I guess if we were arguing to replace normal, then this wouldn’t be a good argument since it’s like 3 times more games than
As a request for an additional rapid setting at either 10m+30s or 5min+30s it definitely makes sense since it’s a significantly high amount of games comparable to the other settings like for 13x13 etc
One can argue in some case that maybe some people choose the current automatch because it’s simpler than figuring out a custom game, and it’s plausible and I guess future statistics with a rapid button enabled will show whether the live 19x19 people move to rapid once it’s on offer.
It appears like maybe in the faster time settings (blitz) the Fischer time control becomes a bit more favourable. It could be that people think Fischer is generally better or it could be that the current 10s byo-yomi is too fast and more inflexible than usual (like a 30s byo-yomi), except maybe in 9x9 games.
In the above for “normal”/live time games, it seems like people don’t as heavily swap to Fischer
I don’t know that any of the above says that rapid should be Fischer time though. The big custom rapid requests we see are byo-yomi.
@AfricanRhino17 whats the largest custom request that uses Fischer in a roughly rapid time setting? Is that easy to compile or pick out?
Fischer has positives and negatives in Go:
positives: more flexible thinking time, you get to manage the whole amount of time, you never waste any time, all can be used compared to losing small or large fractions of a byo-yomi period.
negatives: kos are broken, you can bank insane amounts of time from a ko, you can end up with a lot of useless time fighting a last half point ko, actually having to allocate part of your thinking to managing your clock can be more or less stressful than playing once per e.g. 30s depending on the person.
Hey, you bring many good points that are definitely worth discussing. Let me address them one by one.
Comparing the Normal mode directly to other time settings may not be entirely fair. Normal mode is prominently featured on the main play page, making it the go-to choice for all of the new players. This inherently skews its popularity.
But you’re absolutely right, it is still the most popular choice. That’s precisely why I am not suggesting its removal but rather advocating for a complete revamp of the failing Blitz mode and the addition of a new Rapid mode to better cater to player preferences.
If I was OGS, I would just make a trial month, gather all the data and it would provide clearer insights into how these changes play out.
It’s evident from the data that the current default Blitz setting is failing.
Hardcore blitz players find 5 minutes of main time too much, and would prefer a setting that reflects the true essence of Blitz, closer to 1m+10s.
General Players who are looking for a less intense, yet faster alternative to Normal mode, do not appreciate the pressure of a 10-second Byo-Yomi, it’s too extreme.
The oddity of the current 5m+10s settings is that it somewhat converges to neither preference, making both groups unsatisfied. This is reflected in the popularity of the alternative Fischer settings despite their obscurity among casual users. Most regular people don’t even know that you can change Time Control for automatch and yet Fischer Blitz somehow is still a more popular choice.
I have all the time data at my hand, so if you are interested in some other data insight, ask freely
Regarding custom requests with different Time Controls, they are indeed far less popular compared to Byo-Yomi custom requests. Here are the most popular figures to highlight this:
19x19 Custom Request (Fischer; 10m+30s, up to 600s): 667 games (14 hours/day)
9x9 Custom Request (Fischer; 1m+10s, up to 60s): 1,387 games (9 hours/day)
19x19 Custom Request (Simple; 60s/move): 615 games (8 hours/day)
9x9 Custom Request (Simple; 30s/move): 2,339 games (8 hours/day)
19x19 Custom Request (Fischer; 1m+10s, up to 60s): 579 games (7 hours/day)
19x19 Custom Request (Fischer; 1m+30s, up to 60s): 457 games (6 hours/day)
9x9 Custom Request (Simple; 60s/move): 779 games (6 hours/day)
For comparison, Byo-Yomi’s most popular Custom Request is nearly 15 times more popular than the most popular Fischer custom request, showing a clear preference for Byo-Yomi for manual game creation.
A very small issue I’d have with making the blitz setting longer is how it would affect auto-generated tournaments. (At least I assume these would change, too.) The blitz tournaments are fine with 10s byo-yomis, I think.
These negatives are more like positives lol. I think playing a move every 30s is more stressful than having more time on your hands especially in very tense situations.
