It is currently Sun Sep 22, 2019 5:12 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 111 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next
Author Message
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #41 Posted: Wed Sep 30, 2015 8:15 am 
Honinbo

Posts: 8629
Liked others: 1485
Was liked: 1462
KGS: Kirby
Tygem: 커비라고해
I still feel that a rating snapshot at time labeled X shouldn't include data after X. That way, each rating snapshot is comparable to the snapshot of today, what's listed on the front page of goratings.org. For today's snapshot, future data isn't available, so to make earlier snapshots 1-to-1 comparable, I'd assume the same method to be used for those calculations.

For past snapshots, future data may allow to make a "smoother" prediction of rating, but it is inconsistent with the snapshot as of today, since today's ratings don't have future data to work with.

_________________
it's be happy, not achieve happiness

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #42 Posted: Wed Sep 30, 2015 8:37 am 
Oza
User avatar

Posts: 2152
Location: Tokyo, Japan
Liked others: 1996
Was liked: 1216
Rank: Jp 6 dan
KGS: ez4u
John Fairbairn wrote:
This seems to me to be a variant of the precision versus accuracy debate. I'd have thought precision was more relevant in historical ratings (as here). In other words, we want to precisely know how good someone's peak was, not with the peak worn down by performances when a player was past his prime.

If we include past games but not future, we are not measuring the peak (until we are well past it anyway). We are measuring the average slope leading up to the peak. I believe that the 'intuitiveness' of including past games is just another example of what Kahneman calls 'Answering an Easier Question' in chapter 9 of his book "Thinking, Fast and Slow", which Tami reminded us of a few posts ago in another thread (and I was looking back through earlier today ;-) ).

What do I (or he really) mean? We would like to answer the question,"Who is the strongest Go player in the world right now?" Alternatively, "Where does (insert the name of your favorite player) rank in the world right now?" These are extremely difficult questions due mainly to the immediacy of 'right now'. Ask yourself how we would possibly answer those exact questions. We never do (except inconclusively by arguing back and forth on these forums and any number of bars around the world).

Instead of answering these questions we substitute easier questions such as, "Who is the highest rated player in the world right now based on the XYZ rating algorithm, which uses historical data from the last 180 days (or 1 year, 5 years, or all historical data on games played)?" This is a much easier question and has become the standard approach in chess, tennis, golf, and various other sports and games, including Go. The answer produced is not without controversy due to the presence of competing algorithms and alternative data sources but we are used to this approach. We have 'learned' that this is the proper way these things are done.

We don't like it when some wise guy like Remi wants to rock the boat and include the future too. That undercuts the story that we have been telling ourselves about the 'proper' way to do things. Not only that but once we open that door to the future we also undercut all the current results. The ratings for August will have to be revised when we finally see the results of October's games (face palm)! But in fact isn't that exactly what happens with more informal reputations and interpretations of events? So-and-so wins a couple of titles, goes 25 and 2 over six months and is obviously the next great hope. Then (s)he goes 4 and 15 over the next four months and it becomes 'obvious' that they were just a flash in the pan, beneficiary of a run of luck. Well we have just adjusted our opinion retroactively. Most of us at least will not continue to believe that So-and-so was a tremendously gifted player who unfortunately peaked and declined in a matter of months. We will in fact reevaluate our past judgment and conclude that they were just lucky. Notice that if we have some handy answer (e.g. So-and-so suffers a serious stroke and is clearly mentally and physically impaired), we may be able to get away with continuing to believe that they had the stuff of true greatness back at that precise point in time. That is, the future will never contradict the story that we built in the past.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21


This post by ez4u was liked by 4 people: dfan, hyperpape, macelee, Uberdude
Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #43 Posted: Wed Sep 30, 2015 8:39 am 
Gosei
User avatar

Posts: 1744
Liked others: 696
Was liked: 288
KGS: greendemon
Tygem: greendemon
DGS: smaragdaemon
OGS: emeraldemon
How about this situation: John F finds a never-before-seen game between two players, say Inoue Genan Inseki against Honinbo Jowa in 1823. Because John has become so rich from finding this game, he offers you $1000 if you can guess who won the game before he shows it to everyone. Do you look at the games from 1824 in making your guess? Or do you only look at games played before the new-found game?


