Basho Preparations - 2014 Natsu

[Yubinhaad 11,606]

Some pictures showing the preparation of the dohyo in Dewanoumi-beya's keikoba, which took place on April 15th and 16th.

The dohyo in each heya's keikoba are prepared before the banzuke is released. The work is done by a handful of yobidashi with help from some rikishi of the heya. First, the old tawara are removed and the surface broken up.

Not every heya uses tawara in their keikoba's dohyo, but Dewanoumi-beya does. Only twenty are needed.

The surface is hammered flat with the tako and tataki tools.

Yobidashi Hiromasa and Soichi mark out the dimensions of the dohyo.

The tawara are inserted into the grooves. Dewanoumi chooses to include the tokudawara in their keikoba dohyo.

Soichi places one of the tokudawara, and then steps back to check his work.

Hiromasa adds another tokudawara.

The 15-year-old isn't experienced in this yet and didn't get it quite right, so his senior colleagues help out.

Beer bottles are used to give the tawara the correct curve before being placed in the groove.

Only about a third of the bale is visible; the rest is hidden beneath the surface.

Hiromasa uses a hoe to make some fine adjustments to the groove for this tokudawara.

One of the heya's rikishi helping to secure the tawara.

Final hammering of the surface.

Finally, the heya's own yobidashi Yohei marks out the shikiri-sen and the square hole for the dohyo-matsuri ceremony.

Random shot of the old-school scales at Dewanoumi-beya.

And now some pictures of the dohyo-matsuri ceremony, which took place on April 17th.

Some of the shizumemono which will be wrapped up together and placed in the centre of the dohyo.

Kimura Shotaro, the senior gyoji present for this ceremony, holds the norito sojo, a shinto prayer which will be read out during the ceremony.

The other two participants are Kimura Akijiro and the heya's own gyoji Kimura Chishu. Here Akijiro performs the oharai, a purification ceremony, for the members of the heya.

The shizumemono are placed into the dohyo. They are wrapped in hoshogami, a high quality paper made from mulberry wood.

Kimura Shotaro pours a little sake over the tokudawara.

With a final bow, the ceremony is complete and the dohyo is now ready for the rikishi to practice on.

[Yubinhaad 11,606]

The gomenfuda for this basho went up a couple of weeks ago, installed next to the drum tower outside the Kokugikan.

Work has started on the itabanzuke for this basho. The writers this time are Kimura Kankuro and Shikimori Kazuki. It takes three days for the itabanzuke to be completed, and they start by drawing a grid in pencil as a guideline.

Jonokuchi rikishi are written in the smallest text on the itabanzuke. Six Jonokuchi rikishi take up the same space as the diameter of this 10-yen coin.

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

From the right, Makushita 46-50.

Finally, the recent change of organisation status for the Kyokai meant that the signs around the Kokugikan had to be changed as well.

[Kuroyama 715]

One thing I've been curious about: Why would a heya NOT use tawara in their keikoba? Wouldn't you want to make the dohyo they train in as much like the real thing as possible? How can a deshi develop ring sense if there's nothing to mark its boundaries?

[krindel 671]

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

I cannot imagine any other sport not using a computer for that :-).

[Orion 431]

One thing I've been curious about: Why would a heya NOT use tawara in their keikoba? Wouldn't you want to make the dohyo they train in as much like the real thing as possible? How can a deshi develop ring sense if there's nothing to mark its boundaries?

Using tawara puts the price up, not to mention the man-hours. But I've always thought it was pretty cheapskate just to have a simple slightly-sunken circle.

I was very impressed that they had three gyoji to do the dohyo-matsuri at Dewanoumi. In a heya, it's commonly done by just one. This was just like the real thing that will happen in the Kokugikan itself next Saturday. I suppose this was the first since the new master was really in charge, after his predecessor turned 65 and left the Kyokai.

Orion

[Kuroyama 715]

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

I cannot imagine any other sport not using a computer for that :-).

Major League Baseball uses handwritten lineup cards. But then, that's not calligraphy. And it only has to list 9 players.

On the other hand, MLB probably wouldn't use computers even if it were practical. This is a sport where video review is still HUGELY controversial.

[Orion 431]

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

I cannot imagine any other sport not using a computer for that :-).

But there's the committee of the shimpan oyakata who decide the pairings, plus several gyoji to keep the record, and they are looking at a whole group of names on the list, to set up suitable matches for each day. With a large continuous list on the floor they can get the big picture. After the pairings have been decided, then it's time for one of the gyoji to put it on computer. (Even when computers weren't nearly as common as they are today, the Kyokai was sending selected young and middle-ranking gyoji for computer training.)

