22

This question is inspired by this extract from a question about minimum rating gain:

One of the top 10 players in the world (say Magnus Carlsen), wants to increase his rating to 3000. He has a good friend who has 1200 in rating, and who is willing to play a match of 200 games (2 each day for 100 days) against him. His friend will of course play his best, so this is by no means considered cheating.

According to https://wismuth.com/elo/calculator.html, Magnus will have way more than 99 % chance of winning a single game, but for simplicity, let's say he has 99 % chance of winning a game (which is an understatement). If we then forget about draws (for simplicity), Magnus would be expected a score of 198/200. That means a rating gain of 0.8*198 - 9.2*2 = 140 rating points.

Putting to one side, for the moment, the far fetched nature of the example, is there anything to stop this kind of behaviour?

I know two master level players who have achieved the 3 title norms they need for high level FIDE titles but are still about 150 points short of satisfying the rating requirement for the title. What stops them from following the suggestion and playing a 200 game match against a player rated 1000 points lower? Neither of them is getting any younger and the extra rating points they need seem to be getting harder and harder for them to gain.

3
  • 4
    It should be noted that ELO is designed so that schema doesn't work on average. The stronger player will get a few points each match with large probability, but he will lose a lot of points with little probability. In the long run losing and getting points is expected to stabilise the rating gap between the two players to their actual level - provided that both of them always try to play as well as they can.
    – Pere
    Commented Jan 25, 2022 at 10:45
  • 13
    @Pere the linked question shows why that is not true in practice: FIDE establishes a minimum rate gain of 0.8 for the winner. A pure ELO system without this minimum would give Carlsen 0.00032 rating points per won game, and stabilize the rating gap. The actual FIDE system gives the minimum of 0.8 points, allowing Carlsen to increase his rating indefinitely.
    – wimi
    Commented Jan 25, 2022 at 15:30
  • Just a small mathematical/probabilistic comment: it's a bit misleading to say (technically correctly) that the "expected" number of wins out of 200 would be 200 x 0.99 = 198, since, as they say, "don't expect the expected value...". Yes, in this case, the probability is >1/2 that MC would win at least 198, but the probability is also nearly 1/2 that he'd win fewer. My point is that, although not a bad heuristic or first approximation, this is a bit naive way of thinking about probability and expected values. :) Commented Oct 8 at 1:46

5 Answers 5

17

Tournaments

Not all games affect the rating. FIDE rating is about games in FIDE-sanctioned tournaments pre-registered by the federation in whose territory it is held. In addition, they (FIDE Qualification Commission) have the right to not rate a tournament for arbitrary reasons, and a "tournament" of 200 games between Carlsen and his 1200-rated friend would be a good candidate to simply not be rated even if it would be properly managed and registered otherwise.

Carlsen is free to join various open FIDE tournaments, of course - but those tournament rules will usually ensure that he does not get to arbitrarily choose which players to play against and that he will not be paired with 1200-rated players.

17

Yes, it is possible. And it has been done.

At the highest level such practice would be frowned upon, to say the least, and would only be harmful for the grandmaster gaining valueless rating points with such ridicule practice.

On a lower level, however, a few dozens points can award you a Fide title, and such "farming" by playing in very low level tournaments is seen from time to time. For example, a chess teacher could grab valuable rating points by playing in tournaments against much weaker children on his way to becoming a Fide Master.

13

This was done by Claude Bloodgood, a chess enthusiast who spent most of his life in prison.

In the 90s he regularly played games against other prisoners; since he was a fairly good chess player and they were completely new to the game, his rating rose considerably. By 1996 his USCF rating was 2789, the second highest in the country, despite playing almost entirely other prisoners who barely knew the rules.

2
  • 1
    I changed the firstname of Bloodgood from "Charles" to "Claude", since that the one mentioned on his wikipedia page, and the one he used to sign his book: abebooks.de/servlet/…-ir--kvk-_-30043777656
    – Evargalo
    Commented Mar 23, 2022 at 7:35
  • Unlike FIDE there is no requirement to win against a USCF rated player to get a USCF rating. I don't think it even needs a game against a USCF rated player to get a USCF rating. Commented May 29 at 12:47
10

FIDE ratings are governed by the FIDE Qualification Commission. It is their job to oversee and scrutinise matters relating to titles and ratings. Normally for ratings these things work in an automated, computerised way which doesn't require much oversight. However if something looks wrong regarding either ratings or titles then someone can contact a member of the qualification commission and ask for the matter to be investigated.

In any case anything to do with title norms and title rating requirements will receive a certain minimum level of scrutiny by one or more human beings before being approved and will not be automatically validated by the computer saying "Yes".

In terms of running a very long one-sided match which one player is guaranteed to win purely for the sake of gaining rating points there is an additional hurdle to be overcome in the FIDE Rating Regulations:

6.2 Where a match is over a specific number of games, those played after one player has won shall not be rated. This requirement may be waived by prior request.

Your 200 game match is only going to gain you rating points for the first 101 games. If you apply in advance, as required by the above rule, to have all 200 games rated then the Commission is going to smell a rat and likely deny the event any rating. Furthermore you are likely to be marked by the Commission as a likely cheat and your future moves in the realm of their jurisdiction are likely to receive additional scrutiny.

I don't think this rule was introduced to prevent "rating farming". I think it was introduced because when one player has won enough games to win a match then the they may not try their hardest in the remaining games. Similarly the loser may give up hope and stop trying as we saw in the latest World Championship match even before Carlsen had the match won.

0
8

From 1st January 2022 FIDE have closed this loophole. Their FIDE Rating Regulations effective from 1 January 2022 say this:

8.3.1 For each game played against a rated player, determine the difference in rating between the player and their opponent, D.

A difference in rating of more than 400 points shall be counted for rating purposes as though it were a difference of 400 points. In any tournament, a player may benefit from only one upgrade under this rule, for the game in which the rating difference is greatest.

In other words any subsequent games against players rated more than 400 points less will be rated according to the rating tables. This means, for instance, that if the opponents are more than 735 points lower then zero rating points will be given for beating them.

It's almost as if FIDE read this forum ;-)

1
  • I believe this rule was removed in March 2024 Commented May 29 at 12:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.