Nagoya Banzuke Postmortem

The banzuke committee hard at work

Well, the official rankings for the Nagoya basho are out, and while the crystal ball fared reasonably well, there are some real head-scratchers among the banzuke committee’s decisions. Let’s take a look at what my predictions got right and wrong.

As expected, there were no surprises in the upper ranks from Sekiwake to Yokozuna, where all eight placements went exactly to form. But just below that, we got our first big surprise, with Ryuden (M5w, 10-5) taking the West Komusubi slot in place of Asanoyama (M8w, 12-3). Not only is Asanoyama’s combination of rank and record clearly superior, but he also won the yusho! Oh, and he defeated Ryuden in their head-to-head meeting on Day 9. Neither man has been ranked in San’yaku in the past, so that can’t have given Ryuden the edge either. Pretty much all of the other forecasts I’ve seen also had Asanoyama at Komusubi, so this decision definitely qualifies as a puzzler. EDIT: Now that the Guess The Banzuke results are posted (your humble prognosticator came in 8th, his best result to date), we have some numbers. Of the 83 players, 74 had Ryuden at M1e, and only 9 had him at West Komusubi. As for Asanoyama, 8 had him at M1e, 1 at M2w (???!!!), 40 at West Komusubi, 28 at East Komusubi, and 6 at West Sekiwake. So the committee’s decision was clearly a surprise to those who try to forecast the banzuke on a regular basis.

And the committee’s work did not get any less puzzling in the upper maegashira ranks (M1w-M4w). It was reasonably clear which seven rikishi should be placed here, but their order was anything but. The committee seems to have pretty much randomly drawn names out of a hat, resulting in two pairs of rikishi who posted identical records at the same rank at Natsu (M2e Endo and M2w Daieisho, both 7-8, and M7e Shodai and M7w Meisei, both 10-5) being placed a full rank apart. Aoiyama got the benefit of the doubt sometimes given to make-koshi San’yaku rikishi, only falling from Komusubi to M2e with a 6-9 record, but Ichinojo did not, falling from East Sekiwake to M4w. Ichinojo’s placement was the only one my forecast got right in the nine ranks from West Komusubi to M4.

The forecast improved significantly from there, placing 18 of 24 rikishi from M5 to M16 at the correct rank, and with only one switch of sides (East vs. West). That switch highlights internal inconsistency in the committee’s decision-making. Both they and my prediction had Myogiryu (M5e, 6-9) and Tomokaze (M9w, 8-7) at M7 and Okinoumi (M4e, 5-10) and Onosho (M10w, 8-7) at M8. But who gets the more prestigious East side? Well, you could go with the higher-ranked rikishi, who faced tougher opposition, or you could give it to the rikishi with the winning record. I went back and forth on this, and ended up going with the former for my predictions. The committee, it appears, simply flipped a coin each time, ranking Myogiryu ahead of Tomokaze but Onosho ahead of Okinoumi.

I never even considered ranking Kotoeko, already ridiculously over-promoted to M11e from M15w, where he eked out an 8-7 kachi-koshi, ahead of Yoshikaze (M6w, 4-11), yet that’s exactly what the committee did. I am also puzzled by how Toyonoshima (J1e, 8-7) ended up all the way at M14e, ahead of both Yago, who had a better numerical claim to the rank and was already in the top division, and Kotoyuki, who posted a better rank-and-record combination in Juryo. Kaisei is also lucky to be ranked at M15w (I had him half-a-rank lower) after managing only 3 wins from M8.

Overall, my forecast had 28 of the 42 rikishi at the correct rank, with all but two of these on the correct side. Of the 14 misses, 5 were by half-a-rank, 6 by one rank, and 3 by a rank-and-a-half. Obviously, the biggest beneficiary of the curious decision-making is Ryuden, at the expense of Asanoyama. And rather than distributing the banzuke luck evenly among the upper maegashira, the committee saddled Shodai and Meisei with all of the bad luck, with most of the benefit accruing to Aoiyama and Endo.