Orion

[Naganoyama 5,885]

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

I cannot imagine any other sport not using a computer for that :-).

Major League Baseball uses handwritten lineup cards. But then, that's not calligraphy. And it only has to list 9 players.

On the other hand, MLB probably wouldn't use computers even if it were practical. This is a sport where video review is still HUGELY controversial.

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

I cannot imagine any other sport not using a computer for that :-).

But there's the committee of the shimpan oyakata who decide the pairings, plus several gyoji to keep the record, and they are looking at a whole group of names on the list, to set up suitable matches for each day. With a large continuous list on the floor they can get the big picture. After the pairings have been decided, then it's time for one of the gyoji to put it on computer. (Even when computers weren't nearly as common as they are today, the Kyokai was sending selected young and middle-ranking gyoji for computer training.)

Orion

I think that Orion's view is closer to the mark - the Kyokai is quite advanced when it needs to be - look how long ago they started to use video review compared with quite a lot of other sports.

[Asashosakari 19,096]

Elsewhere, Kimura Satoshi has been writing the names of every rikishi on long rolls of narrow paper. They will be used by the torikumi committee when they are deciding who should fight who during the basho.

I cannot imagine any other sport not using a computer for that :-).

Same here. Especially for the lower divisions where 99% of the matchups are essentially assigned by a fixed system anyway, and it's not a very difficult-to-automate system to begin with. Frankly, I'm amazed that there are so few screw-ups (maybe they do use an automated system to double-check for obvious problems in the intended schedule after it's been input?)...from the last couple of years I vaguely recall one case of jonidan matchups that didn't make any sense, but that's it.

Of course, maybe most screw-ups that are caused by using these paper rolls for record keeping aren't in making the schedules, but rather in making the next banzuke. ;-) (Would explain a few things...) One noteworthy (and pretty major) error caused by these paper transcripts was during the Hatsu 1989 banzuke-making - you'll notice there's a rikishi ranked as Sd101e and an empty spot at Jd44e. They had misread Futagozakura's 7-0 + playoff loss record as a 6-1 and ranked him accordingly. They moved him into that makeshift extra sandanme spot when it was discovered later.

Edit: By the way, one thing I've been wondering and it kind of fits here - can shisho ask the committee that one of their lower-division rikishi be scheduled for a particular day (e.g. "Please put his next match on Day 6, not Day 5") if he has to attend a social function such as a family funeral or the like?

Edited May 5, 2014 by Asashosakari

[Gurowake 4,031]

There would have to be a lot of extra bits in the system to handle all the possible things they take into consideration when determining the match-ups. While *most* is automated, someone has to pick the people who get to visit Juryo, and there's no real obvious pattern beyond the first couple days (and even then, it's not 100% consistent). While the mid-basho days might just be programmable as "random", the last 3 days they tend to have those on the border to promotion to Juryo against Juryo who are on the edge of falling out; while this sort of thing could be programmed, it sounds like a complete nightmare to try to prioritize exactly which match-ups you're most interested in. I suppose it's possible, but it's a much different problem than those typically faced by a computerized Swiss system and already pretty much everyone who wants a program to pair their tournament for them has different requirements that need to be met, necessitating a solution specific to that organization; there's no off-the-shelf software that does exactly what you want, so it's probably not cost efficient.

Plus, Sumo is pretty heavy on tradition. If it's worked this long, why change it?

[Asashosakari 19,096]

There would have to be a lot of extra bits in the system to handle all the possible things they take into consideration when determining the match-ups. While *most* is automated, someone has to pick the people who get to visit Juryo, and there's no real obvious pattern beyond the first couple days (and even then, it's not 100% consistent).

Just needs one simple extra bit: the ability to mark particular rikishi as "do not include" before the automation is set to do its thing. There are really only a small handful of areas which are subject to intervention:

- makushita rikishi sent into juryo (as you mentioned)

- record imbalances, e.g. an odd number of rikishi having a particular record which necessitates pairing across groups; these are generally resolved within the jonokuchi division only

- a few last-round bouts involving some 6-0, 5-1, 1-5 and 0-6 rikishi (when there aren't enough 6-0/0-6 records to match those up directly)

These are well-known and could easily be accounted for in manual fashion beforehand (they already do now!), still leaving basically the entire makushita, sandanme and jonidan schedules to be generated automatically. That's well over 90% of the daily lower division match-making, and that's the lower boundary of what could be automated. I'm not talking about passing the full scheduling responsibility to a computer program (the way some other sports/competitions have done to avoid charges of favouritism), just about automating the proverbial trained-monkey parts of it.

