Chess.com ratings realistic or inflated?

I've been reading through bunches and bunches of topics on the forums on this site about its ratings and how they compare to FIDE. Tons of different opinions are offered, so I'm still confused... but one thing people seem to agree on is that Chess.com's ratings are higher (inflated) than other chess organizations like FIDE. Some say they are inflated by 100 points, some say 200, some say 300... really all over the place. Question: is this alleged inflation really problematic, or does nobody really care about it? In other words, do you think it could lull someone into a false sense of security, or a sense of being better than someone actually is?

Inflated yes, people join lose points and never play again. But you can't compare them because they are different ways to play the game.

Hi Bosco

On the internet we have only a virtual existence. Therefore our rating is virtual too. We like to see our virtual rating because it gives us some idea of our improvement and a comparison with others. This is good and positive! It is purely for personal motivation and satisfaction to know our virtual rating. The bad news is that it means very little in the real world. Any virtual record we set on the internet has as much weight as bubbles in a bathtub. This is because whether we are honest, or dishonest the fact remains the same that our results cannot be verified in any way by simple commonsense. As to the possible inflation of virtual ratings, it is interesting to understand why it would happen exactly (if it is true). It just proves that virtual ratings are meaningless except for personal knowledge in a relative sense. Online cheating is not a big factor so long as the organisation such as Chess.com never introduce monetary or similar rewards for tournaments.

So enjoy online Chess, but never substitute it for the real thing. Your local chess club and organisations are always to be supported first and foremost because that is where ratings can actually be verified. If you do not want to go to the local club and just enjoy online play, make sure with all your energy that no online organisation ever decides to introduce material rewards (money, physical trophies etc) for online play. If they do, you can be sure that cheating will become the statistical certainty that will spoil the experience for everyone.

Cheers

Yes, I've read about that problem, people signing up, losing some games, saying, "This site stinks" or "Oh, I guess I stink," and then chucking their account. But should that really cause much inflation? I would think that would only inflate by a few points here or there, nothing significant, especially at the higher rating levels. Where are people coming up with 200-300+ points of inflation?

They are an accurate way of measuring inside the chess.com pool. They are not accurate in measuring between different pools, as ratings usually aren't.

glider1001: I hear ya on playing the real thing! I like your logic too, as far as no material reward factoring into these games. But is a rating literally worthless here? That seems depressing. Assuming they are inflated, I was thinking that if I work hard to earn a 1800 rating, THAT might be discouraging because it would only be worth a 1500-actual rating. But for that same 1800 to be worth no more than my bath bubbles that have popped into thin air or washed down the drain... that's even more discouraging. I'm not sure I can believe that. An 1800 rating would HAVE to be ballpark-accurate.

But anyway, thanks for the insights!

Right, I've gathered that as well, and my goal isn't to derive some equation to convert from Chess.com to FIDE and vice versa. But as I said above, if I work and attain 1800, is that worth no more than a 1500-FIDE rating?

But as we've already established, the two can't be compared that way. Question is, how real or significant is this inflation, and could it cause me to be deluded into thinking that I'm much better than I ACTUALLY am? 

My chess.com rating you can see, my national rating is 1450, if that helps you anything.

Bosco,

how real you rating is depends on many factors. There are some folks here whose rating matches their USCF one. There are others (like me) where the rating is waaaaaaaaay off. I am currently rated around 2250 here and the highest rating I ever accomplished in Germany was 1800 (my USCF is even lower but also 11 years in the past). Why the difference? Numerous time-outs by opponents (at least some in equal or worse positions), playing against friends, playing folks that play 1200 games at the same time (like Kasparov who I could never dream of beating in a regular one-on-one). Playing thematic tournaments where the choice of opening assures you an advantage with one caller and most importantly taking the time to ponder the move. There are many excellent OTB players here that simply consider chess here a casual thing and are not worried about dropping a game.

Well I'm finally joining the USCF and going to play in a few tournaments in New York. I need 20ish games to get a USCF rating. I'll let you know how far below my meagre chess.com rating (1500) it turns out to be!

Thank you for these viewpoints. These discrepancies BOTHER me, though. Tricklev's rating is inflated by almost 500 points! Yours, Dietmar, by almost as much? One of the things this tells me, honestly, is that I, as someone who cares deeply about rating and improvement, should not waste my time playing on this or any other chess site, because the rating would be essentially a figment of my imagination. I could say, "I'm rated 2100 on Chess.com," and it would be as if I pulled that number out of the air, even if I fought and scraped for it. As you said, others have accurate ratings, but it seems like it's just coincidence. So what then: since I consider an accurate rating important, does that mean Chess.com is simply not for me?

Thanks! Let me know. I hope you buck the trend and garner a HIGHER rating in USCF play. 

Maybe get a rating first, then worry whether it is inflated.

I think the more important question for the player to figure out is what the rating means to them. Even if there were a scientifically perfect way to compare ratings and have a mathmatically precise conversion ratio, we still need to know what to do with that data. If I am worried about rating, why am I worried about rating? If you like chess, play chess! Spend the same amount of time studying as you would have spent thinking about rating conversions and you will get better!

I guess it could throw you off in your first tournament -- but that's about it.  You ask if it's problematic, it's not.  I mean, what's the problem? 

Like Tricklev posted, look at people's national/FIDE rating, then compare to cc and get your own idea.  If you really want to know how strong you are before going to a tournament, join a club and after playing ask them to estimate you and/or ask them what their rating is.

lol that is some terrible logic. Rating is useful because you can match with opponents of approximately your strength. By doing that you should improve. That's how simple it is. Your dramatic conclusions are so out of the blue.

Your rating changes after every game you complete. How is it supposed to be realistic and stable?

Yes, ratings are for pairing purposes.  Games are for improvement.  No one plays online for bragging rights except the weirdos who are above 2600 or so on this site.  And odder still, we all know they cheat... so it's not bragging rights after all, it's probably a combination of neurosis and boredom.

Good answer and suggestion to their questions. I would agree, you can't determine how one rating system compares to another. It is all context of we're you're playing and who's system you're following. If you want to know how good your playing is over the board you have to play people over the board with ratings for it.

 Your actual rating in your head stays the same; the number changes here because it becomes a more-accurate guess as you play more games.