Concerns About Smurfing and Rating Manipulation on Chess.com

Sort:
DrSpudnik

What still has not been addressed is what people gain by intentionally sinking their rating? It's not like there's money at stake here.

angsthe_arc

🫠

GMegasDoux

@25. More to the point, people can seek games with players rated much lower than them without changing their rating. Same with seeking higher rating.

DavidWills99

All I can think is that some folks are more prone to conspiracy thinking. It might be human nature to seek explanation (and justification); an ancient survival technique. Most times, it's simply Occam's Razor - I need to become better at winning and stop looking for boogie men.

OhPawnSaint
ChessMasteryOfficial wrote:

If a player is intentionally lowering their rating to farm wins against lower-rated opponents, you should report them. Chess.com has systems in place to detect and punish sandbagging.

I don't think sandbagging detection or enforcing reports on sandbaggers is great here. Look at the tournament pools and who is winning. I would show one account who is doing that shenanigans since 2023 and still hanging out here but it is not allowed for obvious reasons.

hkbusowkgpa4
Rodrigo-Moraes wrote:

Hey everyone! I’ve been noticing a trend on Chess.com where players who seem highly experienced end up with artificially low ratings, either by creating new accounts or intentionally losing games to drop their rating. It’s frustrating to face someone who plays like a 1500+ but shows a rating of 300–500. It can feel unfair and discouraging to newer players (like me).
Someone says, go and get better, but it's hard to accept me being a ~350 which is literally below day one, play a game with a guy that 2,000 live games on pocket, and the worst part, since now I'm fighting them for 30% of my games I will eventually become these guys too, since I'm losing every game against these smurfs and my ELO keep dropping because of them.
Now I have 385 games and this keeping happening.

Is this reportable?

  • Generally, if you believe someone is clearly manipulating their rating (for example, sandbagging by intentionally losing games), you can report them. However, just because someone starts a new account and has strong skills doesn’t necessarily mean they’re breaking any rules, it might just be a new account for a returning player.

  • If you notice a sudden or suspicious drop from established 1500 down to 300 (or something equally drastic), it’s worth sending a report? 

Now I'm really don't know what to do, buy a course, get better and then after some study go to 800-1000 and this will not happens anymore, but what about new users? They will quit because there is a 25 years old guy in a 5th Grade Elementary School football team, and the Principal don't take any actions?

Even if you did not lose to him badly, you will still have many badly defeats against low rated opponents because your rating is too low. I think that it is much better for you to learn how to improve chess skills.

tom30356

Smurfing is bad?

boriskravitz

My God, please, be a man.

tom30356

boriskravitz wrote: My god, please, be a man. ME: Actually I seen many higher rated players in YT that they made accounts then beat all players, smurfing is bad? They do the bad thing?

Rodrigo-Moraes
tom30356 escreveu:

boriskravitz wrote: My god, please, be a man. ME: Actually I seen many higher rated players in YT that they made accounts then beat all players, smurfing is bad? They do the bad thing?

It is against the rules, for GMs that do SpeedRuns, they need to have a specific account approved by chess.com, and after end the SpeedRun all the opponents receive their ELO points back.

lmh50

Yes, and such people as Anna Cramling have done this, and explicitly explained that all rating points would be restored.

But let's look at the likelihoods. There are about 20,000,000 games played here on a typical day. Let's be really outrageous and estimate that 1000 GMs played 100 games each today in rating-rushes with a view to creating YouTube videos, that means 100,000 contaminated games, or 1 in 200 games was a grand master messing things up (this is obviously a level that only a conspiracy theorist could believe, but let's work with it).

That means that if you play innocently, you will play 199 real games for every game that you lose catastrophically because it's a GM. Looking at what happens to your ratings, ignoring that you are automatically paired with people of similar rating, and assuming that you could drift up and down without affecting your likelihood of winning (i.e. let's assume the results are random because you're always playing someone of identical standard), then your total expected drift in rating is about the square root of the number of games you played, multiplied by the step-size (which is 8 in rating points here). The square root of 199 is about 14, so you'd expect to drift about 113 points in normal games, compared to the 8 points you lost because of the GM. In effect, 93% of your expected change in rating is random, and only 7% is down to the GM.

But actually your overall drift is less than that, because the system has a built-in dampening effect. If you lose some points to the GM, you will get paired against people of a lower standard, so you ought to gain the points again.

Even if the site is full of stressed grand masters desperately trying to bolster their YouTube career, they're doing nothing more than extracting some rating points from the overall system. If they start at 400 and keep playing until they've reached 3000, they have extracted 2600 points from the pool, and given that Chess.com has over 100 million users, who between them must have more than 40,000,000,000 points (I just assumed everyone's a 400 player...), each new GM account is extracting at most 0.0000065% of each other user's overall rating before it maxes out and the GM has to open themselves another account.