This post by emeraldemon was liked by 2 people: dfan, ez4u
Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #44 Posted: Wed Sep 30, 2015 8:44 am 
Honinbo

Posts: 8629
Liked others: 1485
Was liked: 1462
KGS: Kirby
Tygem: 커비라고해
The more I think about this, the more I agree with the way the historical lists are currently implemented. My concern earlier was that there's inconsistency between today's rating list and the historical lists, since future data is not available for today's rating list.

But if you have the data for the historical lists, why not use it?

I guess my position changed, in light of the new data I observed from this discussion :-)

_________________
it's be happy, not achieve happiness


This post by Kirby was liked by 2 people: emeraldemon, wineandgolover
Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #45 Posted: Wed Sep 30, 2015 8:54 am 
Oza
User avatar

Posts: 2152
Location: Tokyo, Japan
Liked others: 1996
Was liked: 1216
Rank: Jp 6 dan
KGS: ez4u
Kirby wrote:
The more I think about this, the more I agree with the way the historical lists are currently implemented. My concern earlier was that there's inconsistency between today's rating list and the historical lists, since future data is not available for today's rating list.

But if you have the data for the historical lists, why not use it?

I guess my position changed, in light of the new data I observed from this discussion :-)

Wow! With that kind of flexibility on display I am worried you might be outgrowing L19.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21


This post by ez4u was liked by: Bonobo
Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #46 Posted: Wed Sep 30, 2015 12:22 pm 
Honinbo

Posts: 8629
Liked others: 1485
Was liked: 1462
KGS: Kirby
Tygem: 커비라고해
ez4u wrote:
Wow! With that kind of flexibility on display I am worried you might be outgrowing L19.


Nah, I've always been flexible. You can just use this discussion as a datapoint to reassess your view of my past disposition ;-)

_________________
it's be happy, not achieve happiness


This post by Kirby was liked by 2 people: Bonobo, ez4u
Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #47 Posted: Thu Oct 08, 2015 2:11 pm 
Lives in sente

Posts: 905
Location: UK
Liked others: 68
Was liked: 469
Rank: 5 dan
KGS: macelee
Thanks for the go rating data feed provided by Remi, I am now able to present the rating data inside Go4Go.net, together with the photos of the top ranked players. This is an ongoing work as I am still compiling and editing the photos which is quite time-consuming. And apparently there's a bug in my responsive theme somehow the page layout isn't right in some browsers. But I am making it available now anyway because this is a very interesting work.

http://www.go4go.net/go/players/rank

Or find it from the Go Players > Top-ranked Players menu.


This post by macelee was liked by 4 people: ez4u, jeromie, Kirby, wineandgolover
Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #48 Posted: Thu Oct 08, 2015 10:33 pm 
Lives in gote

Posts: 547
Liked others: 13
Was liked: 17
Rank: KGS 2dan
KGS: emerus
I predict that Ke Jie holds onto #1 for good this time. The next 3-4 years at least.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #49 Posted: Wed Oct 21, 2015 12:11 pm 
Lives in sente

Posts: 905
Location: UK
Liked others: 68
Was liked: 469
Rank: 5 dan
KGS: macelee
Remi, can you please check why a rather strong player Lee Yeongkyu is missing from your list? He has 454 games in Go4Go database.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #50 Posted: Wed Oct 21, 2015 12:15 pm 
Lives with ko

Posts: 152
Liked others: 29
Was liked: 100
Rank: KGS 4 kyu
macelee wrote:
Remi, can you please check why a rather strong player Lee Yeongkyu is missing from your list? He has 454 games in Go4Go database.

