factor considerations.
... A plethora of factors including illness, injury, the ability to psychologically cope with the big occasion, and even the pool itself through lane bias (3,5) may sabotage a potential PB. The fact that the Olympic games occurs on a quadrennial cycle adds to the pressure of getting the peak right. ...
Article
Full-text available
Olympic games are the pinnacle of elite swimmers' careers where they strive to peak and produce personal best (PB) times. The Olympics provides a unique benchmark to assess successfulness of preparation regimes. The aim of this research was to quantify the likelihood of producing a PB and the expected magnitude of improvement. We analysed relative swim performance gain (RSPG%) or decrement from PB time using data from 793 swimmers (1075 event entries) competing in the 13 male and female individual events of the 2016 Rio Olympics. Analysis of variance (ANOVA) and hierarchical linear regression were performed to estimate the effect of gender, stroke-­‐ type, swimming distance, age, and days since PB, on PB times. We found 339 (31.5%) swimmers produced a PB time during the games. Mean PB improvement was 1.6 ± 3.2% faster, with PB times coming 437 ± 473 days following their previous PB. Gold medallists had ~81% PB success rate. Those failing to set a PB swam 1.5 ± 1.4% slower than their actual PB. The 50m Freestyle was associated with the greatest RSPG% improvement (Males 0.3%, females 1.5% faster). Combining our data with knowledge of actual preparation strategies employed by specific swimmers or teams will allow assessment of the relative success of an approach.
Article
Full-text available
Araştırmanın amacı, günümüze kadar yapılmış Yaz Olimpiyat Oyunlarında Türkiye yüzme milli takımının performanslarının analiz edilmesidir. Türkiye’yi temsil eden milli sporcuların yarışma kategorileri, yarışma sıralamaları, yaş ve cinsiyet değişkenlerinin yıllara göre oluşan dağılımın karşılaştırılması hedeflenmiştir. Günümüze kadar yapılmış olimpiyatlarda yüzmede yer almış Türkiye milli takımı sporcularının yarışma sonuçları incelenmiştir. Veri toplama aracı Uluslararası Olimpiyat Komitesi (IOC) ve FINA (Fédération Internationale de Natation)’nın resmi yarışma sonuçları kullanılmıştır. Sporcuların yarışma sonuçlarındaki sıralamalar yıl, yaş, cinsiyet ve yarışma kategorisi değişkenlerine göre istatistiksel analiz edilmiştir. Yarışma sonuçlarına göre Türkiye’nin günümüze kadar 12 olimpiyatta yüzmede sporcusunun olduğu anlaşılmıştır. Erkek yüzücüler Roma 1960’da, kadın yüzücüler Atlanta 1996’da ilk defa yarışmıştır. Toplam 17 farklı kadın, 32 farklı erkek yüzücü Türkiye’yi temsil etmiştir. Türkiye’nin olimpiyatlarda yüzme yarışmalarında elde ettiği en iyi sonuç “Kadınlar 200 m kurbağalama olimpiyat 9.’luğu”dur. Kadın yüzücülerin erkek yüzücülerden istatistiksel açıdan daha iyi yarışma sonuçlarına sahip olduğu anlaşılmıştır (p0,05). Dört yüzme stilinin 100 m ve 200 m yarışma sonuçları karşılaştırıldığında sırtüstü ile serbest ve kurbağalama ile serbest arasında anlamlı farklılıklar bulunmuştur (p
Article
Performance outcomes at the 2013 World Swimming Championship were previously shown to be biased depending on the swimmer's lane assignment. The purpose of this study was to determine if this kind of bias was unique, and if not, if the bias was related to the temporary or permanent nature of the pool. The effect of lane on the average odd-length split minus the preceding even-length split in the 800- and 1500-m freestyle events, and on the relative change from qualifying to preliminary performance in the 50-m events, was determined for 16 other elite-level competitions. Depending on the swimmers' direction, split times were on average 0.16 s slower or faster in at least one lane at each of the 16 competitions, and in 49% of all lanes analysed. In 5 competitions, swimmers were shown to be faster in a majority of lanes in one direction as compared to the other. Analysis of the 50-m events at these 5 competitions indicate that preliminary performances were between 0.5 and 0.9% slower or faster than qualifying times, which is consistent with the direction effect observed in the distance freestyle events. Further, lane biases occur more often in temporary pools (70% of lanes) than in permanent pools (35% of lanes), with water currents as the most plausible cause. The prevalence of lane bias at elite-level swimming competition highlights the need for the implementation of policies and procedures to prevent such bias from occurring again in the future.
