Invitation to Play GTB - May 2025 - 316 entries - RESULTS!!

[Asashosakari 19,943]

Just now, Reonito said:

This is what I entered.

Thanks for confirming; I was slightly confused before because...

17 minutes ago, Reonito said:

reorders Oshoma-Ura-Tobizaru (5 points)

...if I'm not mistaken that's just 4 points of difference to Gurowake's version. 

 

[Reonito 1,613]

12 minutes ago, Asashosakari said:

...if I'm not mistaken that's just 4 points of difference to Gurowake's version. 

D'oh! I mentally moved Oshoma ahead of the other two rather than swapping him with Tobizaru. My logic was that Oshoma computes half a rank ahead of the other two, and last time they went with the numbers in most such instances, and I believe there was a recent case in which a 7-8 and a 6-9 were tied, and they opted to drop the 7-8 one rank and over-demote the 6-9 by half a rank. Hoping for consistency 

[Sumo Spiffy 621]

I ended up going with the -3.5 for Kinbo and Atami, knowing it's much more of a coin flip with -4 than I would have ever imagined possible, but -4.5... if that happens, we'll have to rethink some core assumptions a lot of us have about their decision-making.

2 hours ago, Reonito said:

D'oh! I mentally moved Oshoma ahead of the other two rather than swapping him with Tobizaru. My logic was that Oshoma computes half a rank ahead of the other two, and last time they went with the numbers in most such instances, and I believe there was a recent case in which a 7-8 and a 6-9 were tied, and they opted to drop the 7-8 one rank and over-demote the 6-9 by half a rank. Hoping for consistency 

Oho and Ura, one year ago. Oho got the regular demotion from 3e to 4e, and Ura went from 1e to 4w.

Edited April 22 by Sumo Spiffy

[Asashosakari 19,943]

Some preview-type remarks...

3 hours ago, Wamahada said:

Line is dead - again more than 300 guesses

316 altogether.

524 players are on the GTB ranking ahead of Natsu basho. By recent participation status:

                    ranked  playing  kyujo
Active last basho:     359      235    124
Kosho  last basho:      23        4     19
Multiple absences:     142       12    130

The kyujo players include last basho's yusho winner Isokiho, one of 14 Natsu absences from the top 100. 

65 unranked players have entered, including - as far as I can tell - 62 new players plus three returnees: boosiv, Ivar (unless it's a coincidentally identical name?), and Ulsimitsuki (albeit seemingly misspelled as Ulsimituski).

In exchange for these incoming players, 20 are set to fall off the rankings by losing their last countable score.

The provisional top 20 after removing the scores from a year ago that will be replaced by the Natsu results (actual pre-Natsu positions and scores in brackets, including all remaining top 20 players):

 1 Asset           4.471  ( 1 / 5.686)
 2 Asashosakari    4.445  ( 2 / 5.599)     2
 3 Bunbukuchagama  4.363  ( 3 / 5.574)     6
 4 CaptSmarta$$    4.332  ( 5 / 5.481)     7
 5 Kale            4.326  ( 4 / 5.509)     8
 6 Goshobiyama     4.324  ( 6 / 5.479)     8
 7 Cheesofuji      4.314  ( 7 / 5.462)     8
 8 Duden           4.304  (14 / 5.370)     9
 9 Hisui           4.304  (11 / 5.394)     9
10 Toshishugisha   4.278  (13 / 5.386)    10
11 Akakyoryu       4.255  (21 / 5.314)    11
12 Kaito           4.246  (16 / 5.353)    12
13 Kachikoshi      4.240  (10 / 5.414)    12
14 Tyerenex        4.236  (23 / 5.300)    12
15 Kobashi         4.208  ( 8 / 5.442)    14
16 Pepenohana      4.197  (25 / 5.268)    14
17 TochiYESshin    4.188  (19 / 5.326)    15
18 Sugakusha       4.187  ( 9 / 5.418)    15
19 Beeftank        4.185  (12 / 5.387)    15
20 Zelerny         4.171  (26 / 5.249)    15
...
21 Reonito         4.166  (17 / 5.342)    16
23 Papayasu        4.135  (18 / 5.328)    17
25 Purichu Ruyaru  4.118  (15 / 5.361)    18
26 Chiyonosawa     4.115  (20 / 5.325)    18

Players with a better provisional than actual ranking are getting to drop a relatively weak result from their totals, players who are worse provisionally are defending a relatively strong score. The final column shows how many approximate game points each player has to make up on the current #1 in order to take over the top spot after Natsu. (Based on an assumed mean score of 50 points = 0.020 ranking points per game point, so these are just estimates.)

