Time and Time Again
It happens after every Slam: The old argument about three sets versus five sets.
Actually, there are two arguments: One about number of sets, and one about who should be paid more ("Men should be paid more because they play best of five" or "men should be paid more because they're better").
The latter arguments are pure economic nonsense. "Better" is a relative term -- and in any case, we don't pay for being "good" at something. A bad bank executive earns more than a good janitor -- indeed, in these days of insider trading, a bad banker may earn more than a good banker! This apart from the fact that "better" tennis is a relative term -- some people think "better" means a particular style, not raw ability to hit or run or whatever.
Similarly the argument about men playing best of five. Players are not paid by the set -- and be very glad they aren't, because if they were, the opportunity for collusion would simply be too great: They'd play three sets instead of two, or five instead of three, just to get the extra money. We see this: Consider how many exhibition matches wind up with the players splitting the first two sets.
It's also wrong because players aren't paid to win -- not directly. Performance isn't keyed to what you do -- if it were, a win at a Slam wouldn't be worth dozens of times more than a win at a Challenger. What players get paid for a match is, in effect, an agreement between them and the tournament, or between them and the Tour. This is why some tournaments have a million dollars in prize money, others far less. This means, yes, that the Slams have the right to pay men more than women -- but they also have the right to pay them less -- as we see at assorted co-ed tournaments. Estoril pays the men more, for instance, but Indian Wells and Miami pay them the same -- and Sydney and Bahia pay the women more. It's just the deal they strike. If the players don't like it, they don't have to play Sydney, or Estoril -- or Wimbledon. From where we sit, equal pay seems only logical, because both men and women are part of the Slams, and without either, it would be a lesser event. But economically, all such arguments are null. Any claim otherwise is based on non-economic arguments -- usually, we suspect, what the watcher likes better.
Which leads to the question of which way is "better for fans": best of three or best of five.
This is much more intricate, because arguments run both ways. Best of five matches can be more dramatic; there is the possibility of a five-hour match, and of a comeback from two sets down. Some people like this. Others -- the author is one -- can't stand sitting still for that long, and much prefer seeing two two-hour matches, involving two different sets of players, than four hours of the same two guys playing the same shots over and over. The author doesn't know which view is more common. He does know that, every time the question comes up, there are significant numbers of people who argue each side. Changing would probably upset more people than it makes happy.
From that standpoint, the current situation is actually ideal, or nearly. If you assume that tennis is a unified sport (usually anathema to either the men or the women, depending on which one is more economically successful at the time), then having one gender play best of three and the other best of five is only right and fair: It lets those who like best-of-five watch best-of-five, and those who like best-of-three watch best-of-three. Any change is going to upset some people and make others happy -- but it will cause more upset than happiness; making (e.g.) the women play five sets will make the fans of five sets slightly happier, but only slightly, and it will make the fans of best-of-three very upset. In effect, the current system is a compromise that everyone can tolerate; why change it?
And it's better that the men be the ones to play best-of-five, simply because (we're told, and we believe based on experience) close women's matches tend to run longer proportionally than men's close matches. This because a close men's match tends to have a lot of easy service holds, while a close women's match has a lot of break points. If we made the women play best of five and the men best of three, it would increase the differential between the longest and shortest matches. This is definitely a bad idea; the TV networks would hate it.
But there is another reason why we can't have everyone play five sets, and it's a legitimate economic issue: Court space.
Mathematically, a "short" best-of-five match (settled in three sets) is half again as long as a "short" best-of-three (settled in two sets). A "long" best-of-five (one that goes to the fifth set) is two-thirds longer than a long best-of-three. That's a lot of extra sets.
But what is the exact balance point? That, really, is the point of this column. We decided to test it -- to take all of the Slam semifinalists from 2001, and see about the length of their three-set versus five-set matches in that year. (We pick the Slam semifinalists because they have significant numbers of five-set matches to measure.)
The players we're looking at are:
Juan Carlos Ferrero
Note: The ATP shows different won/lost figures for many of these players, due primarily (we think) to Davis Cup. These figures are for point-bearing (ATP and Slam) matches.
In any case, here are the results for each player, showing number of matches in each of our five categories: three-set matches settled in two sets, three set matches settled in three sets, five-set matches settled in three, five-setters settled in four, and five-setters settled in five:
No doubt, if we had more time, we could mine a lot of strange information out of this. But we're here specifically to determine the ratio of lengths of best-of-three matches vs. best-of-five. So let's work that out. The table below shows the player, the average number of sets for a best-of-three match, the average for a best-of-five, and the ratio.
Here, again, we could have lots of fun -- e.g. noting that Agassi, Rafter, and Ferrero had the shortest three-set matches, while Corretja and Safin had the longest; but that Agassi, Grosjean, and Corretja (!) had the shortest five-setters, but Ferrero and Henman the longest. But this is still not our point.
If we look at the ratio of three-set matches versus five-set matches, we find that the median ratio is 1.57; the arithmetic mean is 1.56.
That means, if the women were like the men, that if the women are like the men (and we don't know that they would be, but it's likely), then over the 127 matches that constitute a Slam, they would play about 170 extra sets. At 35 minutes per set, that's 99 extra hours of court time. But it's worse than that. 96 of those 127 matches take place over the first two rounds. In other words, over the first two rounds, we're adding 127 sets, or 74 hours of court time. That means, even if we spread the matches over five days instead of four, a Slam would have to add at least one and probably two courts to allow for the extra matches. Plus security, etc. That's a lot of space, transport, and personnel expense.
Plus, one nice thing the women's game has is actual top players who play doubles. Make them play best-of-five, and that vanishes.
What it all says is, the current way -- men playing best-of-five at the Slams, women playing best-of-three -- is the best compromise. No one is perfectly happy (e.g. the author still wishes the men would play best-of-three), but it's a situation everyone -- including the tournaments -- can live with.
Which, in turn, argues for equal pay, just so that we stop arguing over the issue.