Article
Competitive swimmers were asked to swim at a constant velocity (V) for short distances. They wore a collar to which was attached a fine non-elastic steel wire. The wire passed over two wheels of a device attached to one end of the pool. One wheel generated an impulse for every cm of forward movement and another wheel produced an electrical signal which was directly proportional to V. Measurements of distance and time were begun at definable points in the stroke cycle and were discontinued at the end of a predetermined number of strokes. In all of the four competitive strokes, front and back crawl, butterfly, and breaststroke, the V increased as a result of increasing the stroke rate (Ṡ) and decreasing the distance per stroke (d/S). In the front crawl, the male and female swimmers who achieved the fastest V had the longest d/S at slow Ṡ. The faster male swimmers also had greater percent decrease of the d/S at their maximal V than did the less skilled persons. The back crawl was similar to the front crawl except that maximal Ṡ and V were less. Increases of V of the butterfly were related almost entirely to increases in Ṡ. Except at the higest V, d/S was decreased somewhat. In the breaststroke increased V was also associated with increasing Ṡ, but the d/S decreased much more than in the other stroke styles. Fluctuations of velocity during the stroke cycle were least in the front and back crawl (± 15-20%) and greatest in the butterfly and breaststroke (+45-50%). The results were compared to the Ṡ observed and the values for V and d/S calculated for a large group of swimmers competing in the 1976 U.S. Olympic Trials. The implications of the findings for coaching swimmers are discussed.
Article
The purpose of the authors in this study was to predict the mean swim time of the top eight swimmers in swim events at the 2012 Olympic Games based upon prior Olympic performances from 1972 through 2008. Using the mean top eight time across all years, a best fit power curve [time = a × year] was calculated and used to predict the finish time of the finalists for each event. Every event except the women's 100 breaststroke is predicted to be slower in 2012 than it was in 2008. However, the results of the 2008 swimming competition were shown to be biased, likely caused by the now banned “tech suits.” The authors hypothesize that the 2012 Olympic performances will realign with the prediction curves, thus demonstrating the reliability and sensitivity of the model and further confirming the “suit bias” of 2008. Final conclusions will be dependent upon the athletes' performances at the 2012 Games.
Article
Competitive swimmers were asked to swim at a constant velocity (v) for short distances. They wore a collar to which was attached a fine non-elastic steel wire. The wire passed over two wheels of a device attached to one end of the pool. One wheel generated an impulse for every cm of forward movement and another wheel produced an electrical signal which was directly proportional to V. Measurements of distance and time were begun at definable points in the stroke cycle and were discontinued at the end of a predetermined number of strokes. In all of the four competitive strokes, front and back crawl, butterfly, and breaststroke, the V increased as a result of increasing the stroke rate (S) and decreasing the distance per stroke (d/s). In the front crawl, the male and female swimmers who achieved the fastest V had the longest d/S at slow S. The faster male swimmers also had greater percent decrease of the d/S at their maximal V than did the less skilled persons. The back crawl was similar to the front crawl except that maximal S and V were less. Increases of V of the butterfly were related almost entirely to increases in S. Except at the highest V, d/S was decreased somewhat. In the breaststroke increased V was also associated with increasing S, but the d/S decreased much more than in the other stroke styles. Fluctuations of velocity during the stroke cycle were least in the front and back crawl (+/- 15--20%) and greatest in the butterfly and breaststroke (+ 45--50%). The results were compared to the S observed and the values for V and d/S calculated for a large group of swimmers competing in the 1976 U.S. Olympic Trials. The implications of the findings for coaching swimmers are discussed.
Sprinters at the 1996 Atlanta Olympic Games did not hear the starters gun through the loudspeakers in the starting blocks.
  • Julin
Julin L, Dapena J. Sprinters at the 1996 Atlanta Olympic Games did not hear the starter's gun through the loudspeakers in the starting blocks. New Stud Athl. 2003;18(1):23-7.
Bienne: Omega SA; [cited
  • Omega Timing
  • Web Site
Omega Timing Web site [Internet]. Bienne: Omega SA; [cited 2013 Aug 5]. Available from: http://www.omegatiming.com. http://www.acsm-msse.org