GTB invite- Aki 2023 - 202 entries - RESULTS!!

[Ganzohnesushi 522]

57, neither fish nor fowl. Expecting #80 among 203.

Ganzohnesushi

[Sumo Spiffy 569]

60, but it could have been better. I had guys placed in what turned out to be the correct spots for reasons I decided weren't good enough, but a couple of these placements were... very wtf.

[Asashosakari 19,500]

61, even though plenty of my choices were miles away.

[Reonito 1,477]

I got 38/46 points in the first 23 ranks, 20/22 in the last 11, but 1/16 in the middle 8 

[Suwihuto 127]

M9-M11 killed me. Some good areas too, 52 I think, pretty good for me.

[Gurowake 4,088]

I'm counting 63.  84 less 14 for 7 whiffs less 7 for 7 hits.  Mostly off with Ryuden, Shonannoumi, Mitakeumi, and Midorifuji.  They treated the former 2 much better than I expected, and the latter 2 much worse.  If I had noticed I didn't demote Oho at all, I probably would have swapped him with Mitakeumi, ending up with the latter even further out of position, but being right with Oho.

 

Edited August 27, 2023 by Gurowake

[#1HENKAFAN 40]

55 points by my count. I think my first draft was actually closer than my final draft this time so that sucks.

[Hoshotakamoto 259]

On 22/08/2023 at 12:25, Sumo Spiffy said:

When discussing any of this, in terms of precedent, it's worth bearing in mind that the committee has decidedly taken a turn in their approach from the start of this year on. There's only been a few banzukes this year, so there's hardly a scientific sample; however, some of the changes have been consistent enough that looking much further back for ideas to lean on is liable to lead one astray.

This flashed through my head as I was reading how things panned out for Mitakeumi and Midorifuji. I'm in the process of trying to determine how my "heavily influenced by past decisions" (official) banzuke fared vs a hypothetical "naive math / don't overdemote" banzuke.

[Hoshotakamoto 259]

2 hours ago, Ganzohnesushi said:

57, neither fish nor fowl. Expecting #80 among 203.

Ganzohnesushi

I think I have the same score as you and I expect closer to 50-60th.

[Reonito 1,477]

16 minutes ago, Hoshotakamoto said:

I think I have the same score as you and I expect closer to 50-60th.

I think you'll both do better than that.

[Asashosakari 19,500]

I dunno, the top and bottom parts really were quite predictable. I wouldn't be surprised if the median score is close to 50, and that'll just be ~100th place then.

We haven't seen anyone come forward with such a score yet, but I also wouldn't be shocked if the winning result is a 72 or something like that. There are bound to be a fair number of people who correctly predicted that the luck boundary would fall between Shonannoumi/Ryuden on the one side and Hakuoho/Endo on the other, and from there it didn't take that many correct decisions to get pretty close to the real thing. Oddly enough, the big wildcard seems to be Kotoeko. If you got him right, a lot of the MKs fall into place.

[Sumo Spiffy 569]

Kotoeko seems like one of those picks where the people who got it right may well have done it for reasons that skew the rest of the banzuke, so they end up not doing that well.

From Y-M3W (15 guys), the only one that wasn't glaringly obvious was Abi/Asanoyama, and from M12-M17E (11 more), the only screwy outcome was Atami/Chiyoshoma. My guess is most people who had the right idea got Atami/Chiyo switched, so if they got all the rest of those correct, that's 50 points right there. (I swapped Abi and Asanoyama, so that's 48.) Ura and Takayasu should also be gimmes. Then the question becomes how many different variations of the remaining guys do we see?

I'm picking 68 as the winning score.

[Koorifuu 1,003]

I missed out on a few obvious bullseye points, it seems. Ah well. Meisei/Hokutofuji, Aoiyama/Kotoshoho, Chiyoshoma/Atamifuji, Nishikifuji/Myogiryu... all hits.

The only one that feels like an obvious mistake to me is Aoiyama/Kotoshoho, the others require a feel for banzuke nuance that I just can't comprehend (for now, at least).

Still, my main mistake was overestimating the air resistance effect on Mitakeumi's and Midorifuji's falls. Some of the pieces between their actual placement & where I put them fell in place anyway, but not many.

Edited August 28, 2023 by Koorifuu

[Yarimotsu 588]

51 I think. Mostly unhappy with the banzuke committee's logic rather than my own. I don't see how we go from prioritising full demotions for those not in the JOI to this, but we do. A couple of the overpromotions and overdemotions seem egregious based on my idea of how conservative they'd be - but then there's a clear underdemotion I see as well. It starts to look less skewed if you decide you hate hakuoho, maybe I favoured him.