In any case, even if the above three issues didn't exist, some type of manual "do not include" would be required anyway to account for rikishi withdrawing and rejoining the tournament.

Beyond all that, the list of necessary exclusions is fixed and limited to:

- no same-stable matchups (easily checked for as long as the automation has the data available)

- no same-family matchups (would require a simple short exclusion list to be maintained)

- no repeated matchups

That's much less than e.g. the white/black considerations in chess, or the problem of avoiding a meaningless #1-#2 matchup because the winner was determined too early. I reckon I've done more complicated Excel sheets for some of the online sumo games...and that's not because those were particularly difficult to automate, but rather because this here is a pretty simple task as far as I can see.

Edit: Actually, if I recall correctly Doitsuyama did write such a "most of the way" automation a few years ago to help with some data-importing tasks for the Sumo DB...maybe he can tell how complicated it was and what the error rate was.

Edited May 5, 2014 by Asashosakari

[Kuroyama 715]

You wouldn't necessarily use a computer to automate the matches, and I could see considerable resistance to the idea.

But you could use a computer to handle all the "bookkeeping". Suppose the oyakata are looking at all the names on an array of flatpanels, simulating a long paper scroll, and they can arrange the torikumi by dragging them around with a mouse or touchscreen. What the computer could for them then is to perform certain sanity checks, like checking for excluded combinations, asking for confirmation on bouts that seem dubious based on relative rank and record and so on.

[Asashosakari 19,096]

You wouldn't necessarily use a computer to automate the matches, and I could see considerable resistance to the idea.

But you could use a computer to handle all the "bookkeeping". Suppose the oyakata are looking at all the names on an array of flatpanels, simulating a long paper scroll, and they can arrange the torikumi by dragging them around with a mouse or touchscreen. What the computer could for them then is to perform certain sanity checks, like checking for excluded combinations, asking for confirmation on bouts that seem dubious based on relative rank and record and so on.

That's pretty much what I'm getting at for a real-life system...they could have as much or as little manual involvement as they want, though I'm assuming if they did get such a system up and running, they'd soon decide to pass the stuff that's literally 100% straight-forward over to the automation. (Or a hybrid approach: Have the automation suggest each pairing based on general principles, the committee clicks "confirm" if they're okay with it and the system moves on to the next matchup.) With the paper version you're locked into doing everything by hand, so there's no possible way to speed things up (except having more eyes look at it).

The one IMHO major improvement a computerized system could bring is that it can display data selectively: While you're trying to pair all the 2-2 rikishi, you don't need to see any of the 0-4's, 1-3's, 3-1's and 4-0's, but with the paper version you always have everything in full view at the same time. I'm sure having a lot of practice with it helps, but I suspect there's still a lot of "now where's the next rikishi with that record..." pecking around in those scheduling sessions, especially with the more lopsided records in the second week where that next rikishi might be 5 or more ranks away. (And watch out if that's somebody who can't be used as an opponent, now you might have to look through another 5 ranks.)

Edited May 5, 2014 by Asashosakari

[Yubinhaad 11,606]

The dohyo-matsuri took place inside the Kokugikan today, so here are a few pictures.

As always the top ranking rikishi attended. Yoshikaze, Goeido and Kisenosato here.

Kakuryu, Kotoshogiku, Tochiozan and Chiyootori.

The three Yokozuna, plus Asahiyama-oyakata.

While the three gyoji on the dohyo are the focal point of the ceremony, all of their fellow gyoji also attend.

Rijicho Kitanoumi, shimpan director Isegahama and deputy directors Asahiyama and Izutsu.

Three former Ozeki, now all members of the shimpan group - Fujishima (Musoyama), Sanoyama (Chiyotaikai) and Asakayama (Kaio).

Students of the sumo school also attend the ceremony, watching from the second floor.

The participating gyoji were Kimura Shonosuke, Kimura Keinosuke and Kimura Yukihiro.

Kimura Shonosuke places the shizumemono into the centre of the dohyo.

The three Yokozuna look on.

Tate yobidashi Hideo leads the furedaiko parade around the dohyo three times. The drummers were Tasuke and Soichi.

Meanwhile, outside the Kokugikan the itabanzuke, written by Kimura Kankuro and Shikimori Kazuki, is now in place on the drum tower.

Also outside the Kokugikan, Yokozuna Kakuryu's first yusho portrait was unveiled alongside Hakuho's from the Hatsu basho.