12 players who are currently carrying a single zero on their ranking tally are set to replace it with a countable score and will see their ranking position improve significantly (current rank > provisional rank):

Kotokishi (106 > 43)
Hashibamiyama (108 > 46)
Barutonoumi (111 > 49)
Wakatakasake (128 > 81)
Hobbobgob (133 > 97)
Ardonite (132 > 99)
matrix (139 > 114)
AoiyamaFan (146 > 118)
asanosanpo (163 > 136)
Modra (170 > 146)
Kcramsib (176 > 157)
kostuek (183 > 165)

On the other hand, 3 players who currently have the full six countable scores are set to pick up a zero by missing both Haru and Natsu (current rank > approximate next rank...the provisional rank makes no sense for these players):

Krakel (41 > 140)
Miwabe (69 > 163)
Kotononami (129 > 184)

[Asashosakari 19,943]

BTW, the currently highest-ranked player with a zero score is Schwoosh in 102nd place; the zero is for their absence in Aki 2024. In fact, if they had played that edition and done decently well (at least 56 points), they would have become the #1 ranked player one basho later! Alternatively, 60 points in Nagoya would also have done it (with Aki then scored as kosho rather than zero).

[Wakawakawaka 186]

Can always count on Asashosakari for stats you didn't even know you wanted :)

Here's my mild contribution to further that spirit, current 10+ basho KK streaks. Consecutive KK record all time is 29 held by both Asashosakari and Okisuruga.

Goshobiyama    21
Cheesofuji    20
Bunbukuchagama    14
Reonito    13
TochiYESshin    13
Asset    13
Sugakusha    13
AAAAAAAAAAAAAAAAAAAAAA    13
Beeftank    12
Kale    11
Kachikoshi    10

For some reason pasting this as a table blacklisted my IP

Edited April 22 by Wakawakawaka

[Reonito 1,613]

6 hours ago, Asashosakari said:

BTW, the currently highest-ranked player with a zero score is Schwoosh in 102nd place; the zero is for their absence in Aki 2024. In fact, if they had played that edition and done decently well (at least 56 points), they would have become the #1 ranked player one basho later! Alternatively, 60 points in Nagoya would also have done it (with Aki then scored as kosho rather than zero).

The archives are actually up to date!

[Reonito 1,613]

3 minutes ago, Wakawakawaka said:

Here's my mild contribution to further that spirit, current 10+ basho KK streaks.

Before my current streak, I had one of 20 basho (with two kosho), with my disastrous Hatsu 2023 (when I got the komusubi count off by two) in between.

[Asashosakari 19,943]

17 hours ago, Asashosakari said:

11 Akakyoryu       4.255  (21 / 5.314)    11
14 Tyerenex        4.236  (23 / 5.300)    12
16 Pepenohana      4.197  (25 / 5.268)    14
20 Zelerny         4.171  (26 / 5.249)    15
...
21 Reonito         4.166  (17 / 5.342)    16
23 Papayasu        4.135  (18 / 5.328)    17
25 Purichu Ruyaru  4.118  (15 / 5.361)    18
26 Chiyonosawa     4.115  (20 / 5.325)    18

Huh...I only noticed just now that, purely by coincidence, the four players who have provisionally moved up into the top 20 are coming from the exact ranks that the four displaced players have fallen to.

Even more oddly, 22nd / 24th / 27th place have done this:

22 ChewieKen       4.163  (27 / 5.225)    16
24 Kishikaisei     4.120  (22 / 5.311)    18
27 Gurowake        4.113  (24 / 5.285)    18

Consequently, the composition of the top 27 is entirely unchanged between the current and provisional rankings.

[Reonito 1,613]

FWIW this is the consensus banzuke from 12 guesses over on the GSB discord, compiled by @Whatta bout Sumo. There was 100% agreement on 13 ranks, plus another 12 with a single discordant guess, and 3 more with two (for a total of 28 ranks with good agreement). The rikishi with the least agreement were Ura and the bottom three, followed closely by Kinbozan.

Edited April 22 by Reonito

[Kintamayama 45,974]

3 hours ago, Reonito said:

The archives are actually up to date!

Yes, for a couple of months now.

[Bunbukuchagama 827]

3 hours ago, Reonito said:

FWIW this is the consensus banzuke from 12 guesses over on the GSB discord, compiled by @Whatta bout Sumo. There was 100% agreement on 13 ranks, plus another 12 with a single discordant guess, and 3 more with two (for a total of 28 ranks with good agreement). The rikishi with the least agreement were Ura and the bottom three, followed closely by Kinbozan.

This looks remarkably close to my submission; just swap Tamawashi and Abi, Meisei and Kinbozan, and Shishi and Endo. 

[Reonito 1,613]

8 minutes ago, Bunbukuchagama said:

This looks remarkably close to my submission; just swap Tamawashi and Abi, Meisei and Kinbozan, and Shishi and Endo. 

