Category Archives: Michael Hickson

Are You Good at Chess?

This happens to me regularly. I’m on a bus working on tactics puzzles on my phone or I’m in the bookstore browsing a book on endgames, and some stranger will see me and ask: “So…are you good at chess?”

I never know how to answer that question. First, I’m not sure what the person is asking. Do they wonder whether I can beat them, or most people like them, at chess? Or do they wonder whether I can beat top players at chess? Second, I just started playing chess about two years ago, and after three tournaments my provisional rating with the Chess Federation of Canada is 1115. In my experience that’s good enough to beat most casual players most of the time, but it’s still low enough that I’m a below-average club player. To casual players I seem very good, but to tournament players, I look like a complete beginner.

Some will want to dismiss the question “Who is good at chess?” as meaningless or a matter of perspective to which the only possible answer is, “It depends whom you ask.” But I think the question is an important one, and that when it comes up (as it often does on various Internet forums), we should try to give as clear and as justifiable a response as possible, for two reasons.

The first reason is that every beginner wonders whether he is good at chess. We like to be good at the things we work hard at. We stick with those things, while we drop other activities for which we believe we have no talent. Believing that we are good at chess, or that we might one day be good at chess, is an important motivator for us beginners. It would be nice for us to know where the bar lies. Currently in the chess world the only clear benchmarks are the various chess titles obtained by a tiny fraction of all chess players: Expert, Master, Grandmaster, etc. Surely these are all categories that lie far beyond the humbler title “good at chess,” which ought to describe more than the top 2% of players. The absence of consensus in the chess community over what counts as “good”–or is it the snobbish unwillingness to concede that the term might mean anything less than “Master”?–is a motivational stumbling block.

The second reason is that public chess organizations, whether in schools or clubs, need a goal, something they can promise to the students they teach. That goal should not be to produce future Masters—no public program can promise to achieve something that depends on so many factors outside of its control. Yet, once again, in the absence of clear benchmarks below the chess titles, what else can a chess program aim at? In my opinion, the obvious baseline goal of every chess program should be to produce good chess players. So we need consensus over a definition of “good at chess” and the more specific we can be the better, both for individual students and for organizations.  How can we define this term in a way that avoids the problem of perspective?

Here’s my method: The definition of “good at chess” should strike a balance between two competing intuitions. On the one hand, you are good at chess if you can beat the majority of chess players in the world. This intuition will lead to a low bar for “good at chess,” probably somewhere around 1000 in FIDE’s rating system. On the other hand, you are not yet good at chess until you are taken seriously by the game’s Experts. This intuition will require the bar to be higher. What is the Goldilocks rating that captures both intuitions—not too high to be unattainable by most serious players, but not too low to be laughable by the standards of the game’s Experts?

Here’s a suggestion. Let’s say that if your rating is just high enough for us to expect you to beat chess Experts some percentage of the time, then you should be considered “good at chess.” After all, if you can be expected to take a percentage—any percentage, even just 1%—of your games against an Expert, then surely this is a good reason for the Expert to take you seriously. And surely the rest of us should count you as a good chess player.

If you share my intuition that being just good enough to expect to win 1% of the time against a chess Expert is a reasonable criterion for being “good at chess”, then who is good at chess? Well, first we need to know who, exactly, the chess Experts are. In most federations “Expert” is an informal term whose official version is “Candidate Master.” In FIDE, the lowest rating bar for this title is set at 2000. So if we take 2000 to be the lower limit for chess Experts, then who can expect to win 1% of their games against a 2000 player? The answer can be easily calculated using the ratings tables available on FIDE’s website: a person whose rating is 620-735 points below their opponent can expect to win 1% of their games against that opponent.  So…

Here’s the answer you’ve all been waiting for. Who is good at chess? By my reasoning it’s anyone whose FIDE rating is at least in the range of 1265-1380.

This range is both low enough to capture the first intuition and high enough to capture the second. Somebody who is rated above 1265 will crush casual chess players. Such a player won’t exactly strike fear into the 2000 Expert, but he will make the Expert work, and can even expect to win a rare game against him. I cannot imagine any other sport, art, or discipline in which giving the Experts a run for their money wouldn’t be enough to count as “good”!

Now, I’m not suggesting that FIDE institute a new title, “Good at Chess”, for 1265+ players.  We don’t need new official titles, just new ways of presenting the game and its culture to the broader chess-playing public.  On a practical level, I’m recommending a way to talk about the qualitative meanings of ratings and to set minimal goals for chess programs.

I can hear the objections pouring in, and the debate over who is good at chess will inevitably go on. But to conclude, here is the perspective of International Master Jeremy Silman on players whose ratings lie in precisely the range I’ve argued as constituting “good at chess”:

“I remember going to my first tournament at age twelve. It was all quite magical, and as I watched other players’ games in the under 1600 section I recall being amazed at their skills—skills which were far beyond anything I could fully understand at that time. Indeed, my view of 1200-1399 players as being demigods is not that far out of line…If he plays in tournaments, he holds his own against many experienced players. If he competes against non-tournament playing friends, he most likely dominates them” (Jeremy Silman, Complete Endgame Course, Part Three).

If an International Master thinks 1200+ players can be called “demigods,” then I would say it’s safe for the rest of us to call 1265+ players “good”!

Michael Hickson