How Often do Left-Handed Win?

We already took a look at the distribution of left-handed players in the rankings. Another approach is to simply look at how often left-handed players won matches against right-handed players. Or put in a different way, how winning percentages depend on the number on distribution of left-handed players.

One would expect that left-handed players (or pairs with left-handed players) win 50% of the matches assuming there are no systematical differences.

Methodology

We just count how many matches were won by the left-handed players. We include only matches were all players’ handednesses are known in the database and only matches that ended in a proper win, i.e. no withdrawal.

For singles it is rather easy, each match is either between two right-handed players, one left-handed and one right-handed player or two left-handed players. The second case is the only one we are interested right now. We count how many matches were won by the left-handed player, as well as how many rallies were played and how many were won by the left-handed player.

For doubles it’s a bit more complex. Each pair can have either zero, one or two left-handed players. So in total there are six possible combinations, where the pair containing more left-handed players is chosen to be the first one.

  • All players are right-handed, denoted by RR - RR
  • Only one player is right-handed, denoted by LR - RR
  • Both pairs have exactly one left-handed player, denoted by LR - LR
  • One pair consists of two left-handed players while the other one consists of two right-handed players, denoted by LL - RR
  • One pair consists of two left-handed players while the other one has exactly one left-handed player, denoted by LL - LR
  • All four players are left-handed, denoted by LL - LL

For mixed doubles it can be even more complex as we can differentiate between left-handed female and male players. Thus the list of combinations is enlarged to ten. We still sort the pairs such that the pair containing more left-handed players is chosen to be the first one. When one pair contains a left-handed male player and a female right-handed player, while the other pair consists of a right-handed male and a left-handed female player, the pair with the left-handed male player will be chosen as the first pair.

Data

The dataset includes matches from 2010 until today. For more information see the page about the Database.

Results

Singles

In the singles’ disciplines we see that about 53.2% (men’s singles) or 54.3% (women’s singles) of the matches between a left-hander and a right-hander are won by the left-hander. Correspondingly the left-hander won 50.5% (men’s singles) or 50.8% (women’s singles) of all rallies.

I don’t know why there are more matches won by the left-hander. Left-handed players are overrepresented in the higher rankings, but I would expect there to be more weaker players filling up the lower parts of the rankings.

It can also be seen, that the percentage of left-handers among male players is higher than for female players as the percentage of matches involving left-handed players is higher for men’s singles.

Men’s Singles

Combination Matches Won % Rallies Won %
all 67165          
R - R 50154          
L - R 15826 8419 53.2% 1302431 657650 50.5%
L - L 1185          

Women’s Singles

Combination Matches Won % Rallies Won %
all 51043          
R - R 42210          
L - R 8411 4570 54.3% 678184 344544 50.8%
L - L 422          

Level Doubles

The first observation is that pairs with exactly one left-handed player perform better than expected playing against purely right-handed opponents. The percentages for winning a match are 53.1% (men’s doubles) and 52.0% (women’s doubles) and thus comparable to the winning percentages in singles. The same can be said for the percentages of won rallies, where 50.6% (men’s doubles) or 50.4% (women’s doubles) were won be the left-right-handed pair.

Surprisingly pairs consisting of two left-handed players perform worse then expected. Both against two right-handed players and against a left-right-handed combination the winning percentages are clearly below 50%.

One explanation might be that left-handed players are also not accustomed to playing with a left-handed partner. However, it seems that one left-handed player per team is the ideal number for success.

Men’s Doubles

Combination Matches Won % Rallies Won %
all 26354          
RR - RR 17086          
LR - RR 8002 4252 53.1% 673407 340892 50.6%
LR - LR 876          
LL - RR 325 130 40.0% 26823 12902 48.1%
LL - LR 64 23 35.9% 5195 2465 47.4%
LL - LL 1          

Women’s Doubles

Combination Matches Won % Rallies Won %
all 21742          
RR - RR 14667          
LR - RR 6322 3285 52.0% 513667 258966 50.4%
LR - LR 666          
LL - RR 72 21 29.2% 5653 2567 45.4%
LL - LR 15 2 13.3% 1171 517 44.2%
LL - LL 0          

Mixed Doubles

In the mixed doubles discipline, we again observe that a pair with one left-handed player wins more than 50% of the matches, comparable to the other disciplines. The percentages of won matches (52.5%) and won rallies (50.4%) are comparable to the other disciplines.

