How do two-day title matches work?

[Elom]

I remember once reading that it's nearly impossible today to hold an international tournament with a two-day title match and long time settings like the Kisei, Meijin and Houninbou in Japan.

It seems a shame. Maybe a business owner with excellent organisational skills could combine a use of anechoic chambers, signal-free zones and closed accommodation to make it happen one day. It would be nice to alternate it with a popular-style event such as the one I wrote about on Funny Tournament. But it makes me wonder; how do two-day title matches in Japan work?

[EdLee]

Hi Elom,

AFAIK they just do it, like they have been for years.
( I thought a day-2 Meijin game is tonight (PST)? )
Of course, things are different since 2015 AG.
Now anyone with a laptop can analyze the game overnight.
IMO they should just stop it and switch to all one-day games.
With the engines now widely available, someone can do some analysis
to quantitatively verify how the "quality" of the moves compares between 2-day and 1-day games.

( 1-day means ~4 hrs(?) + overtime for each person; 2-day means ~8 hrs(?) + overtime per person ? I wouldn't be surprised if the move qualify difference is 'negligible', in which case there's no good reason to continue 2-day games -- OC, advertisers and sponsors may disagree ).

[Elom]

Sorry, I didn't make it clear that I was wondering more about how they'd normally done so from before AlphaGo.

I found an great SL page explaining the system (for example, it seemed to say that in what we call eight-hour-a-side matches each player actually begins with what we call byo-yomi; 580 periods of 60 seconds for the entire match. Near the end of any players time, the timekeeper can count out the seconds).

But is each player expected to be able to set up and analyse the game on a board between days?

[EdLee]

Hi Elom,

But is each player expected to be able to set up and analyse the game on a board between days?

( Sorry, I misread days as plays. )
Based on every clip I've seen, no. The two people just sat staring at the board, like any formal tourney. Except for the breaks for food, water, and restroom. I don't know the rules for what happens after the sealed move and before it's unsealed.
( But at least one episode of Ninzaburō was about a top Shogi match, and another episode about Go. )

____
The URL tags cannot handle certain characters, such as ō and kanji:
[url=https://en.wikipedia.org/wiki/Furuhata_Ninzaburō]Ninzaburō[/url]
[url=https://ja.wikipedia.org/wiki/古畑任三郎]古畑任三郎[/url]

[moha]

But is each player expected to be able to set up and analyse the game on a board between days?

Although I have no real info on this, I think the very reason for sealing the move is to reduce the advantage/difference for the two sides from this (unavoidable) analysis. I wonder what impact bots have on this today (maybe not too much).

[dfan]

But is each player expected to be able to set up and analyse the game on a board between days?

Although I have no real info on this, I think the very reason for sealing the move is to reduce the advantage/difference for the two sides from this (unavoidable) analysis. I wonder what impact bots have on this today (maybe not too much).

Professional chess has eliminated adjournments since computers became powerful. (The reason you give for the sealed move is also the one in chess.)

It was always kind of unfair anyway, since sometimes one side would have a team of grandmasters working overnight to analyze the adjourned position.

[sorin]

But is each player expected to be able to set up and analyse the game on a board between days?

Yes, of course. The sealed move is meant to make the situation fair, but other than that it is assumed the players can and will analyze the game without any external help during the break.

Pros can analyze the game in their head anyway, the board is just there for convenience

[Bill Spight]

I found an great SL page explaining the system (for example, it seemed to say that in what we call eight-hour-a-side matches each player actually begins with what we call byo-yomi; 580 periods of 60 seconds for the entire match. Near the end of any players time, the timekeeper can count out the seconds).

That many byoyomi periods of 60 seconds could lead to almost interminable games. You don't lose a byoyomi period unless you exceed the time for it. Having read Richard Hunter's article in the British Go Journal and having kibitzed pro games in the same room as they were being played, I suspect that Hunter may have been misled a bit. The only times I observed the time keeper announce no time was when the player had made a play almost immediately, not after several seconds.

