This is the last installment in a four-part series where I am using various statistical techniques, applied to my brand-new Chessmetrics data, to explore the following question:
Was Garry Kasparov the most dominant chess player of all time? If not, who was?
In previous installments we have looked at several metrics for evaluating who was the most dominant player of all time. It seems fairly clear that Bobby Fischer established the largest gap between a #1 player and the rest of the world, but that was only for a few months and then he retired. Emanuel Lasker, on the other hand, had the longest total duration as world champion, as well as the most total months at the top of the rating list. However, both of those included long stretches where he was on top through inertia, rather than through actively and frequently defeating his contemporaries. Not that Lasker necessarily had an alternative, given the times he lived in, but perhaps his durations are not strictly comparable on a one-to-one basis with the durations of excellence achieved in more recent times by Anatoly Karpov and Garry Kasparov. I think if you are trying to find a balance between peak rating gap, and overall career duration of dominance, the best candidate has to be Garry Kasparov. However, I do have another way of looking at all of this...
Each year the Russian chess magazine "64" organizes a vote among the chess community to determine the winner of the Chess Oscar, the "best chess player award". It is always interesting to see the results of the balloting, and the winner always seems like a very reasonable choice, subjectively speaking. Recently, however, someone suggested that I should run some calculations to see whether the Chess Oscar results also seemed reasonable, objectively speaking.
I thought this was a great idea, not the least because it gave me an opportunity to use my new rating formula once again. It is flexible enough that I actually use the exact same formula for the overall rating calculation, plus the single-event performance ratings, plus anything else in between. So I applied my formula on a yearly basis, to see who had the best performance rating in each calendar year. And based upon those rankings, I figuratively awarded a gold medal, silver medal, and bronze medal to the top three performers each year.
This idea actually has a lot of practical appeal. Rather than a scheme like the ACP tour's scoring system, where the points have to be awarded in a somewhat arbitrary (and thus controversial) fashion, why not just pick a performance-rating based approach that also rewards activity? My rating formula does exactly that (you can read more specifics on the Chessmetrics site). And while my official rating lists are based upon a weighted performance over the previous four years (the duration that was found to be most accurate at predicting future results), there is surely a desire for some kind of metric that rewards recent results in an extremely dynamic way, far more dynamically than the existing FIDE ratings.
So, why not go with one year? We could calculate a yearly performance rating, across all of a player's game results during the year, and then rank everyone during the year based upon those yearly performance ratings. Maybe at the end of the year, that ranking would determine the "yearly champion", or maybe it would even determine the automatic seeding into some sort of "yearly world championship" tournament.
Another neat thing about this is that there is no opportunity for a player to just sit out the action and maintain a high rating forever and ever. At the beginning of each year, your yearly performance rating resets again, and you have to start over from scratch. And because the rating formula rewards players who play a lot of games, there would still be incentive to keep playing even if you did manage a fantastic result in your first event of the year.
It's hard to show off the dynamic nature of this measure in recent years, exactly because of the incredible degree to which Karpov and then Kasparov have dominated chess for the past three decades. But if you look at who wins the gold medal each year, going all the way back to the 1840's, you'll see how dynamic the list of winners can be:
|Decade||Year 0||Year 1||Year 2||Year 3||Year 4||Year 5||Year 6||Year 7||Year 8||Year 9|
It is incredible how closely this list matches with the historical results from the (totally subjective) Chess Oscar voting. Over the past ten years, they only disagreed twice. In 1995, Kasparov won the Oscar vote with Karpov finishing second, whereas Karpov had a gold-medal yearly performance rating 13 points ahead of silver-medal-winner Kasparov. And in 1997, Anand won the vote with Kasparov finishing second, but Kasparov had a gold-medal yearly performance rating 19 points above that of silver-medal winner Anand. The other eight years, the Chessmetrics gold-medal winner matched the Chess Oscar winner.
And before that, they matched even better! In fact, there was a perfect match
every single year from 1973 all the way through 1988, at which point the original
Chess Oscars stopped because of the death of the founder. That means the two
approaches have agreed on who was the most successful player of the year, for
24 of the past 26 awards! It is interesting to note that Garry Kasparov had
the best performance rating for every single year in the six-year stretch from
1989 through 1994 when the Chess Oscar was not awarded.
Let's say we were to abolish the current tradition of having the world championship determined by a match. In fact, let's pretend that there never was such a tradition. We'll pretend chess turned out to be more like how golf, or tennis, works currently. If they'd had Chessmetrics yearly performance ratings available way back in the nineteenth century, perhaps they would have determined the world championships based on yearly performance. If the world championships had always been determined in this way, with gold, silver, and bronze medals awarded each year, we wouldn't be talking about the 27-year reign of Lasker, or the injustice of various players never getting a chance (or a second chance) at the title. We might instead be marveling at the seven-year streak of gold medals won by Kasparov from 1988 through 1994, or the record-breaking (at the time) five-year stretch of gold medals won by Karpov from 1973-1977. We might even be waxing nostalgically about the famous "three-peat" jinx in the yearly chess performance race; just look at that list above and see how long it took, and how many failed tries, until someone finally won three straight years. And instead of memorizing world championship dates, we might have memorized these kinds of numbers instead:
For each player, you can see how many yearly gold, silver, and bronze medals they would have won. It includes the breakdown of medal types for everyone who ever won five or more medals in their career. In addition, the next tier of players, with two, three, or four career medals, is listed at the bottom. By the way, in that lower picture, that's José Capablanca on the left and Emanuel Lasker on the right, in case you didn't recognize them. The photo is courtesy of chesschamps.com.
