Chess Ratings - How They Work

• erik
• | Aug 23, 2007
• | 102749 views

Like it or not, we ALL have a chess rating. You may not care at all about your rating, or you may be whining every time it goes down in the slightest. You might be someone who plays a game a year, or someone who plays 1,000 a day. Still, there is a number out there that represents how well you play chess. Well, that's the theory, anyway.

To understand chess ratings you have to understand two things: #1 - that you have a TRUE rating that perfectly represents your strength of play, and #2 - that that TRUE rating will never be known and so we have to use statistics to get as close as possible to the truth. I'm writing this article in response to many people who ask about ratings and need a simple explanation of how they work. (I only know about all this because of a recent super-in-depth statistics course I took and my research in building Chess.com!)

There are two main rating systems, and each one has its merits.

The Elo System (used by the United States Chess Federation, FIDE, and many other online chess sites) is popular for two reason - it has been around for a long time, and it is simple. The idea is this: given two chess players of different strengths, we should be able to calculate the % chance that the better player will win the game. For example, Garry Kasparov has ~100% chance of beating my 4-year-old daughter. But he may only have a ~60% chance of beating another Grandmaster. So when playing that other Grandmaster, if he wins 6 games out of 10, his rating would stay the same. If he won 7 or more, it would go up, and 5 of less, his rating would go down. Basically, the wider the spread of the ratings, the higher percentage of games the higher rated player is expected to win. So to calculate a person's rating after playing a few games you calculate the average ratings of his opponents, and then how many games he was expected to win, and then plug it into a formula that spits out the new rating. Simple enough. Well, it turns out, that is maybe TOO simple.

I wish there was some simple analogy to explain all this, but there isn't. It all comes back to this: you have a theoretically exact chess rating at any given moment, but we don't know what that is and so we have to use math to estimate what it is. There are really smart people out there who work on this stuff for a living, and at the end of it all we get to put their proven methods into our code so that we can all enjoy knowing what little numbers next to our name we deserve.

If you want to read more, check out these articles (WARNING - SEVERE NERD CONTENT AHEAD):

• 4 years ago

Read the last one again, this tie ignore spelling errors (I type too fast) and replace gold with golf!

Cheer

The rascal!

• 4 years ago

Chaps, not sure what your beef is.

Golf has a similar system for handicaping. There is an assumption that knowing your opponents handicap prior to play can have an impact on your play - an expectation factor! However, a good golfer can spot a high handicapper the moment he swings the club - it is the same in chess, 4 moves in and you know.

ratings will never be accurate, they are there merely to encourage improvement. You only get better playing those at your next target level.

RD is based on frequency of playing and also how recent. If you play alot and recently then your RD is low, assuming therefore that your rating is likely to be accurate.

It is the same with a golf handicap. The more you play the more consistent your score will be and the more accurate (and lower) your handicap. A low handicap golfer, following a bad round only loses 0.1 from his handicap whereas a high handicap golfer could lose 0.4.

From what I can see the RD in chess is allowing for the player ability. Of course in chess as in golf there are other factors, an oponent! In golf this can be another golfer OR the weather conditions. If playing (as the professionals do) against PAR of the course, in handicap terms this differs according to the overall scores of the day, so reflecting the weather conditions and this has an adjustment on the handicap movement.

In chess, the opponent may have a high RD (less accurate rating) so win or lose, your own rating will hardly chang and visa versa.

In short, play more and play people at your rating or 1500 above and no more. target your next level - i did that in golf and play of to 5 these days, but did achieve 3. Now that I am to old to swing (hey, hey!) a golf club well, I turn to chess and will apply the same improvement rules.

Just to finish, it's worth pointing out that that gold, despite its accepted handicap system across the globe - 'its still full of bandits!'.

• 4 years ago

don't play if the opponent has a lower rating than you, if you lost the game huge deduction in your rating, if your opponent has higher rating, if you loss minimal deduction in your rating, that what I experience in this site.

• 4 years ago

How do the actual point ratings compare when using the "Elo" and "Glicko" systems?

Would a beginner be rated at 1200 with both?