Edited August 28, 2023 by Yarimotsu

[Bunbukuchagama 739]

2 hours ago, Asashosakari said:

I dunno, the top and bottom parts really were quite predictable. I wouldn't be surprised if the median score is close to 50, and that'll just be ~100th place then.

We haven't seen anyone come forward with such a score yet, but I also wouldn't be shocked if the winning result is a 72 or something like that. There are bound to be a fair number of people who correctly predicted that the luck boundary would fall between Shonannoumi/Ryuden on the one side and Hakuoho/Endo on the other, and from there it didn't take that many correct decisions to get pretty close to the real thing. Oddly enough, the big wildcard seems to be Kotoeko. If you got him right, a lot of the MKs fall into place.

70 here, if my math hasn't failed me. Still, I missed Takanosho at M4, Midorifuji and Mitakeumi placements (and the resulting mess), Atamifuji over Chiyoshoma...

Edited August 28, 2023 by Bunbukuchagama

[Wakawakawaka 157]

And the winner for guess this banzuke's headscratcher = Kinbozan staying at M10e?

62 points here. The good news is my last minute modifications changed my score by exactly 0 points.

Edited August 28, 2023 by Wakawakawaka

[Kintamayama 45,345]

And here are the results. A landslide win for Bunbukuchagama, a good five point lead over second place total newcomer Purichu Ruyaru. 202 is the correct number of participants, a new record by far. 36 new players, 15 of them kachikoshi, which is very nice. 

Milestones:

Kaiowaka- 110th basho

Oskanohana- 60th basho

Shatsume- 40th basho

Suwihuto- 30th basho

Wakatake- 30th basho

Hidenotora - 20th basho

Mr. Rofl - 20th basho

Kobashi- 20th basho

Akakyoryu- 20th basho

Kaito- 20th basho

Thank you for playing and a huge thanks to Andoreasu and Doitsuyama for running the show. 202 players is awesome.

And now- on with the ba-show!!

To see how you did, go here

 

[Jakusotsu 5,995]

Kudos to everybody anticipating the Ryuden-Endo gap.

[Ganzohnesushi 522]

1 hour ago, Jakusotsu said:

Kudos to everybody anticipating the Ryuden-Endo gap.

I had Ryuden at M6e (even half a rank higher) and Endo at M10w, that's 3 points for GTB. But unfortunately I made too many other misjudgements, thus the count stopped at 57.

Ganzohnesushi

[Koorifuu 1,003]

There it is, first KK since Hatsu '13 

[Hoshotakamoto 259]

20 hours ago, Reonito said:

I think you'll both do better than that.

I'm afraid to even ask where I surrendered 4 points that I've forgotten about (53 instead of 57?) but as for the 57 hypothetical placement I think I hit a meta-GTB bullseye. 

[Hoshotakamoto 259]

On 22/08/2023 at 09:03, Bunbukuchagama said:

I would expect both of them getting joi parachutes and landing higher. Where exactly? Well, the answer to this question will decide who wins this game. 

Narrator: it did not decide who won the game.

[Bunbukuchagama 739]

Just now, Hoshotakamoto said:

Narrator: it did not decide who won the game.

Indeed, it didn't. But I missed them very conveniently, in a way that didn't disturb many picks around them. 

[Tigerboy1966 1,451]

I got just about every call wrong but somehow escaped with an 8-7, and cling on to my place in maku'uchi (38th).

[Hoshotakamoto 259]

22 hours ago, Hoshotakamoto said:

This flashed through my head as I was reading how things panned out for Mitakeumi and Midorifuji. I'm in the process of trying to determine how my "heavily influenced by past decisions" (official) banzuke fared vs a hypothetical "naive math / don't overdemote" banzuke.

modulo the fact that I apparently can't count up my official score (a 53 - I saw a 57) it seems that a relatively naive math-the-banzuke effort might have yielded 65 points (gonoyama / takanosho wrong for -4, abi switched vs asanoyama for -2, hokuseiho and a maximally demoted mitakeumi switched for -2, chiyoshoma/atamifuji for -2, endo/kinbozan swapped for -2, ryuden, midorifuji, and kotoeko on the wrong line for -6, hiradoumi on the wrong side for -1) ... so losing 8 to 12 points by staying up all night researching decisions the committee made in the past was probably not a winning proposition.