Thanks!
So essentially more people in the pool means rating is harder to achieve? Since generally majority of the people new to the pool are going to be lower ranked I wonder if it only effects lower ELO players? or ratings across the board are harder to reach than they were a couple of years ago.
I would have thought that the increase in players due to the lockdown would tend to inflate the ratings somewhat, so that a 2020 rating of 1000 wouldn't be worth as much as a 2018 rating of 1000.
My reasoning is that with a wide choice of initial ratings... you can select to start your account at 800, 1200, 1600 or 2000 as I recall... most players would enter at 1200 or so, but the average rating AFTER playing a hundred games or so is around 1000 to 1100. So excess rating points are being pumped into the system by the swarm of new players, and this would tend to inflate (and devalue) existing ratings.
I would have thought that the increase in players due to the lockdown would tend to inflate the ratings somewhat, so that a 2020 rating of 1000 wouldn't be worth as much as a 2018 rating of 1000.
My reasoning is that with a wide choice of initial ratings... you can select to start your account at 800, 1200, 1600 or 2000 as I recall... most players would enter at 1200 or so, but the average rating AFTER playing a hundred games or so is around 1000 to 1100. So excess rating points are being pumped into the system by the swarm of new players, and this would tend to inflate (and devalue) existing ratings.
But you also get lots of cheaters. They steal rating points out of the pool, then get their account banned, thus lowering everyone's ratings.
Overall I'd say that most ratings today are pretty similar to how they were 10 years ago.
Except bullet on this site. Bullet on Chess.com is significantly less inflated compared to how it was 10 years ago.
I do believe average chess strength is increasing. There r so many different ways to study and play now that weren't available decades ago. I know ur only comparing 2 years ago but I'm thinking over the long term.
I think the typical 1000 of today would score well above 50% against a 1000 of say 1990 or before.
But you also get lots of cheaters. They steal rating points out of the pool, then get their account banned, thus lowering everyone's ratings.
That is not how ratings work or banning works.... o_O
When someone is banned, it is like those games never happened. Your score is rebalanced.
I would have thought that the increase in players due to the lockdown would tend to inflate the ratings somewhat, so that a 2020 rating of 1000 wouldn't be worth as much as a 2018 rating of 1000.
My reasoning is that with a wide choice of initial ratings... you can select to start your account at 800, 1200, 1600 or 2000 as I recall... most players would enter at 1200 or so, but the average rating AFTER playing a hundred games or so is around 1000 to 1100. So excess rating points are being pumped into the system by the swarm of new players, and this would tend to inflate (and devalue) existing ratings.
...Could that explain why rating bumped up to 1500 from 1150 one month after I got back into chess, or is it a combination of getting better and the inflated rating.
Well on the one hand there is more training possibilities, even if it's just the free lessons available on the site. It is very difficult to assess what will happen with this though as there are also a lot more distractions that there used to be.
On the other hand, ratings inflation is very appealing for chess sites, at least in the short/medium term. Let's say if you were 1800 five years ago and you're 1800 today - that's not going to cause you to be too excited to come back every day is it? However if you were 1900 today due to ratings inflation, it looks like you have improved a lot and maybe if you try hard for the next few years you might get to 2000 or 2100 when the reality is you didn't improve at all.
Now if you think "but then the ratings lose significance and you eventually have yahoo ratings" that's a good point and for sure they don't want to overdo it. However look at FIDE official ratings - they're constantly inflated over the years. Why do you think this is? It's because they always want people to think the current world champion and players are the best the world has ever seen. It keeps people interested in the game more.
It's taken half a century to inflate about 200 rating points with FIDE, but it's worth it for them as it makes it seem like chess players are getting better, while Kasparov has objectively a lot more claim to the title of best player ever than Carlsen. It also keeps the maximum ever rating record getting broken. chess.com likely have a similar thing going - how boring would it be if decades from now Hikaru still had the highest blitz rating of all time that he got around 2020 and noone could touch him? They will never let that happen, they will keep inflating the ratings pool. They will have a new guy that's marketable for chess there.
If you compare 2018 to 2016, probably it wouldn't vary much.
But 2020 is, uhm... "special". With more people playing chess the distribution is also different. I've found many people (myself included) who wouldn't play as much in normal conditions but with the goddamn virus there are many more people playing while on quarantine, and therefore it is harder to reach the 1000's. The distribution scale changed quite a lot, and that makes it a little more difficult to be there.
But if you want to seek for compensation, there is easier access to content in the form of video streams now than four or five years ago, therefore if you want to know a little more about a certain line (let's say the King's Indian, just to mention one that I have found once and lost), you can just turn youtube on and watch Hanging Pawns playing many variations of the King's Indian, or Varuzhan Akobian, or ChessTV... you name it.
I don't think the quarantine has nothing to do with it. Unless the Coronavirus has now infected the rating system. I never thought a rating system could catch a virus.
Not your question, but I think the rating deflation may happen in tournament chess more so than online (at least in regards to like 1000-2000 USCF) after this pandemic. In online, my guess is there are enough lower-rated players who are just trying it out to see how they like it that the distribution shouldn't change much (just my guess). Then again, like any activity, play in chess probably doesn't get worse over time (i.e. top players today>top players 20 years ago, average players today>average players 20 years ago, etc.)
It doesn't decrease.
Until you reach my age.
... what were we talking about, again?
I think the average chess player of today is better than the average player 20 yrs ago. Online playing has helped a lot. Pre internet it could be difficult to find a similar rated player for a game, now its easy 24/7
The only reason Carlsen is the best is because he has access to an engine. If Bobby had access to extremely powerful engines, he would be the best. Why? Because Bobby was self taught and he annihilated every of his competitors. His peak rating is also quite close to Carlsen's which is impressive because he did not have access to engines and computers and has more limited reading resources as there was no such thing are Amazon back then.
But you also get lots of cheaters. They steal rating points out of the pool, then get their account banned, thus lowering everyone's ratings.
That is not how ratings work or banning works.... o_O
When someone is banned, it is like those games never happened. Your score is rebalanced.
They'll only reapply rating points if it was your most recent game. Otherwise, you're out of luck.
For example, let's say I make a fresh account. My rating starts off as 1800. My first game, I play vs a cheater and lose 400 rating points. Now I'm down to 1400 rating.
Then I play 20 more games, winning most of them, and bringing my rating up to 2000. And then they detect that the player was cheating and they ban him...
Do you really think they're going to give me my 400 rating points back now, thus bringing me up to 2400?
I stand by my point that cheaters steal rating points out of the pool, thus deflating ratings.
However, I'll add that new users signing up to the site, losing a few games and then quitting, adds rating points to the pool, thus inflating ratings.
So it's a question of which effect is more powerful.
Just a question regarding someone that is 1000 ELO today, would their actual chess ability be the same as someone that is 1000 ELO say two years ago? I'm talking about chess.com rating.