37 thoughts on “Nagoya Banzuke Postmortem

  1. Asanoyama’s rank was among the first things I checked, and kind of chuckled. I am actually hoping they have him on a slow promotion track for now, he has a lot of tempering ahead of him.

    I am going to be very interested to see what happens with 4 Ozeki. I hope all of them can last 15 days. With no jungyo, scouting is non-existent right now.

    • For now, I’m just hoping all four can start. Although from the little reporting I’ve seen, Takakeisho is the only (known) question mark.

    • Now that everybody has settled in Nagoya, I predict we’ll be getting more or less daily keiko reports.

  2. I’m left wondering if they’ve retroactively decided that Asanoyama’s ‘win’ against Tochinoshin was, at best, unconvincing and, at worst, didn’t happen at all. Obviously, they can’t officially remove a win from the record, but it’s certainly within their power to re-evaluate dubious results when they put the banzuke together.

    • It’s not something they’ve tended to do in the past, either with dubious results or with fusen wins, but who knows what they were thinking. I blame too much sake at the meeting :lol:

  3. Honestly baffled this is an issue for you…no M8 has ever been promoted to Komusubi with 12 wins in the history of the sport…Kyukutenho won 12 at M7 and wasn’t promoted beyond M1…this 192% follows precedent…this was imho an obvious call…10 wins at M5 is absolutely not worth being leapfrogged by 12 wins at M8.

    • I’m glad this seems obvious to you; it wasn’t to me or to the top players in Guess The Banzuke. In the six-basho era, 25 rikishi ranked M8 or lower have been promoted to Komusubi with 12 or fewer wins. Most recently, Tamawashi was promoted from M9 with 10 wins and Myogiryu from M8 with 9 wins in the same tournament in 2015: http://sumodb.sumogames.de/Query.aspx?show_form=0&form1_rank=M8-M16&form1_wins=8-12&form1_year=1958-2019&form2_rank=K

      And “by the numbers”, Ryuden was 3 ranks ahead of Asanoyama, while Asanoyama’s record was worth 4 extra ranks of advancement.

    • You’re confusing absence of evidence for evidence of absence. Or in other words: Banzuke-making doesn’t work the way you think it does. (And I’m basing that assessment not only on your post here; you’ve made similarly misguided comments before.)

  4. I expected the Yusho to be enough for Asanoyama to be Komusubi; but I’m not shocked or even all that surprised that Ryuden got the nod. He’s been the more consistent Rikishi for a few basho now.

  5. am hoping that this will give my Yago a bit of a scare and kick-start him to string together a few winning tournaments! The time is now my little dragonfly!!!!

  6. I wonder if the committee decided to assign more weight to the quality of the sumo rather than the numerical result. Ryuden has been doing some very impressive technical sumo this year, even in his defeats. That’s why he got the gino-sho. There’s nothing wrong with Asanoyama’s technique but his ranking could be a way of saying “don’t get carried away kid, you got lucky”.

  7. Maybe they thought san’yaku would be too much already for Asanoyama? To me it feels like they try to be more careful with their top talent now, seeing if those rikishi are really capable of great sumo or just got lucky. They also could have promoted takakeisho earlier but they wanted to see one more good basho and he delivered.
    I hope asanoyama can proof himself at m1 but it will be very hard…

    • He’ll face pretty much the same schedule as M1 as he would as a Komusubi. The only difference is the different expectation that comes with a sanyaku rank, but that’s an abstract factor and hard to quantify in any meaningful way.

