Playing strength in tournaments vs the playing pool

@Wasted_Youth said in #9:
> 1. It ́s impossible to inflate the rating pool;; for everyone who gains points, others lose them. Even if 100 GMs were to register tomorrow: their ratings would rocket from the starting points, but thousands of others would take a dive.
>
I don't think you get my point. lets imagine 3 players, A ( 2200 ) B ( 1900 ) and C ( 1905 ) , the ratings being the actual strength of the players.
A plays in hourly arenas
B plays in hourly arenas ,<2000 arenas/ the pool
C plays in <2000 arenas/ the pool
<2000 arenas are basically the same as pools due to lack of berserk.
now say A gets paired with B in a hourly tournament. A, being 300 points higher rated, berserks. now due to A being at a disadvantage, has a higher probability to lose to B, lets say this game A lost due to berserking. B gains 10 points and A loses 10 points. now B plays against C in a Berserk-less game, since B is over-ratted by a bit, C is actually a bit stronger even if the ratings say otherwise. C wins and gains 5 points, B loses 5 points. now A has a rating of 2190, B is 1905, and C is 1910. over a long period of time, A might eventually fall to 2100, B might become 1950, C might become 1955. as you can see, B and C are over-rated by 50 points, while A is underrated by 100 points.
your point stands within one pool, but arenas and the standard pool are different pools, due to the berserk option.

> 2. The number of players who are at the moment berserking in tournaments is negligible compared to the number of players online at any given moment. Let ́s say that in a tournament with 500 players 50 of them are berserking at any one moment; there could easily be 50,000 others online. Even allowing for 10 simultaneous tournaments of that size every hour: the percentage of berserkers is 1%. Not enough to have any effect on the rating pool, even if such a thing were possible.
>
> I ́m still convinced it ́s a psychological thing.
>
even a 0.05% inflation of the general pool and a 0.05% deflation in the tournament pool per day leads to a 0.1% difference, which over the course of just a year leads to a 44% difference, which is huge

> PS I just saw that I didn ́t reply to your idea that regular tournament players are underrated by 150-200 points when compared to the pool. I ́d go along with that, with emphasis on the "regular", but again I think it ́s a mindset thing. It certainly applies to me; usually 1550-1600 rated, but with an average performance in tournaments of 1750. It ́s because I 'm a bit of a happy-go-lucky player who ́s not really bothered about his rating; I often play in the pool when I ́m tired or distracted, or after a couple of beers late at night; that keeps my rating down. Tournaments however I only play when I ́m feeling fit and after having warmed up with a few games first; they ́re my main event. I don ́t think that this applies to the majority of tournament players, as I probably have about a 50% win rate in tournaments against similarly rated players to myself; I do lose occasionally to the odd 1300 player, just as I ́ll occasionally beat an 1800+ opponent.
could be the reason, as I imagine many players are more serious in tournaments than in the pools.

Always take a peak at the profile of the top 3, very often this will answer your question.

@Akarsh_2010 - this works if you reduce it to these three players. However, player A will after loss of rating points through unsuccessful berserking be more likely to win in the pool, whereas B and C will be more likely to lose. That ́s what I meant when I wrote that it ́s impossible to inflate the pool; it ́s a self-regulating thing.

@Akarsh_2010 said in #11:
> I don't think you get my point. lets imagine 3 players, A ( 2200 ) B ( 1900 ) and C ( 1905 ) , the ratings being the actual strength of the players.
> A plays in hourly arenas
> B plays in hourly arenas ,<2000 arenas/ the pool
> C plays in <2000 arenas/ the pool
> <2000 arenas are basically the same as pools due to lack of berserk.
> now say A gets paired with B in a hourly tournament. A, being 300 points higher rated, berserks. now due to A being at a disadvantage, has a higher probability to lose to B, lets say this game A lost due to berserking. B gains 10 points and A loses 10 points. now B plays against C in a Berserk-less game, since B is over-ratted by a bit, C is actually a bit stronger even if the ratings say otherwise. C wins and gains 5 points, B loses 5 points. now A has a rating of 2190, B is 1905, and C is 1910. over a long period of time, A might eventually fall to 2100, B might become 1950, C might become 1955. as you can see, B and C are over-rated by 50 points, while A is underrated by 100 points.

You are forgetting that rating is a 'ladder': yes, the 2100 player may lose rating by berserking against lower rated players, but equally, they will face 2400 rated players who berserk against them. Similarly, players B and C will themselves berserk against 1500 rated players they encounter. These 1500s berserk against 1100s, and so on.

@Wasted_Youth said in #13:
> @Akarsh_2010 - this works if you reduce it to these three players. However, player A will after loss of rating points through unsuccessful berserking be more likely to win in the pool,
Hence the regular tournament player part ( more so regular berserker ), if someone plays tournaments only sometimes this holds, but if someone is plays mostly in tournaments and berserks a lot, thier rating would be lower than their strength

>whereas B and C will be more likely to lose. That ́s what I meant when I wrote that it ́s impossible to inflate the pool; it ́s a self-regulating thing.
B and C losing in the pool does nothing but spread out the extra rating withn the pool, which overall slightly inflates the pool. I took just 3 Players for nothing but simplicity, same would apply for 30, 3000, or even 3 billion.

@TheSquare said in #14:
> You are forgetting that rating is a 'ladder': yes, the 2100 player may lose rating by berserking against lower rated players, but equally, they will face 2400 rated players who berserk against them. Similarly, players B and C will themselves berserk against 1500 rated players they encounter. These 1500s berserk against 1100s, and so on.

This creates nothing but a chain. Yes the 2100 may regain the lost points, but 2400 lost those points and the 1900s gained those points still. Similarly if 1900 berserk against 1500s, they may lose the extra rating, but now the 1500s are overrated, and due to the existence of <xxxx arenas, which thesee 1900/1500 players might play in often and due to lack of berserk, they will lose the points,inflating the pool.

Playing tourney on a regular basis will eventually lower your rating as berserking, violation of fair play regulations, but also the impossiblity to keep up concentration during multiple games in a row as well as many other factors are a thing, whereas by playing the pool it is more likely to get opponents that match their rating because they play lesser games and are more focused for each of them. Also you may be paired with less accounts that are way higher or way lower rated than yourself, like in an arena e.g., which eventually causes less fluctuation in ratings.