Then wouldn't you have to wait until it actually trickled down?

Yes, a very strong one. That means that as the strogest player rank changes, everybody shifts by the same amount. When the strongest player stops playing, if the 2nd strogest player is 2 stones weaker, then suddenly everybody jumps two stones.

Our current system sometimes has some changes when anchors are adjusted, but overall it is far more stable than anything based on a single player would be.

I'm not really fond of the stochastic walk ratings seem to take on KGS even when you don't play and I certainly don't like the expiration to '?' when you don't play for a while, but overall it may be the most accurate system in current use for players who play there regularly.

Other servers have had resets. IGS did some years ago (and I mean before the 2005 change, somewhere in the summer of 2002 when there was a 3-stone shift. I wasn't playing then so I don't know for sure. I guess right before than an IGS 1d was pretty darn kickass relatively speaking.)

If I wanted to get attached to a number, I'd buy a certificate.

_________________
The person who says "don't......" is just showing he doesn't know Go at all, I think. - Tei Meiko 9p

Pippen: my point was that if that's what you want, you can't complain about KGS, since it's just one rating system among many.

Aligning all the ratings systems would require collaboration from around the world, and would also happen to be impossible at the present time.

_________________
When you think your opponent has made a mistake, that's when you should be the most cautious.

It's not really stochastic, is it? It assumes that you would be able to beat the same people you were able to beat when you were last playing.

.

Yes, it assumes that while you were gone that you've been playing on Tygem.

_________________
"When you are learning, it is not so important whether you are right or wrong." - Bill Spight

Well, not necessarily. If your most recent partners decline, you'll decline to. It just assumes that, in the absence of evidence, you can still beat the same people and lose to the same people.

The amount of importance some people are placing on ranks in this thread is quite silly. The most important thing for a ranking system is to be internally consistent, ie trying to have differences in rank equate to 50/50 games with the appropriate number of handicap stones. Why does it matter if there is a "shift" in rankings and some bits associated with your account on a remote server change from 5k to 7k? All that matters is your actual strength which can never be characterized by a single number but by comparison.

_________________
Games * Rank Graph
7k

A system with sudden shifts IS inconsistent.

Robert, like it or not, he's arguing about a reasonably well-defined concept. The fact that you don't like using the word "consistent" in naming that concept (which could naturally be named "momentary internal consistency") is pretty much irrelevant.

_________________
When you think your opponent has made a mistake, that's when you should be the most cautious.

One should not call an inconsistent system consistent.

But he said internally consistent. If everyones rank is affected in the same way by the recalibration, the system will be just as good at predicting an even match as it was before.

_________________
"When you are learning, it is not so important whether you are right or wrong." - Bill Spight

A rating system moving too many players upwards is internally inconsistent because a rating system shall be able to distinguish players instead of creating heaps of fake strong subpopulations.

That sounds scary ...

_________________
Stay out of my territory! (W. White, aka Heisenberg)

You're begging the question! As I pointed out, a good descriptive name for the system he wants is "momentary internal consistency".

_________________
When you think your opponent has made a mistake, that's when you should be the most cautious.

You could be less obtuse, Robert, if instead you said "One should not call an erratic system consistent." That would make it clear that you are, indeed, making a synthetic claim and would point the way in two directions - first, you could clarify the extent to which the KGS system really is erratic, and second, you could clarify why, if we want a consistent system, we should make it less erratic as well.

The problem with inconsistency, as I see it, is that it leads to situations that violate assumed transitive properties. For example, if F always beats W, W always beats H, and H always beats F, it's difficult to apply any meaningful ranking to the {H, F, W} triad.

However, if a ranking system suddenly shifts, that merely means that it has adapted to new information. If it shifts a lot, that either means it gets lots of new information frequently, or it is very sensitive to what little it does get.

This is a more general requirement. Fine.



So the system outputs for each player: "I have adapted to new information! (I do not tell you which information, nor which adaption.)" ;)

IOW, not each adaption is good.

So your objection is not that it's erratic per se, but that wms will not share with you the information that goes into the daily iteration?

It seems to me that the ideal rating & handicapping system would strive to handicap every match to a 50% win rate. If a player's win rate is much higher or lower than 50%, that player is not being well served by the system. (we do occasionally get threads complaining about this, usually "I've won 10 games in a row and my rank hasn't gone up!")

If this is the metric we want to use, it's very easy to check the error: look at the average win-rate of every player over an appreciable number of games, and find the average distance from 50%. There was a competition a while back looking for improvements to ELO that used basically this metric on historical chess data, I believe.