There is so much to be gleaned from this graph that I encourage you to just stare at it for a while and see what you notice. It really is a different way of measuring the accomplishments of the most dominant players in history, but it's also a very good way. One nice thing about having silver and bronze medals is that it leaves room for two or three truly dominant contemporaries to still be rewarded for their excellence, without losing sight of who the top-performing player actually was. I think it is incredible that neither Karpov nor Kasparov ever managed a bronze medal in a single year; it's because the two of them were too busy winning the gold and silver each year. In fact, and this deserves a paragraph of its own:
For a fifteen-year stretch from 1981 through 1995, Anatoly Karpov and Garry Kasparov combined to win all fifteen gold medals, and fourteen of the fifteen silver medals! The only player to briefly join the exclusive K-K club during that time was Vassily Ivanchuk with a silver medal in 1991 thanks to three different 2800+ performances during that one year (out of the six 2800+ performances that he has had in his entire career).
As long as we are pretending things, let's try another fantasy question: what would have happened if Garry Kasparov had never become a serious chessplayer. If Anatoly Karpov had still maintained his same ability and same overall results that he did in real life, then I think it would be a foregone conclusion by now that Karpov was the most dominant chess player of all time. He would have far surpassed almost all of the accomplishments of Emanuel Lasker, except those that were artificially extended due to the infrequency of play during Lasker's time. In fact, had Karpov defeated Kasparov in their first world championship match, it would almost certainly have eclipsed Fischer's main claim to all-time fame, which was his 6-0 match scores against Mark Taimanov and Bent Larsen. Even if Karpov had waited until Game 48 to reach his sixth win by a score of 6-2, it would still show up right now as the best match performance rating of all time, better than Fischer's. And if Karpov had managed to win the match by a 6-0 score before Kasparov reached his first win, then it would have gone down as the only 2900+ performance rating in Chessmetrics history (as long as the match didn't last longer than 66 games!) Imagine the weight of chess history resting on those players' shoulders at the time, had they only known what was at stake... (I'm joking!)
Despite the possible interpretation of "dominant" as suggesting that you must be number one, I nevertheless think you can still be "dominant" even if there is another person who is also very dominant. As an example, in late 1988 there was a 96-point Chessmetrics rating gap between #2 Karpov and the rest of the world, with the #3 spot fluctuating among Valery Salov, Alexander Beliavsky, Vassily Ivanchuk, and Jan Timman. That is easily the biggest gap between #2 and #3 of all time. Was Karpov a dominant player then? Ask his opponents.
I don't think it seems right to penalize Karpov for happening to be a contemporary
of Kasparov. So I would actually place Karpov above both Lasker and Fischer
in the all-time annals of who was most dominant. If you removed Kasparov from
the picture, think how many Karpov silver medals would turn into gold medals.
Eleven, in fact. Look back at that graph and change eleven of Karpov's twelve
silver medals into golds, and remove Garry Kasparov, and then tell me that
anyone else before Anatoly Karpov's time was as dominant as Karpov was, if
not for Garry Kasparov. It may seem too outlandish to talk about "removing
Garry Kasparov from the picture", but I think it does help to clarify
the issue. There simply is no other pair of chess players in history who were
so jointly dominant. There were fourteen different years where they Karpov
and Kasparov, between them, had the two top overall performances for the year.
The next pair who were most "jointly dominant"? Anatoly Karpov and
Viktor Korchnoi, with five years where they won both the gold and the silver.
And of course, once we stop the pretending, and acknowledge that Kasparov did in fact compete, and dominated even the mighty Karpov, then I think it's a no-brainer to answer the overriding question of these articles. If I had to hand out medals for who were the most dominant players of all time, I would give the gold medal to Garry Kasparov, and the silver medal (fittingly) to Anatoly Karpov. And then the bronze medal goes to either Emanuel Lasker or Bobby Fischer, depending on the fine print about whether the most important timeframe is their whole career or their peak year. Admittedly, I think it's pretty clear that for about a year, Bobby Fischer dominated his contemporaries to an extent never seen before or since. It's also clear that if you exclude Kasparov and Karpov from consideration, Emanuel Lasker was number one in the world longer than anyone else, and moves up to the top of the list on several other graphs you have seen throughout the course of these articles. Who deserves the bronze medal. Fischer or Lasker? Lasker or Fischer? And the debate rages on…
I hope you have enjoyed these articles. Please send me email if you'd like to chat about them. In conclusion, let me just take the opportunity to wish Garry Kasparov well, and to say thanks for all that he has done for chess during his competitive career. I've greatly enjoyed playing over his games, following his accomplishments, and reading his books, and I can still do all of those things. As much as I would love for him to remain an active player, I can certainly understand and respect the desire to move on to bigger and better things. Let us hope that this is not just the end of one great story, but also the advancement of another, even greater story.