GTB spoilage

[Hoshotakamoto]

Is the horse out of the barn / should we just go ahead and openly discuss full banzuke predictions without an embargo going forward?

There was a micro-discussion of this in the promotions/demotions thread this morning, and if we develop consensus I would like to move forward with the new paradigm.

However if there is enough pushback then I will continue to post "dumb" picks post-day 15, which is not my first choice. I would rather dissect the full process in the open and then let everyone make their own deviations from whichever process they believe the most in order to submit their picks.

My hope in starting this thread is not to generate a dialogue that leads to inconclusive results. I hope we'll have a policy decision emerge here...

[Suwihuto]

I feel that people posting their own guesses makes it much too easy for everyone to copy the obvious positions across and for them not to make the silly mistakes that creep in with ordering in sanyaku for example.

This would have the effect in theory of moving a lot of the lower scores up to the middle, and it puts more emphasis on those skillful enough to spot where the committee deviates from the obvious to get themselves to the top. In putting that statement in writing, I can see that this is unlikely to make a great difference at the top of the standings unless a good player is generous enough to post their theory; however, would a discussion lead to a crowdsourced consensus eliminating a portion of the skill and understanding required, not to mention the work involved?

[Hoshotakamoto]

To add more context, @Sumo Spiffy is a good player and I don't have to dig into my skillset to watch his video https://www.youtube.com/watch?v=mRZiA9wVEjg so that's part of what I mean when I say is the horse out of the barn. I don't expect my picks to be a carbon copy of his and I would just as soon mull over the topic in public rather than keep my cards close to the vest at this point, especially when the main variable is what flavor of sake the committee is drinking on the day in question.

I hope whatever consensus we reach will apply to future basho and not just be a one-time "well there seem to be a lot of coin flips THIS month, so go ahead FOR NOW" compromise.

[Reonito]

Plenty of public guesses out there. I stopped posting my full guesses on Tachiai before the deadline a while ago, but I still post on the general logic and tricky decisions to be made. Ryan at GSB also stopped releasing the banzuke podcast early, but again, that's a full thought-through guess.

[Bunbukuchagama]

Spiffy already has a video up with a rough banzuke draft. And he does it every time. 

[Tim Sumo]

I think having peoples guesses stay as secret as possible until deadline would make the game more fun. Spiffy isn’t number 1 or anything so if you want to go copy them it’s at your own risk and I doubt many are doing that anyway. 

[Ganzohnesushi]

41 minutes ago, Tim Sumo said:

I think having peoples guesses stay as secret as possible until deadline would make the game more fun. Spiffy isn’t number 1 or anything so if you want to go copy them it’s at your own risk and I doubt many are doing that anyway. 

Why should I copy one's guesses when I have the right to make my own mistakes.

Ganzohnesushi

[Jakusotsu]

Now I feel like invoking my right to copy other ones' mistakes...

[Bunbukuchagama]

54 minutes ago, Tim Sumo said:

I think having peoples guesses stay as secret as possible until deadline would make the game more fun. Spiffy isn’t number 1 or anything so if you want to go copy them it’s at your own risk and I doubt many are doing that anyway. 

Nodody is discussing outright copying - although even of everybody's submissions were to become magically available to someone, the question "Who to copy?" would remain unanswered.

But obvious mistakes would become rare - and the game "floor" would rise. We are trying to determine if it is a bad thing.

[Hoshotakamoto]

I just feel slightly hamstrung for doing a thing in public up to 14 days in a row and then doing a severely brain damaged version of it on the 15th day to avoid the appearance of impropriety.

EDIT: I also question to some extent whether or not it damages my ability to be taken seriously in discussions about the juryo line etc when you look at my picks to cross the maegashira line and they are total garbage.

Edited January 29, 2024 by Hoshotakamoto

[Tsuchinoninjin]

Not to divulge a nonsecret, but in interest of transparency you can just search google for 予想番付 and get back ample public predictions. I never look at them myself really, the one time I checked two of them and they would have gotten an mk score, that was a while ago.

[Reonito]

2 minutes ago, Hoshotakamoto said:

I just feel slightly hamstrung for doing a thing in public up to 14 days in a row and then doing a severely brain damaged version of it on the 15th day to avoid the appearance of impropriety.

EDIT: I also question to some extent whether or not it damages my ability to be taken seriously in discussions about the juryo line etc when you look at my picks to cross the maegashira line and they are total garbage.

I mean, we have a whole topic to discuss division exchanges and the like, so it's not like the forum is dedicated to silence on any questions that might affect GTB. 

[Tsuchinoninjin]

As an addendum to the above post I have never manually tabulated my score correctly so please take my judgement of other banzukes with a grain of salt

[Tetsuba]

I never publish my GTB submission because I always do it at the last time 

[Bunbukuchagama]

5 minutes ago, Tetsuba said:

I never publish my GTB submission because I always do it at the last time 

Mine is done weeks before deadline, but I spend those weeks agonizing over a couple of choices. 

[Sumo Spiffy]

5 hours ago, Tim Sumo said:

I think having peoples guesses stay as secret as possible until deadline would make the game more fun. Spiffy isn’t number 1 or anything so if you want to go copy them it’s at your own risk and I doubt many are doing that anyway. 

hey hey hey it's not that risky

[Sumo Spiffy]

FWIW, I used to post my actual guess just after the GTB deadline, but at that point nobody really cared because they all wanted to talk about their own. So, now I usually post a few days before the deadline. I'm pretty sure it hasn't done much to affect scoring.

The instant banzuke after the basho comes with the permanent disclaimer that it's a draft and almost assuredly wrong in some clear way or another that I'm overlooking in the moment. Yeah, it's designed to be helpful in the basics, but I'd rather educate than maximize my competitiveness.

[Reonito]

2 minutes ago, Sumo Spiffy said:

Yeah, it's designed to be helpful in the basics, but I'd rather educate than maximize my competitiveness.

There are many more people who are interested in following projected banzuke movements than there are GTB players...

[Hoshotakamoto]

I'm just gonna reverse my long-standing policy and go deep on GTB analysis in the math thread exploring every outcome in the sumo multiverse starting in about 24 hours if no one protests before then.

 

EDIT: it's been 36 hours, forging ahead.

Edited January 31, 2024 by Hoshotakamoto

[Sumo Spiffy]

3 minutes ago, Hoshotakamoto said:

I'm just gonna reverse my long-standing policy and go deep on GTB analysis in the math thread exploring every outcome in the sumo multiverse starting in about 24 hours if no one protests before then.

Really, the deeper you go, the less of a problem it should be. Everyone only gets one entry, so even if a lot of people look at what you write and want to make use of it, they still have to choose from the various ideas.

[Wakawakawaka]

Interesting observation, just checked the Kyushu23 GTB entries. There were 194, all were unique.

[Seiyashi]

4 hours ago, Wakawakawaka said:

Interesting observation, just checked the Kyushu23 GTB entries. There were 194, all were unique.

Considering that in theory there are something like 42! ways to arrange a banzuke (although much less if you have any modicum of sense), it's not entirely surprising.

[Jakusotsu]

23 minutes ago, Seiyashi said:

Considering that in theory there are something like 42! ways to arrange a banzuke (although much less if you have any modicum of sense), it's not entirely surprising.

I still think it is.

[Sokkenaiyama]

On 30/01/2024 at 01:42, Tetsuba said:

I never publish my GTB submission because I always do it at the last time 

I've been out of the gaming business for years now, but this is so damn accurate. 

[Athenayama]

11 minutes ago, Sokkenaiyama said:

I've been out of the gaming business for years now

(Off topic of this thread) And why don't you just come back?