Interestingly, none of the individual guesses matched the consensus (or each other) exactly. A couple were one swap away; mine swaps Ura/Tobizaru and Meisei/Kinbozan.

[Bunbukuchagama 827]

3 minutes ago, Reonito said:

Interestingly, none of the individual guesses matched the consensus (or each other) exactly. A couple were one swap away; mine swaps Ura/Tobizaru and Meisei/Kinbozan.

Nobody is leading the hive mind, it's a true collective. 

 

[Kachikoshi 45]

3 minutes ago, Reonito said:

Interestingly, none of the individual guesses matched the consensus (or each other) exactly. A couple were one swap away; mine swaps Ura/Tobizaru and Meisei/Kinbozan.

In that case you and I have exactly the same guess. And it looks like I sent in my pick ever so slightly before you :)

[Reonito 1,613]

12 minutes ago, Kachikoshi said:

In that case you and I have exactly the same guess. And it looks like I sent in my pick ever so slightly before you :)

I'd settle for losing the yusho on a time of submission tiebreaker 

https://tenor.com/bhEjn.gif

Edited April 22 by Reonito

[Asashosakari 19,943]

Asset posted their guess on Reddit.

That M16w-M18e solution is certainly not the consensus, so that could make or break their #1 status.

[Gurowake 4,123]

4 hours ago, Reonito said:

The rikishi with the least agreement were Ura and the bottom three, followed closely by Kinbozan.

How would there be more agreement on Meisei than Kinbozan?  Meisei can't be moved one spot further up and Kinbozan can't be moved one spot further down.  Who else are they putting in Kinbozan's spot other than Meisei?

 

[Bunbukuchagama 827]

2 minutes ago, Gurowake said:

How would there be more agreement on Meisei than Kinbozan?  Meisei can't be moved one spot further up and Kinbozan can't be moved one spot further down.  Who else are they putting in Kinbozan's spot other than Meisei?

 

Midorifuji or Aonishiki? I think Spiffy was pushing the "they never overdemote by a full rank" narrative. 

[Asashosakari 19,943]

2 minutes ago, Gurowake said:

How would there be more agreement on Meisei than Kinbozan?  Meisei can't be moved one spot further up and Kinbozan can't be moved one spot further down.  Who else are they putting in Kinbozan's spot other than Meisei?

Putting Kinbozan as high as M7w isn't out of place when somebody has also been lenient with Tobizaru and Ura.

[Reonito 1,613]

11 minutes ago, Gurowake said:

How would there be more agreement on Meisei than Kinbozan?  Meisei can't be moved one spot further up and Kinbozan can't be moved one spot further down.  Who else are they putting in Kinbozan's spot other than Meisei?

 

7 of 12 guesses had Kinbozan at 10e, 8 of 12 had Meisei at 9w. The other 4 Meisei guesses were all 10e as you'd expect. 3 of the other 5 Kinbozan guesses were the expected 9w flip with Meisei. One had him at 9e (ahead of both Midorifuji and Meisei), and one at 11w (???!!!).

Edited April 22 by Reonito

[Reonito 1,613]

5 minutes ago, Bunbukuchagama said:

Midorifuji or Aonishiki? I think Spiffy was pushing the "they never overdemote by a full rank" narrative. 

It happens, if not very often, though you'd have to go further back in time to find an overdemotion by more than a full rank (this is just for M5w 6-9, the db is not well-suited to the more general question without brute-forcing it).

[Bunbukuchagama 827]

15 minutes ago, Reonito said:

It happens, if not very often, though you'd have to go further back in time to find an overdemotion by more than a full rank (this is just for M5w 6-9, the db is not well-suited to the more general question without brute-forcing it).

There is nothing special about M5w 6-9 (with a non-joi schedule). If you extend the search over several ranks down, you will find plenty of 1.5 rank overdemotions and even an occasional 2 ranks. 

[Reonito 1,613]

55 minutes ago, Bunbukuchagama said:

There is nothing special about M5w 6-9 (with a non-joi schedule). If you extend the search over several ranks down, you will find plenty of 1.5 rank overdemotions and even an occasional 2 ranks. 

Yeah I agree; they've been pretty inconsistent about whether they give any joi-adjacent preference to 5w, so we'll see what happens. Had he been 6w, I'm pretty sure I would have swapped him with Meisei in my guess.

[Whatta bout Sumo 14]

9 hours ago, Gurowake said:

How would there be more agreement on Meisei than Kinbozan?  Meisei can't be moved one spot further up and Kinbozan can't be moved one spot further down.  Who else are they putting in Kinbozan's spot other than Meisei?

 

We got one more person from our guessing community to submit to the average Banzuke, so there are 13 now, but it did not change the average Banzuke between us. Here is the full compilation:

Edited April 23 by Whatta bout Sumo