But unlike in level doubles, a pair of two left-handed players does not perform worse. Teams of two left-handed players won 57.2% of the matches against a solely right-handed opponent pair and 50% against a pair that consists of one left- and one right-handed player.

Why the combination of two left-handed players performs better than in the level doubles is unclear. It could be due to the more tactical matter of mixed doubles, as there is already an important distinction between the two players of a pair, a difference in handedness does not pose an added difficulty. Thus it might be a lesser problem for mixed doubles players to play with a partner of unaccustomed handedness.

When we differentiate between male and female players, we find that in a left-right-handed combination, a left-handed male player increases the winning percentage more than a left-handed female player. Against solely right-handed opponents, a pair with a left-handed male player won 53.5% of the matches, while a pair with a left-handed female player won only 51.5% of the matches. Thus adding a left-handed male player increased the winning percentage by 3.5%, while adding a left-handed female player only increased the winning percentage by 1.5%. The difference is even more pronounced against a solely left-handed pairing. It has to be said though that the absolute numbers are fairly low, so it’s unclear if one can draw general conclusions from this. A left-handed male and a right-handed female player won 58.2% of the matches against two left-handed opponents, while a right-handed male and a left-handed female won only 38.3% of the matches.

It can also be seen that percentage of male players who are left-handed is bigger than the percentage of female players who are left-handed. For example of the matches with exactly one left-handed player, in 4024 matches the left-hander is male while only in 3865 matches the left-hander is female. But the difference is not as big as could be expected from the numbers of players in the database. As a side note, there seem to be more solely left-handed mixed pairings. While in level doubles there was only exactly one match where all four players were left-handed, in mixed doubles there were five matches with four left-handed players. Also the other numbers of matches including one pair with two left-handed players are higher for mixed doubles than for both level doubles disciplines.

Without Differentiation

Combination Matches Won % Rallies Won %
all 28713          
RR - RR 19364          
LR - RR 7889 4142 52.5% 661389 333628 50.4%
LR - LR 778          
LL - RR 563 322 57.2% 46923 23960 51.1%
LL - LR 114 57 50.0% 9616 4847 50.4%
LL - LL 5          

With Differentiation

Combination Matches Won % Rallies Won %
all 28713          
male R, female R - male R, female R 19364          
male L, female R - male R, female R 4024 2151 53.5% 336369 169910 50.5%
male R, female L - male R, female R 3865 1991 51.5% 325020 163718 50.4%
male L, female L - male R, female R 563 322 57.2% 46923 23960 51.1%
male L, female R - male R, female L 416 226 54.3% 34814 17604 50.6%
male L, female R - male L, female R 191          
male R, female L - male R, female L 171          
male L, female L - male L, female R 67 28 41.8% 5706 2829 49.6%
male L, female L - male R, female L 47 29 61.7% 3910 2018 51.6%
male L, female L - male L, female L 5          

Summary

It was found that left-handed singles players win more than 50% of their matches, in both singles disciplines about 53-54%. In doubles a combination of one left-handed and one right-handed player seems to be the most successful combination with a winning percentage against a solely right-handed pair of 52-53%. For a pairing of two left-handed players the observed patterns are contradictory: In level doubles pairs of two left-handed players seem to be at a disadvantage while in mixed doubles these pairs perform better than expected. However the numbers of matches with at least two left-handed players are still fairly low. A very successful pair of two left-handers could change the numbers in favour of solely left-handed pairs.

It was also found that a male left-handed player increases the winning percentage more than a left-handed female player.

The reason for left-handers winning more than 50% of the matches remains unclear. As the matches are taken from high level badminton tournaments around the world, one would expect an over-representation of left-handed players among the top players that generates a winning percentage of more than 50% for the left-handers. But at the same time, one would expect that for lower levels this should be decreased by introducing left-handed players who without their advantage due to their handedness would be at a lower level. Thus these weaker left-handed players should lose enough matches to decrease the percentage of won matches. It remains unclear whether the over-performance is due to the over-representation at the top level or whether this is an effect that remains at all levels of competition.