Are We Actually Getting Better at Chess?

How do you tell if we’re getting better at a sport or a game, when you can never pit players from different eras against each other? For example, could Tom Brady carve up Pittsburgh’s Steel Curtain defense of the 1970s? Who would win one-on-one, Michael Jordan (in his prime) or LeBron James? Could Justin Verlander strike out Babe Ruth? Outside of video games, we’ll never know.

The obvious exceptions are track and field events, where accomplishments are measured in time and distance. And in those cases, we actually have been consistently chipping away at records, running faster, jumping higher etc. Though the recent use of performance-enhancing drugs has certainly tainted that “progress.”

But what about chess? Players are judged by a sophisticated rating system, though there’s a thought that scores have been inflated recently. A pair of academics have set out to address this by comparing the quality of play over the years. In their recent paper, Kenneth W. Regan, a computer science professor at the University of Buffalo, and Guy Haworth, an engineering professor at the University of Reading, examine the quality of players’ moves, rather than win-or-lose outcomes. Their conclusion is that yes, we are getting better at chess.

Read the full study here. The abstract is below:

This paper develops and tests formulas for representing playing strength at chess by the quality of moves played, rather than by the results of games. Intrinsic quality is estimated via evaluations given by computer chess programs run to high depth, ideally so that their playing strength is sufficiently far ahead of the best human players as to be a ‘relatively omniscient’ guide. Several formulas, each having intrinsic skill parameters s for sensitivity and c for consistency, are argued theoretically and tested by regression on large sets of tournament games played by humans of varying strength as measured by the internationally standard Elo rating system. This establishes a correspondence between Elo rating and the parameters. A smooth correspondence is shown between statistical results and the century points on the Elo scale, and ratings are shown to have stayed quite constant over time. That is, there has been little or no ‘rating inflation’. The theory and empirical results are transferable to other rational- choice settings in which the alternatives have well-defined utilities, but in which complexity and bounded information constrain the perception of the utility values.

At the outset, Regan and Haworth pose four questions, the two most fundamental of which are:

1. Has there been ‘inflation’—or deflation—in the chess Elo rating system over the past forty years?
2. Were the top players of earlier times as strong as the top players of today?

And here’s their conclusion:

…there has been little or no ‘inflation’ in ratings over time—if anything there has been deflation. This runs counter to conventional wisdom, but is predicted by population models on which rating systems have been based [Gli99].

The results also support a ‘no’ answer to question 2. In the 1970’s there were only two players with ratings over 2700, namely Bobby Fischer and Anatoly Karpov, and there were periods as late as 1981 when no one had a rating over 2700 (see [Wee00]). In the past decade, however, there have usually been thirty or more players with such ratings. Thus the lack of inflation implies that those players are better than all but Fischer and Karpov. Extrapolated backwards, this would be consistent with the findings of [DHMG07], which (like some recent competitions to improve on the Elo system) are based only on the results of games, not on intrinsic decision-making.


[HT: Tyler Cowen]

Mike B

Performance-enhancing drugs didn't "taint" progress, they are the very definition of progress!! If humans should not be allowed to use technical aids to improve their physical abilities then why are runners allowed to wear shoes? Why are they able to drink anything except water or food they haven't gathered or hunted with their own hands?

Anyway it's clear that chess playing will always improve because the skill of future players is built on the shoulders of past players. In fact any sport with a strong mental component will have that part improving over time. Humans don't just evolve through their changing DNA, thoughts and ideas, transmitted between groups and generations, evolve as well creating ever more sophisticated ways to play and win a game. The best part is that at some point Chess will be completely solved such that no further improvement is possible and at that point our civilization can declare that mission accomplished and shift resources to mastering other tasks and skills.



Yes. Absolutely correct. The way a chess amster prepares, using book and software, which did not exist few decades or years back, does give advantage to current players; The number of chess players with 2700 ratings exceed 40; See it on
Now even 2800 are 4-5 players.

Joshua Northey

This would seem pretty obvious. To hold otherwise would be to hold that chess scholarship/theory/whatever hasn't advanced in 40 years, which is a rather silly contention.


"Though the recent use of performance-enhancing drugs has certainly tainted that 'progress.'"

Only if you buy into a lot of mythology. The use of performance-enhancing drugs is not recent. People have been using them as long as they've been around. Amphetamines, for example, were very popular among 19th-century cyclists. Early Olympians were known to use various stimulants to enhance their performances.

And why focus on drugs? Modern runners have track shoes and training techniques that earlier runners (many of whom ran barefoot) could never have imagined. You might even say the incentives have improved: a modern successful athlete, even an amateur, can look forward to worldwide fame and thousands upon thousands of cheering fans at every event. Might that "enhance" a performance too?


Jordan would beat James.


If you ever watched the ESPN Classic Sports channel, it's obvious that the professional athletes of today are superior to those of the past.


Elo is a relative measurement, it measures how strong you are relative to other players of the time. It doesn't make sense to compare top players of different generations by Elo. That said modern players have a huge advantage in computer analysis and tactics trainers like the one at

Caleb b

One "sport" not racing. The last sub-2minute Kentucky Derby is still Secretariat. Odd, no? I think it's because the sport is fixed, but I might just be a disgruntled bettor.

John Cochrane

Long Jump has advanced nearly none since Beamon in 68. Mike Powell surpassed it in 91 and since then, nothing.


True, but long jump is an exception - which is due, I think, to the enormity of Beamon's result. At a rare olympic at altitude, with the wind exactly right (+2.0), hitting the board perfectly, Beamon had the jump of his life. With all those factors coming together, it was a freak event, absolutely shattering the then world record of 8.33.
If it hadn't happened - say his last step happened to be 1 cm longer - the record would have slowly progressed to Powell's 8.95, and the development in long jump would have looked normal.