http://www.goratings.org/players/303.html
He does not appear in the current rating list because he has played no game in more than one year. He still appears in the historical ratings lists in the years when he was active.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #51 Posted: Wed Oct 21, 2015 12:32 pm 
Lives in sente

Posts: 905
Location: UK
Liked others: 68
Was liked: 469
Rank: 5 dan
KGS: macelee
OK. He might be in his mandatory service in the army as many have to do.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #52 Posted: Wed Oct 21, 2015 1:01 pm 
Honinbo

Posts: 8629
Liked others: 1485
Was liked: 1462
KGS: Kirby
Tygem: 커비라고해
macelee wrote:
OK. He might be in his mandatory service in the army as many have to do.


From here his profile (https://www.baduk.or.kr/info/player_vie ... e=10000261):

Quote:
12.26 논산훈련소 입소 - >공익근무요원으로 복무


So I guess he's serving as a public service worker. I don't know about his particular situation, but typically you have to serve for 2 to 3 years. Since he entered at the end of 2013, maybe we can look forward to more games from him in the near future.

_________________
it's be happy, not achieve happiness

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #53 Posted: Wed Oct 21, 2015 1:29 pm 
Lives in sente

Posts: 905
Location: UK
Liked others: 68
Was liked: 469
Rank: 5 dan
KGS: macelee
I think strong players are very unlikely to just go away and stop playing. They return sooner or later. Maybe it is sensible to just keep them in the main list but noting the date of his last available game.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #54 Posted: Wed Oct 21, 2015 6:02 pm 
Lives in gote

Posts: 628
Location: Austin, Texas, USA
Liked others: 45
Was liked: 209
How about an option or different page that includes all players?

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #55 Posted: Wed Oct 21, 2015 8:13 pm 
Oza
User avatar

Posts: 2152
Location: Tokyo, Japan
Liked others: 1996
Was liked: 1216
Rank: Jp 6 dan
KGS: ez4u
macelee wrote:
I think strong players are very unlikely to just go away and stop playing. They return sooner or later. Maybe it is sensible to just keep them in the main list but noting the date of his last available game.

Of course strong players (and weak players) routinely go away and stop playing. It is called getting old ( :sad: ) and happens far more often than players stopping and then coming back. An 'active' list makes sense to me. The Chinese pro rating list operates that way although I have no idea what parameters are used.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #56 Posted: Wed Oct 21, 2015 11:21 pm 
Lives in sente

Posts: 905
Location: UK
Liked others: 68
Was liked: 469
Rank: 5 dan
KGS: macelee
ez4u wrote:
macelee wrote:
I think strong players are very unlikely to just go away and stop playing. They return sooner or later. Maybe it is sensible to just keep them in the main list but noting the date of his last available game.

Of course strong players (and weak players) routinely go away and stop playing. It is called getting old ( :sad: ) and happens far more often than players stopping and then coming back. An 'active' list makes sense to me. The Chinese pro rating list operates that way although I have no idea what parameters are used.


The Chinese system are simple - players need to register every year with the Go Association if then intend to keep playing. If they don't, they are consider inactive and move to another list.

There is another factor to consider. The Chinese list is calculated in such as way that new games won't affect old ratings. This isn't the case in Remi's algorithm.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #57 Posted: Thu Oct 22, 2015 1:03 am 
Lives with ko

Posts: 152
Liked others: 29
Was liked: 100
Rank: KGS 4 kyu
I increased the inactivity threshold to 1000 days, so he is back into the list.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #58 Posted: Thu Oct 29, 2015 10:39 am 
Lives in sente

Posts: 905
Location: UK
Liked others: 68
Was liked: 469
Rank: 5 dan
KGS: macelee
Hi Rémi, I wonder if you are able to conduct an experiment using your algorithm.

