I keep seeing claims that the KGS rank system encourages users to create new accounts in order to get a more accurate rank. But, I wonder how well this belief is justified?

As a player, not as an admin, I have seen a few users create new accounts because they think they deserve a higher rank. At first, they actually do get a higher rank. But, after several ranked games with the new account, their rank stabilizes to what it was with the previous account. The rank graph shows only a tiny, if any, difference.

However, my experience is very limited here. The few cases I have seen are insufficient for me to conclude the premise that creating a new account to get a more accurate rank is false. As an admin, I have seen several users who are "account abandoners" but I have not investigated to see if they change accounts in order to obtain improved ranks. Furthermore, those with the most accounts have the fewest ranked games in each of their accounts, and the number of ranked games in these accounts is too few to draw conclusions that changing accounts gives them a rank that is both higher and stable.

Yes, the KGS rank system does take into account all ranked games played in the last six months. However, the weight of a game result decays exponentially, so recent games have more influence on users' ranks.

In short, I do not believe the claim that the KGS rank system encourages users to make new accounts in order to get a higher rank. But, my experience is insufficient to decide either way.

If one does have a stable rank with a large history of ranked games and one does think one's rank should be different, I think it would be better to try to play games with handicaps that reflect what one thinks one's rank should be. For example, if you think you should be one stone better, as Black, take one fewer stone in handicap, or offer one more as White.

As was said in the other thread, the KGS rank system is not intended to correspond with rank systems elsewhere. It is intended to help users find fairly matched games.

KGS does not encourage it. Only users' egos do that.

_________________
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).

I've never done it but I've seen one or two dan players do it specifically to see if they could hold a rank higher than their current one if they were playing a lot and felt their history was holding their rank down a bit (i.e. they thought their rank would go up on their current account but that it would take a fair few games to happen because of the time window vs how active they were). Both times I saw someone do this they held the new rank without problems for at least a few months of heavy play.

I can understand it for people who play a lot and I mean a lot, and who if they gain strength have to rack up a lot of wins to change their rank due to how many games they've played at their current rank. Really though the only examples I've seen have been around 4 or 5 dan KGS were in fairness it's not exactly a rank a month so someone could spend rather a long time at a certain rank whilst improving and most of the time it was curiosity to see whether they could hold the new rank or not rather than the need to rank up a bit faster that drove them.

I don't see the argument for 5k players though. It's less likely that they're improving and their rank not budging for 3-4 months if they were playing a lot.

I have found a lot of people do create new accounts to get a more dynamic rank change. In the end, as you point out, most do end up in the same spot.

Other servers have encouraged me to also make new accounts since it can take many more games to move ranks there. There are pros and cons of various systems. It would probably be best in my opinion to keep KGS able to quickly get to a correct range but be a bit more dynamic on changes than it currently is.

I was going to say that I think it is not the KGS rank system so much as users repeating the claim. You make a good point users' egos is part of it.


I found a page at Sensei's, which makes some claims about the rank system. I do not know how authorative the editors are, though.
http://senseis.xmp.net/?KGSRatingMath

I've seen players with 10 game rated win streaks and not rank up. I think an ideal system would be a bit more dynamic than that for example. There's nothing wrong with more players ping ponging between ranks I think.

Depending on how frequently you play, 10 games in a row may still be a statistical anomaly. You never hear anyone complaining that they kept their rank after losing 10 games in a row either, after all.

Sure, but it's no problem going up a rank and down a rank when they win or lose games. It's just a more dynamic system than what KGS currently has and more similar to other servers.

UnclMartin, you are asking for people's opinions here, but what I said is based on a fact - you can know this if you understand how the KGS ranking system works. If someone plays a lot of games, they have a lot of game history. Because that game history is included in the rank calculation, the more games that are played, the more weight history has on one's rank.

If someone doesn't improve and is always a 5k, playing a lot of games just reenforces this. But if someone improves from, say 5k to 1d, they necessarily have to play enough games to counter the effect of their history as a 5k. There are some people that don't improve fast, and as ranks move up, this becomes more of a reality. But it is a *fact* that using history in the calculation requires more games to force the rank to change.

This phenomenon does not occur in a system like Tygem, which only counts the last 20 games. If you play 10,000 games as a 5k, and then start studying a lot and improve in strength, within 20 games, you can change your rank. With KGS, this won't work.

Some people prefer this type of a system - I'm not making a claim against that. But if you suddenly improve, your game history hurts you a lot more on KGS than it does on other types of systems.

_________________
be immersed

I have seen a decent number of players on KGS that started out as beginners, got really into Go, and played hundreds of games every month. In that case somewhere in the 10k-5k range their rating starts to get extremely stuck. In those cases I advise them to make a new account. Most of the ones who do immediately go up 1-2 ranks, and it doesn't take long until they go up another rank.

Overall I don't think it's such a big deal that the rating system does this, it's only going to happen once or twice in a player's lifetime.

The "more weight" thing is not exactly accurate. If you play games at a consistent rate, your rank will change at the same rate regardless of how many games you are playing. If you suddenly became significantly stronger, it may take you about 10-14 days to promote, assuming you do not take any special steps correct it. After that initial lag period your progress toward your true rating will accelerate until you get close.