  8. “Hello, Shimpan cho?”
    “Oh, Takasago oyakata. Yes, this is Onomatsu. I’m just about to enter the banzuke committee meeting. Can you call me later?”
    “No, It’s about the banzuke. I want you not to promote Asanoyama to Komusubi”
    “Say again? You don’t want your deshi to be Komusubi?”
    “Right. I know it’s highly irregular, but… I already have three victory parades, six yusho parties, five public interviews and three TV appearances scheduled for him. I want him to have some practice when we get to Nagoya. If he starts san-yaku celebrations there, he’s going to go 0-15 in honbasho”.
    “I see. But… wouldn’t he be disappointed?”
    “No. He actually suggested it.”
    “That’s very… diligent of him.”
    “Oh, his reasons are a bit different. He wants to… ahem. He wants a k… a ki…”
    “Out with it!”
    “He wants a kinboshi”.
    “A kinboshi?”
    “Yes. He says he wants a kinboshi off Hakuho before it’s too late.”
    “Hahahahaha… you’re serious?”
    “Well, he seems to think it’s possible”.
    “OK, Takasago. You got it. Your man is M1. This I want to see. I’ll be scheduling myself as head shimpan for that bout”.
    “Um, Onomatsu… I’d rather you didn’t…”
    “Good bye, Takasago!”

  9. Ryuden won 4 of 6 bouts against sanyaku opponents. Asanoyoma won two out of three, one of which was dubious. One could say that he could only fight whoever he was scheduled against, which is true, but if the Committee decided they wanted another look before promoting him, their decision doesn’t look irrational. The question will be answered at Nagoya.

  10. Ok I’ve been doing a post-mortem on my own prediction – I got 26 at the right rank of which 23 were bullseyes.

    My main problem areas were KW to M4E and M10E to M11W where I got nothing at all.

    In the K1 to M4E area the only pattern I can identify is that there appears to have been some kind of parachute in operation for those in and around the joi that softened their demotion relative to winning records coming from lower down the banzuke. Hence the ‘lenient’ demotions of Hokutofuji, Aoiyama, Endo and Daieisho and lower than I expected promotions of Meisei and Shodai. You could extend this logic to the Ryuden / Asanoyama decision and redfearn4’s point would appear to support that.

    In the M10E to M11W area the level of promotion of Takagenji caused me some issues. I didn’t have much of a clue where to place him as it could have been as high as M7. But I doubted they would go higher than Shimanoumi who got a (harsh) M12E last time. In the end he got M10W so higher than I expected. But the relative positions of Kotoeko, Yoshikaze and Nishikigi were surprising as on the face of it Kotoeko had the worst rank & record combination. I can only suppose that there was some kind of additional ‘punishment’ being handed out for really bad losing records.

    This may also explain why Kaisei ended up below Toyonoshima. But then he ended up above Kotoyuki who on the face of it was very harshly treated. My only theory is that his promotion was restricted due to coming up from a middling juryo rank.

    I’d like to say that I will learn from all this and improve my performance next time, but then the Banzuke committee will doubtless apply some completely different logic next time out!

    • This was a tough one. Looking back on my placements, the only thing I wish I’d done differently was to place Hokutofuji at M1w (which I considered) rather than M2e. The other committee decisions didn’t follow many of the usual patterns, and strongly disagreed with the majority of players in GTB, the best of whom in my opinion are more thoughtful and informed about banzuke-making than the actual banzuke committee. I’ll note that only 5 of 83 players had Kotoeko as high as the committee ranked him, and only 3 had Toyonoshima as high as the committee ranked him, and I don’t think those guesses came from top players.

      • Also I’ve just realised this must be the first time the words ‘Shodai’ and ‘bad banzuke luck’ are appearing in the same sentence. I had assumed they would consider fast tracking him to Ozeki after the last basho!

