Stop lowering the rating deviation minimum!

It's an atrocity. I remember back in the day when it seemed that you could never go under 60 rd no matter how many games you played. Every match used to give you a boatload of points (or take those points away). Do you remember that smug feeling of having a solid winstreak and a horde of points to show for it, or the despair after being ground into a -100 point paste on a bad day? It felt so alive.

Now? I have a lousy 46.30 rating deviation. I spend 6 minutes of my life playing a game just to gain/lose 6 rating, are you serious? Trying to gain rating feels like trudging through the muck of the glicko swamplands. Finally cutting a path through the dense shrubbery of game after game while being harassed by a mosquito swarm of ragequitters -- only to discover that wow, congrats, +4 rating cause you played some dude 100 points below you. And I look at the commits and it's been lowered again, with plans to go lower! (github.com/ornicar/lila/commit/d71dce286e932162ee8be17ca1bf348b41b91f4b) You have to be kidding me. Is having slightly more accurate ratings really worth giving up the feeling of gaining a fat stack of rating? I don't think it is.

Shouldn't chess be exciting? Don't you love the feeling, the rush as you greedily admire those points you so rightfully earned after trouncing an inferior opponent? The anger after making that one stupid blunder against your low-leveled opponent that punished you by mauling your rating with a fat -15? The feeling of being alive as you ride on the rating rollercoaster?

It's time to do what's right for lichess. It's time to raise the minimum rating deviation back up to 60.

agree

"Is having slightly more accurate ratings really worth giving up the feeling..."

Well isn't it? You want to pretend you're doing better? You can do that without playing chess. Or maybe there's a plugin which shows a higher score than you actually have.

Sure, let's also disable functionality which depends upon ratings since we wouldn't want to break anything:
* lobby seek filters and pairing pools
* tournament restrictions, pairings and tiebreaks
* leaderboards
* cheat detection (probably)

@Doofenshmirtz dude, you do realize that you'll also potentially have lower scores than what you "actually have", right?

@Toadofsky Do you think that somehow all of these things didn't work back when rating deviation was higher?

#5 @fpvbmct Decreasing the RD floor improved all of these things; but yes, I am of the opinion that these things worked poorly before the reduction.

It may be true, that these things were improved, yet I do agree with the OP at the same time... getting on a winning steak used to be more rewarding, sometimes even boosting you over 150 points if you win enough.

At the same time however, it's probably better that we have a lower deviation because it's then harder for people who are creating an artificial rating by just grinding against much, much lower rated people to get a high rating. I remember many people were really mad about how some of these people were on the leaderboard, so I guess this does help fix those issues.

#7 Indeed. There is further room for improvement of the rating system (see online-go.com which I've referenced dozens of times) but it's difficult to develop and even more difficult to test.

@Toadofsky Why is online-go's system better in your opinion?

OGS explains forums.online-go.com/t/ogs-has-a-new-glicko-2-based-rating-system/13058 :

Grouping as envisioned by Prof. Glickman essentially works by considering many games at once, which for in-person tournaments is not a problem. For online play where we have a lot of ad-hoc games, we needed to get a little creative, so our solution is as follows: We maintain a running tally of up to 15 games recently played, along with a “base” glicko-2 rating and a “current” rating. Your “current” rating is the rating that everyone sees and is used when match making and computing handicaps. When you finish a game, we add the game to the list of games, and compute your new “current” rating by applying the results of those games to your “base” rating. When recomputing, we also use your opponents’ “current” ratings as opposed to simply the ratings they were when the game originally concluded. Using this technique consistently produced better results in our experiments, and we later found that Prof. Glickman applied some techniques akin to this in his Glicko-Boost work in 2010, which made us feel better about doing this as opposed to simply using the original ratings. Once 15 games has been played, the “current” rating is finalized and the list is reset. We’ll also finalize the “current” rating after 30 days, regardless of how many games have been played.