Now Iyama Yuta is in red-hot form, winning all his games over the past three months. This put him at No. 5 on your list (in fact he temporarily went to No. 4 above Shi Yue two days ago). I wonder how reliable your algorithm reflects his real strength. After all his games are mostly against opponents significantly weaker than him. And each year he only plays very few games against strong Korean and Chinese players.

My data does not support Iyama being at top 5. His score against the remaining 19 players in top 20 is a pathetic 12-21 (36.3%). It is also quite surprising that he never played any games again 6 of the top 20 players.

I believe your rating list quite realistically reflects the relative strength among Korean and Chinese players. They so frequently play each other, not just during international events, but also the Chinese league. With large sample size, the dynamics of your algorithm should work really well. But because there are much fewer games connecting Japanese players with the rest, the Japanese ratings are harder to justify.

I propose you run the following experiment. Create a number of (say 50 or 100) fictional games between top 10 Japanese players (not including Iyama) and Korean/Chinese opponents of similar ratings. Randomly generate the outcome. The purpose of this is to make more links between Japanese players and the rest. Then recalculate everything to see if this has any impact on Iyama's rating.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #59 Posted: Thu Oct 29, 2015 11:47 am 
Lives with ko

Posts: 152
Liked others: 29
Was liked: 100
Rank: KGS 4 kyu
macelee wrote:
Hi Rémi, I wonder if you are able to conduct an experiment using your algorithm.

Now Iyama Yuta is in red-hot form, winning all his games over the past three months. This put him at No. 5 on your list (in fact he temporarily went to No. 4 above Shi Yue two days ago). I wonder how reliable your algorithm reflects his real strength. After all his games are mostly against opponents significantly weaker than him. And each year he only plays very few games against strong Korean and Chinese players.

My data does not support Iyama being at top 5. His score against the remaining 19 players in top 20 is a pathetic 12-21 (36.3%). It is also quite surprising that he never played any games again 6 of the top 20 players.

I believe your rating list quite realistically reflects the relative strength among Korean and Chinese players. They so frequently play each other, not just during international events, but also the Chinese league. With large sample size, the dynamics of your algorithm should work really well. But because there are much fewer games connecting Japanese players with the rest, the Japanese ratings are harder to justify.

I propose you run the following experiment. Create a number of (say 50 or 100) fictional games between top 10 Japanese players (not including Iyama) and Korean/Chinese opponents of similar ratings. Randomly generate the outcome. The purpose of this is to make more links between Japanese players and the rest. Then recalculate everything to see if this has any impact on Iyama's rating.


Hi,

WHR is designed in such a way that these additional games would not change the ratings if the results are generated according to the current ratings. Iyama's rating is based mainly on his games against Japanese players. If the ratings of the Japanese players don't change, then his rating won't change either.

What I could do is display the likelihood of superiority matrix.

You can see such a matrix at the bottom of that page:
http://www.computerchess.org.uk/ccrl/4040/

Likelihood of superiority of Japanese over non-Japanese should rarely be close to 100%, because they have so few international games.

I am busy with much more important projects now, so it won't be in the short term.

Top
 Profile  
 
Offline
 Post subject: Re: http://www.goratings.org/ now has historical ratings lis
Post #60 Posted: Thu Oct 29, 2015 4:32 pm 
Tengen

Posts: 4268
Location: North Carolina
Liked others: 447
Was liked: 702
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 5k
WHR is described in public--I even believe there might be open source implementations. I've often thought it would be interesting to do simulations on separated populations that play each other infrequently--how will your predictions do? In particular, what happens to a player like Iyama who dominates his local competition?

But you'd have to, y'know...actually run those simulations, and I haven't done that.

Another test would be to see the accuracy of predictions with regard to international matches. Is it within the normal range?

P.S. Surely there's a timing problem with citing your data. Both goratings and Dr. Bae Taeil's data show Iyama getting even stronger in the past year or two. But his international record reflects more than 5 years of play.

_________________
Occupy Babel!

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 111 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group