You can expedite this process by playing games handicapped where you feel you should be rated, and the system will correct without penalizing those who help you get your new rating.. Realistically, if you are playing 55 games/day as you suggest, I would hardly consider a ten game winning streak a sign of improvement.

Do you think there's something significantly wrong with having three ranks during those 55 games/day?

I don't know if this is the case for KGS but it is annoying on IGS that the promotion bar is higher(demotion is lower) than the actual rank, ostensibly so that players ranks are stable and not oscillating, but it is kind of annoying to be stuck just under the point threshold then cross it and win like six or seven in a row. Especially if you are looking at stable point where you could be a 1kyu for years when you could just as well be a 1 dan.

I think that a rating system using nothing but game results as input data that would vary an otherwise consistent player by 3 stones in a day using a sample set of less than 1% of their games would have very poor predictive capabilities. I also think that it would imply that the stone-based handicap system is poorly calibrated under that system.

Let me clarify what I mean by "more weight". On Tygem, I have no fear of playing ranked games. If I get into a slump, I know that if I get my act together, within 20 games, I can move my rank back up.

On KGS, I *am* afraid to play ranked games. That's because, if I screw up, it will be harder in the future to achieve the same rank. If I have a bad day and lose 10 games, then even if I have a good day tomorrow, it takes longer to get my rank back up.

So while some people have different rating system preferences, my opinion is simple: I'm afraid to play a lot on KGS. I'm not afraid to play a lot on Tygem. Therefore, I prefer Tygem's system.

---

This begs the question, why am I afraid to play? Isn't it my "ego" that's making me want to keep my rank up?

To that, I can direct you to the advice that many have given about playing stronger players. People don't often say to play games as a way to improve, but when they do, it's almost universally that you should play strong players. With KGS, if I screw up because I'm having a bad day, I will not have the opportunity to play strong players for a much longer time, given that I play a lot of games.

Now if someone is not interested in improving, or if they want to be the same rank for a long time, then KGS is the system for them - play a lot of games, and the rank will converge.

The rank still converges on Tygem, too, for these players, but allows more volitility for those that are interested in challenging themselves the most.

_________________
be immersed

@Mef: And I won't argue with you that the KGS ranking system is a feasible one. I simply argue that other systems, such as Tygems, are more volitile. And for that reason, I have no fear of getting stuck at a rank and having to play 10s or 100s of games to change it.

On KGS, you might note, I have played maybe less than 10 games in the last few months, because I want to play only when I am confident that I can have a good game. Otherwise, I'll have to win a ton more games to be able to play stronger players. I've played more than 10 in a day on Tygem. There's your difference.

_________________
be immersed

Except that this is not true. As long as you are playing games at a consistent rate, you will see no difference in how your rank moves. Much like when you do well you feel our rank doesn't go up as much as you feel it should...When you have the slump on KGS, your rank will not go go down as much as you feel it should. Any disparity you think exists is from selective perception.




This is also not true. Tygem's system promotes Brownian motion more than it promotes convergence. If you will allow me a thought experiment I can demonstrate why:

Imagine you have a group of 11 friends who are all 10k, and they decide they want to improve so they all sign up as 1d on Tygem. In order to track their progress against themselves, every week they each play everyone else in the group once, then play 5 games against other people of the same rank. If they always win against each other 50% of the time, and lose against everyone else they play, then they will all maintain 33% win rates and they will never demote. Note: this is with no malicious intent by any player as well as every player always trying their best to win every game.

What's more -- this type of example could be used to keep any arbitrarily strong or arbitrarily weak group at any arbitrary rank. You could quite literally maintain a set of random playing bots at 5d. This is all without foul play on the part of the players. This example also allows you to assume that everyone else on the server is properly ranked. Once you introduce inaccuracy and random fluctuations even more crazy things could happen.

This is because tygem assumes that any given player has a 50% chance of winning a rated game at any given time. All wins and losses are treated equally.

Any ranking system that attempts to have predictive power will take that group of 11 and have them spiral down relative to the aggregate until they settle at where they should be.

Surely Mef, (if I understand it correctly) the crucial flaw in KGS's system is that it isn't the rank a person had when you beat them but the rank you had now that is counted in the calculation? There's an assumption in the system that someone I beat two months ago I could beat today even if they've improved. Or if they got worse the opposite effect would occur. This strikes me as a rather strange way of calculating someone's rank.

If I misunderstand it please correct me.

You are correct with one caveat. This is all true, in the absence of new information being available to the system. If both you and your old opponent have continued playing games (against each other, or against other people) in those two intervening months, that single game will only carry about 1/4 of the weight of a new game being played. This is actually a major strength in KGS's algorithm that allows it to provide reasonable ratings with limited data. On other servers you may need to play 20 games before you can have a solid rank. On KGS it is usually accomplished with 2-4 games. There was a study done using KGS data that compared time-dependent algorithms to incremental algorithms (Elo, Glicko, etc) and the former were found to have superior predictive capability.

KGS has succeeded in getting you to treat every single rated game as seriously as possible.