Revisiting Automatch Time Settings: Data-Backed Proposal for New Automatch Settings on OGS

Are your sure that having a “prefer Fischer” option like we do right now splits the pool? I believe if you select “prefer Fischer” you will still match with Byo-Yomi games, you’re effectively in the same pool, no?

If you select “demand Fischer” then that’s mutually exclusive with someone who’s set “demand Byo-Yomi” but as long as those options are rare it doesn’t negatively affect the pool and largely just penalises the players that choose to set such restrictive options.

3 Likes

I’m not sure how the “preferred” option works. I assume it still tries to satisfy your preference first and only opens up to other settings after a while if unsuccessful?

2 Likes

I guess the KPMC is now using Fischer time and it’s mentioned the AGA is moving toward Fischer here

I suppose the EGC has been using Fischer in Toulouse and Leipzig, and those potential 3-4 hour games have been interesting (thinking of Leipzig).


I suppose we can compare the tournaments of the EGC for time controls European Tournament Calendar

Best I can understand/find info:
Fischer ~ 33ish
Byo~ 40
Canadian ~ 14 (but they might be as backups in some cases)
I don’t understand Tilburg.

Probably you can debate about inclusions/exclusions or double counting when the same city has multiple tournaments.

I guess it’s still probably about even though.

Fischer
  1. Orsay
  2. Olomouc
  3. Grenoble
  4. Trenčianske Teplice
  5. Paris/Levallois?
  6. Lille
  7. Legnano
  8. Hamburg EYGC
  9. Hamburg Euro Pro Championship
  10. Madrid
  11. Helsinki
  12. Paris International
  13. Dubrovnik Pair go
  14. Esch-Belval (Luxembourg)
  15. Oslo
  16. Prague
  17. Leipzig
  18. Strasbourg
  19. Not the London Open
  20. Wiselka
  21. Turku
  22. Osnabrück
  23. Patra
  24. Leipzig Bergmansturnier
  25. Utrecht
  26. Pardubice
  27. Toulouse EGC
  28. Edinburgh
  29. Bratislava Women’s championship
  30. Halle (S.)
  31. Rzeszów
  32. Frankfurt
  33. Mannheim
Byo-yomi
  1. Stockholm
  2. Olomouc
  3. Sokolov
  4. Mollet del Vallès
  5. Vatra Dornei
  6. Trenčianske Teplice
  7. Tampere
  8. Nicosia
  9. Vienna
  10. Murcia
  11. Nijmegen
  12. Wroclaw
  13. Rzeszów
  14. Seville
  15. Pisa
  16. Budapest
  17. Vilnius
  18. Madrid open
  19. Porec
  20. Strasbourg
  21. Budapest Grand Prix
  22. Lisbon
  23. Kraków
  24. Warsaw
  25. Vienna
  26. Macerata
  27. Čistabora, Molėtai dist. Summer go
  28. Beek Zomergo
  29. Zaostrog Sego + Open
  30. Vatra Dornei open Wygc
  31. Tatry
  32. Vatra Dornei Japanese amabassador cup
  33. Stockholm
  34. Tuchola
  35. Wrocław
  36. Pisa
  37. Dublin
  38. Istanbul
  39. Budapest Cup
  40. Budapest Championship
Canadian
  1. Copenhagen
  2. Olomouc
  3. Hamburg moustrap
  4. Nijmegen
  5. Bled
  6. Graz
  7. Leuven
  8. Berlin Chinacup
  9. Madrid open
  10. Hamburg Monkey jump
  11. Not the London Open (If needed)
  12. Beek Zomergo (If agreed)
  13. Damüls
  14. Brussels
Absolute
  1. Tilburg

It’s actually the case also that in some tournaments only the top boards use Fischer and the rest use byo-yomi. Like that I’ve included a tournament under more than one list. Or in some cases they just mention it depends on what clocks they have available like in Olomouc.

5 Likes

That’s very interesting, if those three are moving towards Fischer seems like it might make sense to switch that to be the default. I happen to also think Fischer time is superior from a game design perspective for a lot of reasons, so if there’s organizational pressure to move that direction too, it might be something to try.

12 Likes

Tilburg is 30 minutes absolute time per person, a common format for a 5 round 1 day tournament in the Netherlands.

Despite the organising bodies pushing for a certain time control it isn’t always adhered to for all tournaments, this is partially due to the material available to clubs. There’s still tournaments where at least part of the boards have analog clocks (digital ones for top boards and analog for the rest). With analog clocks often you will see canadian in tournaments.

3 Likes

When people create their own game requests (not counting automatch, just manual game creation):

  • About 63% go for Byo-Yomi
  • Fischer time comes in second at around 23%
  • Simple time is third with about 12%
  • The rest (Absolute, Canadian, and no time control) make up less than 3% combined
3 Likes

I think Fischer is the forward looking choice.

I also think that people get confused comparing main time and increments between byo yomi and Fischer.

People will need to get used to the change and I think they will soon enough. But some thought about time and increments (and caps I guess) would be needed.

I prefer longer main time and shorter increment for Fischer.

OT moan: I think we have twisted the time settings. Really we should be playing within the main time and using the byo yomi etc to elegantly finish games where some more moves are needed to do this. But people worked out that you can actually use byo yomi for a lot of the game rather than just for finishing up and so a culture has developed where byo yomi increment is all important.

8 Likes

I agree with Fischer 100%. Many reasons to as mentioned above.

I think we should make 5min default but I know many who prefer 20m so I think that button should still exist as “Classic” rather than default.

