Bloated FIDE Ratings of the Elite

Kramnik alternates between being a Super GM and someone who doesn't like the game of chess. Topalov is streaky like Ivanchuk. Carlsen made Vallejo Pons look like a 2200 and then played like Tony Miles vs Adams...

It is tough to tell, I tend to watch Ivanchuk, Shirov, and Nakamura more as they tend to appear in both elite and non elite tournaments. I think how they perform shows roughly what an accurate rating is.

Interesting.

Reb and Pogo agree and they shoulod know.

So give us a few names of players who "should be" around 2800 as well if they hadn't been excluded from the elite clique. Honestly who plays at that level except for the current guys?

 You completely misunderstand my point. I am not claiming there should be more 2800 players but less, even none !  If the current elite players were to play in 2 or 3 big Open swiss events each year their ratings wouldnt be so bloated as they would lose to sub 2700 GMs and sometimes even to sub 2600 GMs...... this is why they dont play in them. If elite players only play against other elite players its easy to maintain a bloated rating......

The rating formulas used by FIDE and USCF are usually more indicative of relative strength rather than daily performance, anyways.  For OTB situations, it wouldn't make much sense to have huge swings in individual ratings based on 1 or 2 tournaments, only after say, 50-100 games have been played.

Given a long streak of bad performance, players ratings ought to come down on their own.

Match play does have additional factors. Draw-by-agreement is common once a match has been decided, to save energy for future rounds. The team comes first, not the board prizes.

 Hmm.. I wasn't under the impression that a 2800 couldn't lose to a sub 2700- in fact, statistically they should only score 75% vs someone who is 2600.

As far as how the 2700ish GM's do like Adams or Pons - we can look at their recent results in less prestigious events. Both Scored at about their expectation vs a mixed field of opponents.

So it seems to me in relation to each other they are rated right. Now as to the question of 2800 being an inflated number compared to previous era's I will leave that up to Jeff Sonas and co.

It makes sense though, when you artificially alter the pool of players the ratings wont be accurate.

It's hard to not think of the top 10 as monstrously strong though.  Do you also think Kasparov was below the relative 2800 level in the early 2000s?

hmmm the politics of envy

premise dismissed

>:)

I don't understand. Nobody is as good as the three current 2800+ guys, yet you say they aren't worthy of having the top three ratings in the world?

er.. but why would they?

Alexander Khalifman also thought that the top guys ratings were inflated, when he became "world champion" as well as Rustam. Both held a bare 2700 rating on beating the worlds elite.

Winning a knockout tournament is nothing like playing a match.

well yes it seems to be a common theme for those who dont quite make it to the very top, ranked around 50 at his peak, the biggest prizes just out of his grasp, must have been frustrating

>:) 

I was under the impression that you played matches in the knockout cycle.

The percentages don't apply at the high end of the rating distribution; the population is too small to interpolate/extrapolate reliably.

I supposed, as in first to 8 in a 15 game match would win the match and "knockout" the other guy... but you see how this is different than a knockout tournament right?

It may be Carlsen has a legitimate 2800 playing strength, the point is that once you reach the top, your rating (which is a relative measurement) is skewed in the first place (the same with people at the lowest end).  Add to this changing the pool of players afterwards (the top 10-15 playing in elite round robin events with each other) and again ratings will tend to become skewed vs playing in a larger pool.

Of course it's a bit speculative to say no one deserves a 2800+ rating currently, but the point is, whatever the rating, they're inflated at the top, a claim which makes a lot of sense to me anyway.

It probably doesn't make much sense to worry about who's in the penthouse, either, when you occupy the apartment right next to it.  At some point, a number will become just a number (compared to the rest of us).