sorin wrote:jlt wrote:sorin wrote:I don't know about Lee Changho in particular, but I heard from multiple sources that playing with short time limits does very little to the quality of a pro's game in general.
I heard that pros estimate the difference to be only 1-2 points.
That sounds counter-intuitive to me. Amateurs play much better with longer thinking times. Leelazero plays much better with 10000 visits than with 100 visits. How can that be different for pros? What if they have to solve a complicated life-and-death problem? Does it mean that if they can't find the solution in 30 seconds, then they won't find it in 30 minutes either?
It sounds counter-intuitive to me too, but I read it enough that I come to accept it.
It tells how strong pros are, how fast they can read.
It also tells how large of a gap 1-2 points is for pros
I hesitate to offer much of an opinion, but here goes. Partly this is based upon research done years ago on chess.
1) If a play depends upon local reading, up to 40 ply, top pros will normally see the right play almost instantaneously. A lot of that ability has to do with previous experience against other pros. (Based upon chess research.)
2) (Based upon environmental go games, at a pace of 2-3 hours total time for a game or endgame.) I think that the pros are optimistic.
Example 1. In one endgame tournament (Edit: Everyone started from the same early endgame position), one pro let Rui Naiwei kill one of his groups. All the pros got a big laugh over that.
Example 2. In the first environmental go game, both Jujo and Naiwei overlooked a play that gained more than 3 pts., even making plays that gained only 1 pt., and finally Jujo made the wrong local play. The potential swing from these errors was up to 5 pts. That evening I discovered Jujo's error, after much study. The next morning, as I was showing the position to Jujo, Naiwei passed by, looked at the board for two seconds, and rattled off a sequence that she had overlooked during the game. (Edit: The point being that her ability to see the correct sequence very quickly did not let her see it during actual play.)
Example 3. In reviewing another Jujo-Naiwei game, I discovered a ko at the end of a 23 ply sequence. It took me an hour and a half to find it and verify the sequence. The next day Jujo came up to me and proudly told me that Naiwei had discovered a ko in that area. I don't know how long it took her, presumably much less time than it had taken me, but nobody had seen it during the game. I think it's a good bet that Naiwei would have found it during the game, given enough clock time.
3) Over decades, pros collectively can discover joseki mistakes on the order of 1 pt. or so. Modern top bots have shown that they still make sizable errors of which they have been unaware. Whether they can reduce these errors with more clock time is unclear, but possible.
