please rollback the rating range change!

What people don't understand is that the other system allowed for exploitation!

By being able to set your opponent's range to much higher than your own, a new account with 1500? could only get very high rated opponents (and did, this has been seen countless times) and win and gain a very high rating. There are tons of ways to exploit the old system, and many people did! This is why we can't have nice things.

You can complain all you want about the new system, but the facts stay the same: It's harder to exploit this one than the older one, which is a good thing.

#9

If you go to the lobby, challenges with a rating difference of over 500 still appear as you can still set the range between 600 and 2900, so technically an elite GM picking up your challenge might still happen, but yeah you can all go back to chess dot lag and be happy with its disconnects and the site telling you you have abandoned and hence lost the game if you press F5 when those disconnects happen, but maybe if you complain to FIDE they might decide that the game you lost due to disconnect is a draw, after all.

@TCF_Namelecc But the new high rating would still be provisional, which is understood to be potentially very inaccurate. They would need to win vs the high rated players many times to get a high rating without the ?. And if they can do that, they probably are good and deserve the high rating.
Does it matter if sometimes people can have a higher than true provisional rating for one game because they were lucky?

@TCF_Namelecc I sincerely consider your "exploitation" argument to be rather invalid. If a new account with 1500? actually won against these very high rated opponents, maybe it was a good player indeed and deserved such a high provisional rating. Look at this from the other side - why should a grandmaster joining Lichess be forced to play some 1500 rated opponents on the beginning? Anyway, even if the new system makes the ratings a little bit more precise (which I seriously doubt), it is definitely not worth destroying our game experience. Remember, rating is just a number, nothing real depends on it.

@MrBarnesMoustache You apparently confuse two things. If you go to the lobby, you can pick up only challenges given by other people, with their specified boundaries. A top GM will never see and pick up my challenge, because I am not permitted to create a challenge with range including his rating. Did I explain it clearly enough? And the second part of your post, urging me to move to the other website because I dare to criticise changes here, is plainly pathetic - I hope that you see it now.

And I am telling you I can see challenges with over 500 points of rating difference in the lobby. "Did I explain it clearly enough?"

@MrBarnesMoustache Yes, I do see them too, but - according to the mechanism I explained - they theoretically shouldn't be visible after this update and I don't know why they are. Maybe it is some form of a bug which, alas, will be eventually removed. Maybe these challenges are from people with mobile Lichess app which they didn't update. We have no way to know, because the developers did not bother to explain any details of this update. Sad, isn't it?

Yup, it should be some kind of bug, because those challenges very rarely appear and by mentioning this, we might actually be helping Lichess to debug it xd

to Namelecc: What does exploit mean!?

ALSO, WHY IS CHESS.COM 100x more exploitable (for boosting) than lichess even though chess.com only allows you to play people -/+ 400 points (max) above your level!?

Outside the justification of the change, what worries me is the authoritarian and high-handed manner this change was incorporated. Is it too much to expect some form of transparency, let alone having a civilized discourse with the community discussing the drivers and benefits for this change? What's the point of being a non-profit community of developers and patrons if such crucial decisions are going to be taken arbitarily behind closed doors?

This is exactly the kind of autocratic and unaccountable behavior that has alienated several AAA gaming cos from their player base. Its not like there was a scarcity of opportunity to discuss this matter beforehand, but the fact that there was no effort from leadership to hold a discourse with the community asking for feedback or better ways of dealing with whatever issues were meant to be addresssed by this action. To a certain extent it could be understood that it may not have been an entirely productive endeavor to involve the community beforehand but atleast doing so would have demonstrated an accountability towards the various developers, patrons and players as stakeholders. Who knows the community may have churned out some great ideas towards addressing whatever issues they were dealing with here.

Coming back to the justification of this change, I don't see how this addresses sandbagging or boosting. A player can still cherry pick the opponents they want to play against by refusing to start the match when they are setup against any unfavorable opponents. This change just makes it more time consuming if someone were inclined to do so. Besides, the rating dropped/gained from losing/winning against a -/+500 opponent would be the same as against winning/losing against +/-1000 opponnents assuming their RDs are stable at around 45.

If someone created a fresh account and boosted their rating by winning consecutively against a few top rated players taking advantage of very high initial RDs then the problem lies somewhere else. The provisional nature of the high RDs make it a moot point whether one sandbags/boosts using +/-500 opponnets or +/-1000 opponents. If the issue was to address the sandbagging/boosting then its the RD system that needs fixing not the pairing methodolgy.

I've already expressed my view previously on this, and I'm not going to go over it again. I just want to demonstrate something to forum members.

These are my games since the matching change was made:

2090 vs 2145 - lichess.org/5dzp2avptLl7

2085 vs 2135 - lichess.org/gspqgGCwQtmu

2079 vs 2112 - lichess.org/U1hHhQrRFDrJ

2086 vs 2170 - lichess.org/thTi9K6afbA4

2082 vs 2112 - lichess.org/rB8yBOlZ9tQg

2088 vs 2213 - lichess.org/xI83iYYz2kw8

2084 vs 2120 - lichess.org/0tFTMsRXbS7W

2079 vs 2125 - lichess.org/MgjNxQk7EeDr

2086 vs 2140 - lichess.org/FwjAKhgdtRKT

2092 vs 2101 - lichess.org/kENzHzZK3n74

There has been only one that has been over 100 points outside of my rating range, and I haven't had one player in the 2250+ group that I enjoyed playing against. You simply won't be matched against higher rated players any more. It doesn't matter how you set the sliders, the site's algorithms will match you against people in your own rating range. This would always have been the case, but you could circumnavigate this by specifying a particular rating. Now you no longer have that option, so you won't be able to have the experience of playing higher rated players outside of tournaments.

Anyway, they've made their decision, I guess I'll have to try to improve.