Another week, another Rally. This time Kamakshi and I talk about the controversial-to-laughable new computer GOAT survey, whether GOATs really do exist, and who we might want playing to save our lives. Here's my first post; she'll have her return tomorrow.
*
Kamakshi,
Weren’t we just talking about bonus points the other day? They may not be used on the tours anymore, but they made a big comeback this week when we got the news that a computer had decided that Jimmy Connors was the greatest tennis player of the Open era. Programmed by someone named Boffin Filippo Radicchi, it awarded bonus points for wins over highly ranked opponents. Connors’ longevity obviously helped him beat a high number of those players, more than, say, Roger Federer, who came in at an absurd No. 7, has had a chance to play. The programmer even said that younger players will be punished for not having played as many matches as the legends. This is clearly the case when it comes to Rafael Nadal, who may be the top quality-win player ever—he has 14 of them over the guy I think is the best of all time, and who has been ranked No. 1 for as many weeks as anyone.
Jimmy Connors may have been the most successful, over the longest period, of any Open era player so far. Nobody took as many punches and got off the mat as many times as he did. But he’s not the best of the last 40 years. The flaw at the digital heart of this program is that it rewards players for beating certain opponents, while the players themselves care about winning certain tournaments much more than they do about beating highly ranked players—it is kind of fitting that Connors, who loved to make matches personal, should fare the best under this system. Otherwise, it makes for some pretty good laughs: Guillermo Vilas, Bjorn Borg’s whipping boy, ends up six places ahead of Borg himself; ditto for Andre Agassi, who clocks in three spots ahead of his nemesis, Pete Sampras. And poor Ivan Lendl, who is ranked second, might be wondering what he had to do to get ahead of Jimbo—Lendl only beat him the last 17 times they played.
That a computer could be so off-base shows again the logical impossibility of any GOAT debate. In 2005, for the 40th anniversary of Tennis magazine, the editors did a countdown of the 40 best players of those four decades. We looked, roughly, at Slam wins, time at No. 1, and total titles. There were some misunderstandings along the way—we judged Rod Laver only on his record within those 40 years, from 1965 on, so he ended up being No. 8, I think, one behind Connors, a fact that set a lot of people off. And to make it more fun, and more trouble, we mixed men and women. It was obviously tough to compare them. In the end, we had to choose between Navratilova, Graf, and Sampras for No. 1. Graf and Navratilova had the stats (I tried my best to keep the thought of Seles’ stabbing out of my mind, even though it almost certainly elevated Steffi’s totals), but we thought Sampras’ achievement relative to his competition was greater, so we went with him. More outrage. Someone wrote to tell me that I had done irreparable harm to the sport by choosing Sampras over Graf—I wrote the Sampras entry, so I guess that’s why I was blamed. And it’s true, Steffi’s at-least-four-wins-at-all-four-majors stat is hard to argue with.
Later that year, a statistician and tennis fan who was outraged by the seemingly arbitrary way we went about this came up with a purely statistical method for finding out who “the best male player never to win a Slam” was. Except that, as with Boffin’s computer, a human had to decide which stats counted most and which could be ignored. In other words, it wasn’t any more “pure” or logical than six editors sitting around a table batting names back and forth. His stats ended up being weighted in favor of shorter careers, so Miloslav Mecir, who had his cut short due to injury, was determined to be the best never. Stats don’t lie by themselves, but they also don’t exist by themselves. Someone has to choose them, and everything is subjective after that.
The bigger problem with the GOAT debate in tennis is that its stats, like its players, don’t cross all eras. The Slams, while they are more valued now than they were in, say, the 1970s, have always been the most important events; every tennis player has dreamed of winning Wimbledon. But the majors can’t tell the whole story because, to take one prime example, Pancho Gonzalez, by consensus the best player in the world throughout the 1950s, didn’t play in any of them for 20 years. And, as we know, Laver didn’t play in them during five prime years of his career. There are no what-ifs allowed in this debate—we can’t speculate that Laver would have won 20 majors and have that number mean anything—but it does mean that Slams can’t really be the be-all and end-all, because that would mean anointing a greatest tennis player ever while ignoring the existence of Pancho Gonzalez and downgrading Rod Laver.
Which is pretty much what I did a couple of years ago when I wrote another Tennis magazine article claiming that Federer, with his 15th Slam win, had earned the GOAT title. Logically, there can be no such person, but I've always justified having the debate in other ways. Namely, our minds make these comparisons anyway; everyone is entitled to an opinion; we all know it’s a parlor game and just for fun; and why should we deny ourselves that fun, while also realizing that any GOAT claims we make are going to be flawed. I do consider Federer the best ever, by the measures we can use. But watching Pete Sampras on Monday, I thought again that at his peak, when he was at his best (I think of his 1999 Wimbledon win over Agassi), no one in history, including Federer at his best, could have beaten him. He had the ultimate first-strike (with the serve), quick-strike (with the forehand) game. If Sampras locked you down, you weren’t going to escape.
That, of course, just brings up another question: Is it possible for two players to be “at their best” against each other? If they’re evenly matched to start, it doesn’t seem like it would work that way in the zero-sum system of tennis. If one guy is bringing his best stuff, chances are the other guy is going to be struggling; if Sampras is on, even the world’s best returner isn’t going to be able to do much about it. The “at his best” argument favors the big servers and erratic shot-makers too much to be mistaken for a GOAT argument. Have Federer and Nadal ever both played their best against each other? The finest I’ve seen Nadal play against Federer was the 2008 French final, which he won easily. The best I’ve seen Federer play against Nadal were their two matches at the Masters Cup in Shanghai, which he won in straight sets (though their were moments in one of those, the one that ended 4 and 3 for Federer, when they both giving all they had at the same time; an awesome sight).
As for Jimbo, he inspired our friend Tom Perrotta, who wrote about this study for the Wall St. Journal, to bring up one other side debate in the GOAT debate: Who would you have play a tennis match for your life? He thought Connors was a pretty good choice, and this computer program does measure ability in individual matches. I agree that Connors makes sense, but first I have to come to grips with the whole idea in the first place: Why would someone be playing a tennis match for my life? Where would I be watching from? How nervous would I be? Most important, who would my designated choice be playing?
To “play for your life” you want someone tough, a gutty competitor who knows how to win; someone, above all, whom you can count on. Connors himself chose Gonzalez when asked this question, then later added John McEnroe to his list (he put a lot more trust in his old rival Johnny Mac than I thought he would). I might say Sampras, but he would have to be “at his best”; often he wasn’t. Jimbo is a solid choice, as long as he’s not playing Ivan Lendl. Or Borg. Or McEnroe. From that era, I would go with Borg over Jimmy; the Angelic Assassin was a killer in his own right. Federer? He’s the best, he's the GOAT—but for m**y life? When the chips were truly down? I’ll take Nadal.
How about you, Kamakshi, which player would you count on the most?
Steve