I think this Platform is rigged (kind of)

 

No way.

ouch

@b1zmark

Your insights page will have things like: your accuracy diminishes in mid game by this exact percentage
, you lose x% of the time when your opponent plays this opening, here’s EVERY TIME you missed a fork (taken from your actual gameplay), here’s your win / loss ratio based on time of day… it’s a lot of metrics and very detailed. Regardless of whether or not YOU have access to this page (some comments are saying it’s for premium only) obviously chess.com has access to it. So the profile I’m talking about is built, we can move past that.

Matching players can be as simple as - Johnny loses 80% of games when his opponent plays this opening, we’ll pair him with ppl who play that opening in 85% or more games. Something like this or any other host of predictable metrics you gather from monitoring thousands of games from one player. You simply pin specific strengths vs specific weaknesses, extremely simple. If they have the power to create the insights profile, why would you think they wouldn’t monetize this information?

Are they collecting it for fun?

Also noteworthy, chess.com has 77 million users and over 200 employees, it’s a huge operation. I’d be shocked if they weren’t doing something like this. If they aren’t but read this and begin to, dm me for where to send my check. 😘

That seems reasonable, but the win/lose percentage thing isn’t really feasible. Let’s say I lose 60% of my games against 1.Nf3. But on the next move I score well against move X and bad against move Y. Chess.com will have to find an opponent that plays line Y, and this can keep branching out, taking exponential time.

But I understand that’s an example so I don’t care.

What I wanted to say is that I don’t think the people at chess.com are vegetables. If they really did implement something like that, then it would be obvious. Because if it wasn’t obvious, then people can’t associate premium with rating gain. But if it was obvious, then it is equally obvious that they would be biased against free members, which would lose them customers.

At the end of the day, rating gain centers on improvement in chess, not with how bad the matchmaking system is.

Maybe you play better because you paid for it.

I've been on this site for a long time (through other accounts, this one was made recently), and I know 100% that that's not true. I had previously purchased a membership and my rating has grown not because of more favorable pairing, but instead because of lessons and tactics training that chess.com has to offer. I disagree with the original statement, and if chess.com is doing this, it would be violating the fair play policy that defines the integrity of this site. Membership does not lead to better pairings, pairings are made by an algorithm and the players online.

I don't know about matchups being manipulated by account status, but it seems obvious to me that the rated puzzles are structured to increase engagement. When you do poorly at your own level, the algorithm that automates what puzzles you get gives you easy puzzles (for your level, of a type you have already shown yourself to handle well) to bump your puzzle rating up, and it seems (I am sure of the former, less sure of this) to feed puzzles in a rhythm that creates a structure for the user of intermittent reward, i.e. you will hit a fairly high rating relative to your actual ability, early on in using the puzzles, and then algorithm will manipulate the types of puzzles you get so that you chase that rating, but tend to fall short of it.

This is logically what you would do if you were designing the rated puzzle algorithm to increase (manipulate) user engagement, and at the same time create more interest in paying higher subscription rates for the "premium"/unlimited puzzle access. IMO, that whole phenomenon is lame, fundamentally dishonest, and a problem because it is a corruption of the process relative to a truly randomized pool of problems. It is basically a form of scamming the user, and a product of a criminal mentality, at the same time normalizing that kind of business practice (which as most adults know is harming society significantly).

Yes, I know: my player rating makes my comment suspect. My puzzle rating is 2100+. I just rarely play live games against real opponents.

To Administrators: I am specifically not enrolling at the higher (Platinum or whatever the level is with unlimited puzzles) *because* I find the algorithmic manipulation of the user distasteful. No doubt it is effective on children and the stupid, but the manipulation is extremely obvious, and that for an ethically-minded, adult user, is offensive. I absolutely would increase my subscription level to be able to do more puzzles, if the manipulation did not exist. You might distribute an alert to puzzle users if you correct this practice. I would increase my subscription rate, just to be able to do all the puzzles I want, daily, if the manipulation did not exist.

AFAIK, there are three player pools. Premium, normal, and fair play.

Players who bought premium because they could, and not because they cared, tend to make up the vast majority of the premium pool, especially below 1200 chess.com blitz. Now you are, in theory, exclusively facing a very small subset of players who have a mindset of "I can afford it, therefore I should buy it" and either not enough self-control to stop themselves from buying another subscription service if they're wrong or enough money that it really isn't even worth mentioning the price tag. Thusly, they usually don't care to improve (case in point, OP), and so you end up facing the same group of players who are also not improving, and so the ELO to skill ratio ends up slightly skewed compared to the normal pool.