Not about the longer time settings but about how much time you have on your hands to allocate freely.
Just compare having 30s to play for each move and having 300s to play 10 moves. Some moves don’t need 30s while some moves need more than 30s. Total amount of time is the same.
But what can happen in fischer is that a) the increment is usually less like 15 or 20s than the 30s of byo-yomi to keep the average game times similar, and then b) if you really do use up your time down to the increment you’re now stuck in effectively a much smaller byo-yomi scenario “living off the increment” until you can play a forced sequence fast enough or a ko to bank back up to a comfortable amount. That can more stressful than byo-yomi I find personally.
That’s true and I would say that’s a situation you should avoid because it would be no different from standard byoyomi. However, the good thing about Fischer is that you can get out of it, whereas for byoyomi you are stuck in that situation forever.
I guess it depends on whether one is considering also simplifying the automatch down to one option per time control like Blitz (Fischer), Rapid (?), Live (Byo-yomi) etc.
If we are working to reduce automatch pain, I think it would also be a good idea to look at combining the auto match into the custom games.
Have auto match games be posted on the custom game list under the appropriate settings. Then if someone is only looking at the custom game list due to previously mentioned issues they will see it pop up.
Thanks for the detailed analysis @AfricanRhino17, and thanks everyone that chimed in here, it’s apparent this is quite the popular idea.
I ran some numbers to double check and they of course agree with what you are seeing. For the curious here’s what I’m seeing for 2024 so far in terms of the top time control options for non blitz “live” games
19x19
The custom default is the speed settings that are default when you create a custom challenge. Automatch default is what’s used when you click the “Normal” button to auto-match.
13x13
9x9
So it certainly seems like 5m+?x30s has some very significant pull towards it, biasing towards 5m+5x30s but 5m+3x30s has a good following too. For 13x13, 5m+5x30s was the only significant custom setting for awhile down the list, and given that it’s closer to the default setting of 10m+5x30s, one could imagine that people would be less inclined to bother setting up a custom game to shed that extra 5m. For 9x9, basically relatively few people bother to customize that, which seems to jive with folks being reasonably happy with 5m+5x30s. Further the fact that the automatch to custom ratio is higher for 9x9 might indicate that folks are overall happier with that setting in general and don’t feel the need to bother with a custom game. (Though that’s partially conflated with 9x9 being the default size so with absolutely no customization or clicking anything extra, that’s the game that you’re going to get. I suspect some of both.)
All of this, and the feedback in this forum seem to indicate to me that 5m+5x30s is the sweet spot for timing and we need to better support that as an option.
Honestly it seems to me like we should definitely make this an automatch option and the default for the custom challenge. With that it seems like it’d make sense for Blitz to be changed to 1m+3x10s or something along those lines while we’re at it.
The real question in my mind is if we keep the longer 20m+5x30s as an automatch option or not.
The second question is if we keep fischer as a time control option. It does have some popularity, sitting about half way between 5m+5x30s and 5m+3x30s, but given the overall general bias towards Byo-Yomi it seems like maybe it’d be better to just go with Byo-Yomi for automatch and keep things simple and the challenge pools less fragmented. Canadian should be scrapped as an automatch option it’d seem.
Regarding a new Rapid automatch button, should we add a new Radid option or replace our “Normal” with 5m+5x30s to avoid splitting the challenge pool?
Split the challenge pool and add a “Rapid” 5m+30s or 5m+3x30s option, keeping the 20m+5x30s option
Replace the 20m+5x30s “Normal” auto-match with 5m+5x30s, making it the new “Normal” (or renaming it to “Rapid” or something else)
0voters
Regarding the Fischer automatch time control option for live (blitz/rapid/normal) games, should we remove Fischer as an option and always use Byo-Yomi? (Note: Fischer will be kept as an option, or perhaps the only option, for correspondence games.)
Remove the Automatch Fischer time control option, simplifying the system and reducing the number of challenge pools
Keep the Fischer automatch time control option as is
I agree Fischer is objectively best, but people are allowed to be wrong and prefer inferior timing, and I think it is better for the site to try to maximise automatch pool sizes.
