Interesting article, it does seem like you'd have to have newer juniors coming in faster and faster to get the effect seen, however. And the one graph shown doesn't look like the recent covid bump in interest would explain it.
Playing where I do I often find that stronger players will drop out of the chess scene. The wins are worth little in terms of elo and losses hurt a lot. My area it's fair share of 1400-1900 players but if a player reaches 2000 most will quit playing rated games.
They just find the stress of having to win every game too much. When you're the top dog everyone is gunning for you. They'll research all your games, your openings and tendencies and it's hard to stay on top. This also has downstream effects of "under-rated" players because when players play the same people over and over again they're unlikely to gain or lose much rating.
Older players that return to tournament chess often have one tournament that goes badly and they don't return. The accessibility of chess resources means that blunders that used to happen all over are reduced, and since online is easier to get a game than over the board, the people that show up to a otb tournament are out to test their mettle. I don't think there are many "casual" tournament players anymore.
Thanks -- good article. My understanding is that the USCF is look at it. Not sure what happens next. But I do have a question, mathematically related. You wrote:
"It’s also tempting to read these results to mean that you, yourself, are underrated. Unfortunately, unless you are a rapidly improving junior, I don’t think this is all that likely to be true....With that in mind, my biggest suggestion is this: if you’re underrated, prove it!"
But how can we prove it if we keep getting paired up with these underrated juniors? I play at least one every tournament (many times, I play more than one). Earlier this month, I got beat by a fairly newbie 11-year old who played way above his rating (imho). We've all got stories like that. I play a section up -- I can beat an adult who's 200-300 points higher than me 30% to 40% of the time lately -- so much rating should be going up, but... the kids who are approximately my level to only 100 points higher than me I find to be much tougher (and often lose).
Do you have any idea if something like the Glicko system used by Lichess (and probably other online platforms) would solve this issue? I get the impression the volatility mechanic would solve this "inertia" of junior players, as long as some scheme for assigning deviation to already rated players would be devised.
However none of those ratingsystems solves the issue of players suddenly playing much better than their rating which happens nowadays. For that you need a totally different approach.
From 2014-2018 and 2020-2021 I haven't been playing and yes, the pool is much stronger. In 2013 I peaked at 1785 and winning lower rated players I found easy. Now it has change, playing a +1700 is a tough struggle and +1600 is not easy at all. In the way I also lost about 200 rating points playing some 60 games in 2019, 2022 and 7 games in 2023. I think I am underrated but the word doesn't obsess me since now I have the chance to inflict big upsets from time to time, and is funny to see opponents in the verge of tears punished by a player 200 pts. lower.
Anyway I understand people annoyed because 10 little Indians are stealing their points. My suggestion to FIDE is to narrow the 400 gap to 250 or even 200. That will allow the stronger part to win more than 1 point no matter how low the rating of the opponent was. Actually the rule of 400 means that if you play somebody with a more of 400 gap, automatically it reverts to 400. But winning a guy with -400 rating also gives you a miserable prize of 0.5 or similar points. A win is a win, no matter how low rated is the opponent, and deserves more than 1 point. 1.5 would be fine.
Interesting article, it does seem like you'd have to have newer juniors coming in faster and faster to get the effect seen, however. And the one graph shown doesn't look like the recent covid bump in interest would explain it.
Playing where I do I often find that stronger players will drop out of the chess scene. The wins are worth little in terms of elo and losses hurt a lot. My area it's fair share of 1400-1900 players but if a player reaches 2000 most will quit playing rated games.
They just find the stress of having to win every game too much. When you're the top dog everyone is gunning for you. They'll research all your games, your openings and tendencies and it's hard to stay on top. This also has downstream effects of "under-rated" players because when players play the same people over and over again they're unlikely to gain or lose much rating.
Older players that return to tournament chess often have one tournament that goes badly and they don't return. The accessibility of chess resources means that blunders that used to happen all over are reduced, and since online is easier to get a game than over the board, the people that show up to a otb tournament are out to test their mettle. I don't think there are many "casual" tournament players anymore.
At least that's my read of things.
I didn’t drop out, I’m just taking a break ok?? 😂😂😂
Thanks -- good article. My understanding is that the USCF is look at it. Not sure what happens next. But I do have a question, mathematically related. You wrote:
"It’s also tempting to read these results to mean that you, yourself, are underrated. Unfortunately, unless you are a rapidly improving junior, I don’t think this is all that likely to be true....With that in mind, my biggest suggestion is this: if you’re underrated, prove it!"
But how can we prove it if we keep getting paired up with these underrated juniors? I play at least one every tournament (many times, I play more than one). Earlier this month, I got beat by a fairly newbie 11-year old who played way above his rating (imho). We've all got stories like that. I play a section up -- I can beat an adult who's 200-300 points higher than me 30% to 40% of the time lately -- so much rating should be going up, but... the kids who are approximately my level to only 100 points higher than me I find to be much tougher (and often lose).
Do you have any idea if something like the Glicko system used by Lichess (and probably other online platforms) would solve this issue? I get the impression the volatility mechanic would solve this "inertia" of junior players, as long as some scheme for assigning deviation to already rated players would be devised.
Lichess uses Glicko 2 while Chess.com uses Glicko 1. I guess there are probably even more Glicko systems. In fact 12 years ago there was a competition about to build the most accurate ratingsystem (https://en.chessbase.com/post/the-deloitte-fide-che-rating-challenge).
However none of those ratingsystems solves the issue of players suddenly playing much better than their rating which happens nowadays. For that you need a totally different approach.
From 2014-2018 and 2020-2021 I haven't been playing and yes, the pool is much stronger. In 2013 I peaked at 1785 and winning lower rated players I found easy. Now it has change, playing a +1700 is a tough struggle and +1600 is not easy at all. In the way I also lost about 200 rating points playing some 60 games in 2019, 2022 and 7 games in 2023. I think I am underrated but the word doesn't obsess me since now I have the chance to inflict big upsets from time to time, and is funny to see opponents in the verge of tears punished by a player 200 pts. lower.
Anyway I understand people annoyed because 10 little Indians are stealing their points. My suggestion to FIDE is to narrow the 400 gap to 250 or even 200. That will allow the stronger part to win more than 1 point no matter how low the rating of the opponent was. Actually the rule of 400 means that if you play somebody with a more of 400 gap, automatically it reverts to 400. But winning a guy with -400 rating also gives you a miserable prize of 0.5 or similar points. A win is a win, no matter how low rated is the opponent, and deserves more than 1 point. 1.5 would be fine.