What question are we actually asking?
Watching a contest, it is easy to attach meaning to the most visible difference: one judoka is in blue and the other in white. But three different questions can hide inside 'does blue win more?' We might be counting victories by colour, asking whether the two colour groups contain comparable athletes, or asking whether changing only the colour would change an athlete's chance of winning. Those questions need different evidence. A count can answer the first while leaving the others unresolved.
This article addresses the research question, not what judogi a particular event requires. It is an explanation of published analysis, not a new analysis of every match video. The useful distinction is between a visible association and an effect caused by colour. Keeping that distinction clear makes a headline easier to evaluate and a statistic safer to quote.
What did the large study examine?
Dijkstra, Preenen and van Essen's 2018 study reports 45,874 contests from 162 tournaments. It compared outcomes around the 2011 change in which colour the first-called athlete wore. Open-weight and repechage contests were excluded. The abstract describes 2008–2014, but one body paragraph says 2008–2012; that inconsistency remains unresolved here. Before the change, the first-called athlete wore blue; from the 2011 World Championships, that athlete wore white.
The authors used mixed-effects logistic regression: colour and round were fixed effects, while tournament and the identity of the first-called athlete were random effects. This addresses specified sources of variation, not every possible confounder. This makes the paper more informative than simply adding up blue victories at one event, but the number alone is not a quality guarantee. A reader still needs to know how contests were selected, what an observation represents and which differences the model considered. A contest is not a unique person: the same judoka can appear repeatedly. Nor is a collection of international events a random sample of everyone who practises judo.
Sources: Frontiers in Psychology
Why does the colour assignment matter?
Imagine a completely fictional tournament in which stronger seeds are often placed on the blue side of the draw. Blue could accumulate more wins even if fabric colour changes nothing about perception, confidence or refereeing. Now imagine that the organiser reverses the colour labels while leaving the same athletes in those draw positions. A similar advantage could move to white. In this example, colour is a label attached to an existing difference, not its cause.
The example is deliberately invented and contains no measured result from the paper. Its purpose is to make the comparison problem visible. Before treating a colour total as an effect, ask what else travels with the colour: draw position, athlete identity, competitive level or the tournament's allocation procedure. Recording those conditions is more useful than assuming the two groups are interchangeable because the bouts take place on the same mat.
What do the reported estimates mean?
For the colour switch, Table 1A reports odds ratios of 1.04 for men (95% confidence interval 0.90–1.20) and 1.19 for women (0.96–1.47). Neither result was statistically significant at the paper's threshold. These are model estimates for the specified comparison, not blue's winning percentages.
An odds ratio should not be read as a percentage-point gain. A ratio compares odds, whereas a probability describes a share of possible outcomes. Without a starting probability and a clearly defined reference group, converting the ratio into a claim such as 'four per cent more likely to win' is misleading. For a general reader, the most useful takeaway is the uncertainty around the estimate and the exact question it addresses, rather than a dramatic number detached from its model.
Sources: Frontiers in Psychology
Does a nonsignificant result prove no effect?
No. Failing to detect a difference and establishing that two conditions are practically equivalent are different conclusions. A nonsignificant result does not license an unlimited statement about every athlete, every tournament format or every possible perceptual mechanism. Equally, uncertainty is not positive evidence that blue offers a hidden advantage. The reasonable response is to keep the conclusion at the level supported by the comparison.
When reading an interval, think about the range of estimates compatible with the model and data under its assumptions. Do not select only its most exciting end. A cautious summary can still be clear: this analysis did not establish the proposed colour advantage. That answer is more precise than either 'blue always helps' or 'science has proved colour irrelevant forever'.
Is being called first itself an advantage?
The original paper distinguishes a first-called-athlete association from a colour effect. That distinction does not mean a competitor can create an advantage merely by being announced first. Labels, positions and athlete characteristics can be related without any of them acting as a simple switch that produces victory.