I’m a big fan of most things that increases automatic speeds. Once concern of mine is Handicap though. Personally I think handicap is very useful for increasing the range of opponents but only see negative feelings about it.

I wonder if people would prefer even games against stronger or weaker players even if the outcome is almost certainly going to be the stronger player. I also wonder if people would prefer the range of ranks to be less to force even games even if their opponent pool is less.

Then there are Dan players who will only play even but won’t play on OGS because it takes too long lol. Though Dan players are such a low percentage I’m not sure how much their opinion matters to the overall opinions of the community.

2 Likes

Daily reminder to add the number of players around your rank seeking the game to the buttons blitz, rapid, etc.

1 Like

Silly question but wouldn’t that number mostly be “0” or “1”?

Players around your rank are often also around each other’s rank, so you would expect them to be paired if there’s more than 1.

2 Likes

probably a rolling average over a certain number of minutes would be better in that case. Say the number of unique users queueing over the time period.

5 Likes

Like this, yes. It could help with statistics and motivating people to click on the button. Or demotivating…

It could also help choosing the mode with the most players if we go with both Rapid and Normal game modes.

2 Likes

I would be interested in a “estimated time to match” or even just a binary indicator that somebody is waiting for me :slight_smile:

8 Likes

I love fisher time.
I hate solving late-game-corner-invasion-tsumegos in 30 seconds.

10 Likes

I’m in two minds about this.
While I’m comfortable with Byo-Yomi, I’m intrigued by the idea of making Fischer the default. It feels like a modern approach with future potential, aligning well with OGS’s philosophy. The fact that major organizations are transitioning to Fischer adds weight to this option.

That said, a sudden shift to Fischer could be jarring for some players and definitely going to have some shock effect. To mitigate it, it would be crucial to retain the option to choose Time Control preferences in the settings. This way, players can go back and play Byo-Yomi if they feel uncomfortable.

On the other hand, I’d support switching to Fischer sooner rather than later.
In the recent “Blitz Western Go Challenge” organized by Clossius, many participants quickly shifted to prefer Blitz Fischer settings. This suggests that while there might be initial resistance, players could adapt fairly quickly. Fischer is simpler to understand and explain, which could be less intimidating for newcomers.

Perhaps the best course of action right now is to just introduce the new Rapid option. We could then revisit the idea of defaulting to Fischer later on, once players have had time to adapt. The data makes it clear that the current Blitz setting isn’t popular, and there’s a strong preference for quicker game options.

As for keeping Fischer AM option or not. While Byo-Yomi is favored by a majority, the 23% of players preferring Fischer isn’t negligible.

Given the divided opinions in the ongoing polls, it’s clear this isn’t a straightforward decision. Here’s how I see it:

  1. Fischer as Default: This could simplify things for new users who might even already be familiar with similar time controls from chess. It’s intuitive and ensures a more streamlined experience.

  2. Keep Byo-Yomi: This would suit the majority of existing player base and those transitioning from other Go servers like Fox or Tygem, where Byo-Yomi is standard.

Thinking it through, it might be more beneficial to cater to new and starting players. The existing Go community will likely adapt very quickly, whether they’re used to Byo-Yomi or not.

In the end, I don’t believe the difference will be earth-shattering. Most players will adapt and be fine with either choice.

5 Likes

A few thoughts

Yes but

I think people would get used to it. And if they really hate it, there’s custom…

I’m not sure about this. There will be a good chunk of the current byo yomi people who don’t actually care and whatever the default it will suit them.
And then there are those who think byo yomi suits them but actually find that they prefer Fischer once they are used to it.

But basically I agree with this:

7 Likes

I suspect that byo-yomi is more easily understandable by beginners. In the context of live games, it is certainly easier to manage, IMHO. If the voice is turned on, one always knows how much byo-yomi time is left; with Fischer, one has to check between every move. I expect that good readers (more advanced players in general) would prefer Fischer, which allows them to manipulate the available time. In contrast, I expect beginners and poor readers would prefer byo-yomi, since manipulating the time has far less value for them and is difficult to do.

The primary problem with long main time is that it magnifies the bad effects of an escaper who holds the tab open. With 5m + 5x30s, for example, a timeout is likely to waste little more than 5 minutes at worst. I think that shorter main time will also lead to fewer escaping (Stopped Playing) reports. After all, having someone bug out with 20 minutes left is a lot more annoying and provocative than an escaper with only 5 minutes on the clock.

3 Likes

I’m not sure what the means or looks like in practice

But we are taking about fairly rapid games here. The timeout problem you describe might be worse with byo yomi since you have to wait for all the periods to elapse whereas with Fischer time is not being banked if the player is not playing.
For a rapid game we could also use a fairly low cap of a few minutes.

1 Like

From my experience “escapers” are very rare, if at all existent, at upper sdk level, thought I don’t have as much experience in other ranks. I would more likely lose on time because of connections “move submission error” issues than encounter escaper problem.

Seems pretty speculative to me. The counterargument to this’s people who chose longer time control are more committal, thus will escape less often, which is also a speculation. What I want to say is that we’ll need data to back up your argument.

There should be a button to “accept win by timeout” in that case which would resolve that issue.

You have misinterpreted both statements that you quote. On the first it appears that you are placing a negative connotation on the word “manipulate.” You can manipulate the available time under Fischer by taking less time on easy moves and more time on hard moves—that’s the main value of Fischer.

My point about main time has nothing to do with byo-yomi versus Fischer; it is strictly about the consequences of large main time in byo-yomi.

1 Like