x100 peaks on the ratings distribution graph continued...

@Jimothy-Jangles said in #10:
> why does the game get more difficult to win when I'm in the 50 to 75 zone, isn't that what the graph shows?
No, it does not. Just like it's more likely players will rest after getting over a milestone, it's more likely they will have more incentive to play when they get close under it so that they are less likely to stay there. I don't say everyone thinks and acts like that but those anomalies are not too big either.

I'm giving up, I really prefer a simple, logical and consistent explanation over artificial hypotheses which can be easily proven wrong by reading about the algorithms and inspecting the code. But perhaps one more hint: have you noticed that these peaks are best seen in bullet and blitz ratings (on the right side), much less so in rapid (where they essentially stop around 2000) and are almost misssing from the right side of the classical graph (but are, in more complicated form, present on the left side instead)? Assuming the natural psychological explanation, it makes pretty good sense. But I really don't see how I would explain it if I assumed some fictitious irregularities in rating calculation process.

@mkubecek said in #7:
> Some people insist that using these graphs with straight line segments connecting points representing discrete values (or summed over intervals) are Bad ThingTM and that bar graphs or at least isolated dots should be always used instead. I always found it a bit "Sheldony" as I believed that noone could really think that those segmented lines represent actual dependency of continuous values. I may have been wrong... :-(

yes, but I would expect to see a more rounded approximation. That's my whole point. 25s, 10s, 5s, 1s, whatever - it's still way too much a peak (and the curious dip preceding it) for me to be happy with the resting point idea. I do agree though that the 100 mark is a point where we sometimes stop playing for longer (probably bc we're proud of the rating and don't want it to slip) - but the proportion of people doing this compared to people not doing this at other points would have to be particularly large for this idea to add up. That's what I can't comprehend. I get the idea though. but surely something else is contributing to the 100s peak reason aswell?? My main question anyway is regarding the increase in the slope decline from the 50 to 75 mark. Can anyone help me here. Surely whatever the reason is for this is related to the reason for the 100s peaks (or at least in part).

@mkubecek said in #11:

> I'm giving up, I really prefer a simple, logical and consistent explanation over artificial hypotheses which can be easily proven wrong by reading about the algorithms and inspecting the code. But perhaps one more hint: have you noticed that these peaks are best seen in bullet and blitz ratings (on the right side), much less so in rapid (where they essentially stop around 2000) and are almost misssing from the right side of the classical graph (but are, in more complicated form, present on the left side instead)? Assuming the natural psychological explanation, it makes pretty good sense. But I really don't see how I would explain it if I assumed some fictitious irregularities in rating calculation process.

yeah, but the peaks don't stop at 2000 on rapid - they're visually consistently relative towards the end of the graph (2800) - less pronounced because of less people, but visually an equal proportion throughout. So we're still doing this resting all the way up to that level? I don't buy it. I'm not suggesting anything about some fictitious irregularities in rating calculation process! I too favour the simple solution over the complicated, true, but sometimes it's the simple solution that is the assumption. I think there's simply a more complicated reason that we've all looked over. - I can see people getting frustrated with me, but this currently trending assumption is full of holes. Let's see some evidence. Someone help me out with how to view these resting points, and I'll try my hardest to compile a graph of them. And with any luck, it'll turn out that I'm completely wrong. I'd like that. But like I said before, there are just too many people not doing that relatively to people that are for the peak to be so pronounced. If the peak was more insignificant, I'd stop going on about it! Let me prove me wrong. someone help me please.

@griffindabeast said in #9:
> @Jimothy-Jangles I try to bring up my ratings 100 at a time and I won’t touch them untill I bring up the other ratings. I would like to maybe get 2000 in all variants and never touch any of them again besides crazyhouse. I’ll play blitz/bullet again once I hit 2200 in rapid. I’m sure it’s a common thought. Same on chess.com I hit 2000 blitz/2000rapid and I won’t touch them again untill 2000 bullet.

mate, if I was 2000 on rapid I'd probably stop playing chess completely. I think I just play online chess to make sure I'm better than average so that if there happens to be a chess board in someone's house, I've got a fair chance of winning - or more importantly, not completely embarrassing myself! Although, saying that, I can't stop playing (even though I actually don't particularly enjoy it - what's that about?). I've never got higher than mid 1800s, and that was only due to the fact that it was early days (with the big rating changes per win) and I just got lucky a few times. Have a good day my friend.

@Jimothy-Jangles trust me it’s overrated. I feel the people at my level are actually very bad at the game still.

This has gone long enough. simplest explanation is correct i.e people take breaks when reaching round numbers. lets not spam with this fruitless discussion anymore.