Meanwhile, the normal pool also contains this subset of players who remain content to go 49/51 and stick around the 1000 mark or lower rather than improving, but it's a small minority or a plurality at most rather than the vast majority it is in premium, as most players in the normal pool are not content with their current rating or skill level. This leads to a phenomenon where the average skill level of... say... an 800 goes up faster than the ELO does, as they end up facing self-improvers and stronger players at the same ELO values that a premium player would be fighting exclusively casual players.

The fair play pool is an unknown to me. From what I can tell, most people who complain about cheating have been found later to have been cheating themselves, leading me to believe that Chess.com flags suspicious accounts and pairs them with other suspicious accounts, and the same seems to be true of stallers, quitters, and aborters. Therefore, I have no clue as to the relative strength of the fair play pool compared to the other two, nor do I have any desire to find out.

If you wanted to separate players based on characteristics of their account you would do it only as a bias not as distinct pools, in order to keep the ratings synchronised.

I feel the exact same. What really bothers me are the swings where you win/lose like 10 out 10 or 18 out of 20. This happens so many times, it is rediculous. It should statistically not be possible to happen that often if you are matched randomly with opponents that are somewhat in your ability range. You win 200 points in 30 games, than it is 50/50 for 10-20 games, and then you can't win a game for 30 games, even if your life depended on it. This cycle repeats and repeats and repeats

You biggest winning streak is 9 and biggest losing streak is 5. Streaks naturally happen in play. It happens over the board as well.

Some Aspie level listening skills on this thread (i.e. good at listening for what you want to hear, that pertains to your already existing beliefs, not that great at hearing the main point of what someone else wrote, or processing that through a perspective other than your already existing one).

Here's a useful thought for the day for all: some of you, probably most, are too young to have spent much time in chess parks in cities, where lots of players of all levels from all over the world, many of them senior citizens, gather to play. One thing you notice "IRL" when you go to a park like that: a lot of high level chess players have very low level personalities.

You would think that for those people - having come into their ability levels through lots of playing in such parks - it would have occurred to them to ask themselves why so many of the good older players have such poor personalities (aggressive for no reason, unlikeable in general society, totally unaware of any of that).

That is a very good question, that would help a lot of people to ask, if they did.

The same is true for online chess. The whole of the internet is poisonous. Look at this thread. Chess is extra poisonous. Why?

I am old, for me that is a rhetorical question. I already know why. That's contained in this post, as already written.

It would be very weird and illogical for this site not to "optimize" user engagement by manipulating all its algorithms, and to minimize disengagement that would occur due to users not liking to have negative feelings (of a type that would lead to logging off or cancelling subscriptions) while using the site.

That is basically policy on every search engine and social media site in use today. It would not only be illogical but incompetent by today's exploitative (of you, the user) business norms for the site not to be manipulating your and my brains into greater engagement by manipulating as aspects of the user experience that influence engagement as possible.

That's another thing that any intelligent person should be able to reason out on their own, though, bafflingly, most won't.

The question is, is that the kind of world you want to live in, one where your every apparent choice and use of attention and time is manipulated by an algorithm designed to manipulate you without your knowledge or wish for it to be so?

The reason I'm commenting is that, older than many, I don't think that understanding that in full anybody wants to live in a world like that, because within it every single person is a slave, including the people designing that world.

So what can you do about it, to prevent it? Don't live in denial, see the obvious for what it is, and make it known that you don't want to live in a world like that, to the degree that objection is a norm, instead of complicity.

It would be stupid for the site to do so. There's a recent interview with @erik where it's mentioned that players winning their first game are more likely to stick around, statistically speaking, but that it would be morally problematic (I think he said "icky") to do it.

That said, Chess, in and of itself, already has the psychological mechanisms in place for many people to play just one more game, to attempt to get their highest rating or get their rating back from a slump. A lot of games have that without needing any algorithm or trick to try to increase that engagement.

This is not unique to Chess, of course, but the game has the same effect on the brain reward system that other games (or other activities) have, endemic to the act of actually playing.

Of course ratings are manipulated -clearly- ignore the ratings and keep playing, just like you would if you played a stranger in the park. I can speak to the the lower rating levels: if they want you to increase your rating - you will. If they want you to decrease - you will. How is that done? the "500" rated player you're playing could be truly a higher or lower rated player - also manipulated or just a bot represented however they choose. The algorithm is simple, if you want to push a player down match them with an underrated better player. They have all the data from every game you ever played, and you don't think they use it to their advantage? - don't be naive. They are only trying to get you to buy their product or if you paid your money reward your ego for dollars "well spent". My Ford is better than your Chevy...

No it's not...where's the Kool-aid? We the Sheeple.

Why did you bump this

What?

So people know there is potential rigging going on

someone test this theory by gifting me premium membership :]

Except, the site doesn't do any of that.🤔