Also, I was informed, but did not observe this myself, that some games had two forms of byoyomi, first, byoyomi periods with five minutes to make a move, followed by periods with one minute to make a move. That scheme does not fit with the idea of hundreds of one minute byoyomi periods.

However, if it was customary for the time keeper to keep track of time used by rounding down to the minutes used for each play, instead of rounding up or down to the nearest minute, then that is indeed like having hundreds of one minute byoyomi periods, as Hunter described it. It is also similar to Fischer Timing, with the time increased on average 30 sec. per move. A player who played 100 moves before going into byoyomi would have added an average of 50 min. to his time.

But is each player expected to be able to set up and analyse the game on a board between days?

Before modern timed events, it was often the case that important games were played in several sessions, even over months. Nobody expected the players not to analyze the game during breaks, or to do so alone. With modern adjourned games, those expectations did not change. In the AI era, we might expect pros to play very well on the second day.

[Bill Spight]

But is each player expected to be able to set up and analyse the game on a board between days?

Yes, of course. The sealed move is meant to make the situation fair, but other than that it is assumed the players can and will analyze the game without any external help during the break.

Is this a new expectation? Maeda was rumored to have found a move for Shusai in a famous game against Go Seigen, right?

[wolfking]

But is each player expected to be able to set up and analyse the game on a board between days?

Yes, of course. The sealed move is meant to make the situation fair, but other than that it is assumed the players can and will analyze the game without any external help during the break.

Is this a new expectation? Maeda was rumored to have found a move for Shusai in a famous game against Go Seigen, right?

I believe when the game is adjourned the players will sleep, eat etc in close proximity to each other and avoid outside contact. Not sure what they do with cell phones.

[Calvin Clark]

Are you sure that the demise of 2-day matches is mostly about preventing outside influence? I thought it was more about sponsors being keen on TV time controls. In any case, the go world doesn't seem to have gotten as weirdly paranoid as in top chess. No is claiming coded messages being sent in players' noodles. But maybe we don't hear those stories.

[John Fairbairn]

In any case, the go world doesn't seem to have gotten as weirdly paranoid as in top chess.

Aren't you forgetting the Shusai/Maeda case?

Go boards were provided in hotel rooms for players who wished to think overnight, so no-one even thought of objecting to them using the overnight time to think. And many players did. But a typical outcome was that they felt very tired the next day and made mistakes in the endgame. That would still apply today.

Outside help was precluded because the players were in the same hotels and outsiders were barred (this was called kanzume - being sealed in a tin can). But that was before the age of cell phones, of course.

However, one significant difference from chess, at least in domestic go, is that players belong to a guild. It behoves them to take a freemason approach to their career. I'd have thought they would hardly dare risk their membership of that guild by inviting suspicion through taking too many toilet breaks during play. And if we are talking just about the sealed move, well, that's just one move, and players usually aim for quiescence on the board anyway as the time control approaches.

[moha]

Professional chess has eliminated adjournments since computers became powerful.

This may also have to do with chess games taking less moves, so more natural to finish same day.

[Elom]

I doubt that the sponsors of Japanese two-day final tournament have much to worry about anytime soon. But international tournaments organised in Korea or China may be very different.

I don't mean to go into the attic and pull out the well-trodden 'Japanese professionals do poorly in international tournaments partly because of shorter time limits' rug, but I do wish there was at least one tournament specifically for long Japanese-style time settings. I'd appreciate it as I do fast tournaments— each has things the other doesn't. Is it hardly fair that one should dominate so much over the other? I guess it's a sign of the times.

[dfan]

Professional chess has eliminated adjournments since computers became powerful.

This may also have to do with chess games taking less moves, so more natural to finish same day.

Regardless of the number of moves in a chess game, adjournments used to be common and now they are extinct, due to the superhuman analysis abilities of chess engines.

Serious chess games used to regularly last over six hours, and now that is about the upper limit of their duration, since it's desirable to finish the game in one sitting.