A sensible explanation keeps several possibilities open until the evidence separates them. The athletes may differ before the contest, the event format may distribute them unevenly, or another condition may be linked to their assigned position. Listing alternatives is not the same as confirming them. When someone attributes a victory to the announcement order, ask what comparison would leave the other conditions sufficiently alike to test that claim.
Sources: Frontiers in Psychology
What did the later commentary add?
Cárdenes and colleagues' 2018 commentary discussed a possible referee-position explanation and described a separate duration analysis of 6,410 contests. Its referee explanation was a conjecture requiring further research, not a demonstrated cause of the first-called association. The 6,410 contests were analysed from the same datasheets used in the original study, so this is related analysis rather than an independent sample.
A commentary can be useful because it asks questions that a large dataset does not automatically resolve. But its status matters: it is not another independent experiment confirming the same effect. A duration analysis also cannot simply be substituted for a direct measurement of athlete ability. Short and long contests may raise useful questions; they do not, by themselves, tell us why a particular encounter ended when it did. Keep the commentary beside the original paper, with each one's role clearly labelled.
Sources: Frontiers in Psychology
Can we combine numbers from different papers?
Not merely because they concern judogi. First check whether the studies include the same event types, years, weight categories, sexes and rounds. Next check the outcome: total victories, a modelled colour-switch estimate and a duration distribution are different quantities. Adding their denominators together can make the evidence look larger while concealing overlap or incompatible questions.
A useful source comparison has four columns in your notes: what was observed, who or what was included, what was compared, and what remains unresolved. If two papers use the same underlying contests, treat them as related analyses rather than two independent samples. If a paper's dates disagree internally, preserve that discrepancy instead of quietly choosing the date range that makes the story simpler.
What should a spectator take from this?
Treat colour as an identifier while watching, not a complete explanation of the result. Describe what you can actually see before adding an interpretation: the athletes' positions, an attempted action, the opponent's response and the recorded outcome. A single successful throw cannot identify the effect of a judogi colour, just as one defeat cannot establish that a particular athlete lacked ability.
This is not an instruction to ignore psychology or perception. It is a reminder to match the claim to the observation. If you cannot see a grip clearly, say so. If you are discussing a disputed decision, distinguish the visible sequence from the rule interpretation. This article does not assess officials or provide a current competition-rule ruling.
How can you quote the finding responsibly?
Prefer a sentence that names the study, describes the comparison and retains its uncertainty. Avoid a headline that converts an observational result into a guarantee. When using a number, put the unit next to it: contests are not athletes, an odds ratio is not a percentage, and a study period is not an access date.
Link to the original paper so readers can inspect its methods and table, then use this explanation for the comparison and interpretation. Do not describe these figures as Judo Knowledge's own research. The unresolved date inconsistency should travel with a detailed dataset description; a citation should not make an uncertain record appear cleaner than the source. If a correction or a clarified dataset becomes available, update the explanation because the evidence changed, not simply to refresh the date.
What would make the answer stronger?
A stronger next step would make selection, allocation and the comparison transparent, examine alternative explanations and report uncertainty rather than only a winning total. It would also make clear which claims concern the historical dataset and which might apply elsewhere. More rows can help with precision, but they cannot automatically remove a systematic difference between the groups.
For now, the practical knowledge is how to resist an attractive shortcut: the colour we notice is not necessarily the condition that explains the result. You can understand that point without buying anything, changing your training or choosing a judogi for a supposed competitive edge. Use the source trail to inspect the evidence, and the glossary if an unfamiliar judo term interrupts your reading.
Sources
- Does Blue Uniform Color Enhance Winning Probability in Judo Contests? — Frontiers in Psychology. Sources checked:
- Commentary: Does Blue Uniform Color Enhance Winning Probability in Judo Contests? — Frontiers in Psychology. Sources checked:
Sources & editorial approach
We link factual claims to their original sources and distinguish research findings from interpretation. Study populations, dates and limitations matter.
Translations are prepared with AI assistance. They are not certified by a federation or independently reviewed by native-speaking judo experts. Articles do not replace qualified instruction.