• 4 years ago

Hi,

The system makes sense to me and I like the critiques I've read, especially JollyPlayer's one which challenges the model. Either way, it sounds as though the Glicko System gives a reasonable methodology for determining ratings and gauging relative ability.

BUT, I do agree with one reviewer that it's important that a player's rating at the beginning of a game should be used. This avoids the possibility of someone using gamesmanship when multiple games are being played, with them deciding that their rating could be affected by either a quick or delayed resignation.

Just as importantly, if the ratings at the beginning of a game are being used, this should be clearly advertised to avoid any such gamesmanship.

Skanski Fiddler

Numbers nerd and part time bon viveur

• 4 years ago

make sense!

• 4 years ago

Are you guys berserk?! I am an octogenerian and I should be able to make out some sense in your system, sadly the whole thing seems to be one huge nonsensical mess.Some people have asked valid  questions but they are not being answered, what kind of site are running? How about some answers !

• 4 years ago

I don't know if anyone of you knows about Statistics methods and concepts... There's a new theory called Item Theory Response, which explains the relativity of the scores always depends not only by the instrument ( or formula ) but also a determinative features of the Sample... It is well said above by other participant that there is a problem with the K-Factor. Indeed, FIDE scores are really so stronger... the same Glicko System loses Internal Validity or verifiability criterion when we use the same validation system in two different samples. Then, according to IRT Concept, a measure instrument has only validity within a determinate context.

Best regards...

• 4 years ago

wow!

• 4 years ago

How do the rating adjustments work during live chess? I have played players with a rating 150 points above mine and received no alteration to my rating when they resigned well into the game. In other instances, I have seen a change of 20-40 points for a resignation from another player much closer to my rating. Do the different ways of winning (by time, resignation, or checkmate) take different effects on my rating?

• 4 years ago

Groovy Baby, Yeah!

Jollyplayer triggered an interesting thought if I may quote:

"Glickman's adjustment is an adjustment for time.  His hypothesis?  The person who has been rated the most recently probably has a more accurate score than someone who has not played recently."

Couldn't the data being collected on the performance of players on this site be used to test the validity of the hypothesis?  What insights can we gain in learning more about RD factors in other tasks requiring analytical and or creative thinking?

• 4 years ago

Haha lol the severe nerd content :p

• 4 years ago

Why not the folowing:

If the assigned opponent shows blank - no rating, don't display the game creator's either.

Alternately, display not rated,or not known

• 4 years ago

One (quite important) thing this article doesn't explain is whether chess.com uses the ratings of both players *when the game starts* to calculate the final deltas, or the ratings when the game ends.  I feel the former is by far the fairer alternative.

Of course this choice only exists in correspondence chess or similar situations when one player can play multiple games at once (and get them rated).  It doesn't occur in classic OTB chess.

I'd be grateful for someone to elaborate on this, and yes, it's personal.

• 4 years ago

wow! Now I understood!

• 4 years ago

Nice,  things are a bit clearer.

• 4 years ago

The Statistical Data & Probability in Chess Rating & Chess period will become more Accurate when  " No Draw " is a Rule & Chess is only Win, Lose,or Stalemate. I Love the Game of Chess & this is my Thought OK

• 4 years ago

Hey! I  for some odd reaqson 150-250- ratings points were illegally taken away from me. I sent you this complant .about two weeks ago! When do get an answer or a resolvement?

• 4 years ago

Well, at least there's an explanation how the players are rated. Actually, i couldn't care less about my rating. I play chess to enjoy and feel the thrill. If possilble, I would like to win, but  you can't beat them all. Besides, I don't think I could make it to the Master's level. So, after all the fuss about ratings...have fun!!!

• 4 years ago

after reading the article, I still have questions.  I am fairly active, usually have about 10 games going, yet my rating seems very volatile.  I have seen my rating jump 40 points while playing someone close to my own rating. for instance, I just beat a player and my rating went up 15 points.  his rating was very close to mine when the game started, but at this point, I've been on a winning streak, and have about 300 points on him. which brings up my second question:  is the rating difference calculated at the beginning of a game, or at the end?  it would seem like it shouldn't matter, but if it is at the end, that brings up gamesmanship issues, where someone might prolong a lost game, or resign quickly, depending on what he thinks might happen in other games I have pending.

thanks.

jr