      • I’ll have you know that about the only way the top GTB players are more “thoughtful and informed” is that they actually try to be consistent. I mean, I guess that’s what you mean, but the fact is that the banzuke committee does not make their processes known to anyone, so we can only really guess on what they’re making decisions based on. We can only use as comparisons previous outcomes, and some of the reasons behind those outcomes might not be what we might initially think. Like back in 2014 when Goeido was inexplicably ranked ahead of Tochiozan despite a much worse KK when they were both at Sekiwake, it took us years to realize that it wasn’t because Goeido was closer to becoming Ozeki (and interestingly enough, he did get promoted that basho), but because the committee has decided now to not re-arrange KK Sekiwake, which we found out when Takayasu, Kotoshogiku, and Tamawashi all had KKs at Sekiwake. But then promptly reversed that as being the overriding criteria with Mitakeumi’s yusho, and reversed the apparent “only change on a Yusho win” with Tamawashi’s Yusho. From all of that, it is crystal clear that they can make any decision they want to, and it all has to do with how discussion goes when they are in the room making it, not based at all on any particular precedents. Obviously because the people in the committee only change very slowly and are by nature conservative as Sumo elders, we can learn a lot from precedent, but we can’t always assume that it’s of the utmost relevance.

        • Fair enough, “consistent” would have been a better choice of words. What’s interesting is that the decisions are quite consistent and predictable ~90-95% of the time, and then occasionally they do something that deviates from precedent, and isn’t even internally consistent for different instances of the same scenario within one banzuke (e.g. the order of Myogiryu vs. Tomokaze and Okinoumi vs. Onosho this time).

          I assume you play GTB, but under a shikona that’s different from your handle here?

          • I’m a regular contributor to SumoForum who is currently one of the top GTB players. I already had a WordPress account, like, a decade before this website launched, which was well before I started following sumo. I wasn’t going to create a new one. It doesn’t really matter who I am; I don’t trade on my reputation.

      • Oh, and you can click on the name of a rikishi in the left-most column of the selection stats and see who picked what for each rikishi.

  11. To me it looks like there is too much discussion going on from the “western” point of view which relies only on numbers and seeks to quantify anything and everything. Japanese (and Eastern for that matter) way of thinking is not only all about numbers and it relies on other, non-quantifiable, influencers – something that Japanese Sumo followers have no problem with

    • I’m not sure it’s right to characterize a quantitative approach as purely “western” when sumo is and always has been a heavily data-driven sport. Numbers are everywhere in sumo, from the 33/45 requirement for a sekiwake to make Ozeki or the 10-wins to return after demotion, to the convoluted way they measure a rikishi’s bonuses. The Japanese love data and stats as much as the rest of us. They may also consider other factors, like hinkaku when they promote an Ozeki to Yokozuna, but then so do most western fans I’m familiar with. Plenty has been said about the quality of Asanoyama’s sumo in May.

      • Actually, the 33/45 requirement doesn’t exist other than in the imaginations of some sumo fans, and maybe some sports writers who convinced themselves that correlation is the same as causation. The 10-wins back from Ozekiwake is, indeed, in the rules. But even Yokozuna promotion criteria have the “or equivalent” subclause. The convoluted bonuses are not so much about the sports or statistics as it is about money, and Japanese do love making business stuff as complicated as possible.

        Basically, I’d say something in the middle: they don’t mind numbers and statistics, but they always want to have humans make decisions by consensus, based on their “ki”, rather than by formula which would leave the human factor out.

        • It may exist only in people’s minds, but most of those minds are those of Japanese people, ergo the point stands that a focus on numbers is not a western way of thinking.

        • I’m quite sure 33/45 is also on the minds of those on the shimpan committee, if only because sustained external discussion of it over the years has created a feedback loop back to those in charge. They wouldn’t feel the need to publicly justify their decisions when they go against that guideline (in both directions), but they do.

  12. Via the Sumo Forum:

    An article about the banzuke committee decision to make Ryuden komusubi and not Asanoyama mentions that the representative for Takasago-ichimon was physically weak deputy chief Nishikido alone, while Ryuden’s Nishonoseki-ichimon is represented by same deputy chief Takadagawa (Ryuden’s shisho) and chief Onomatsu. Takadagawa commented “I have nothing to do with it. (also deputy chief) Fujishima and Asakayama endorsed Ryuden.” and escaped from the media. http://npn.co.jp/article/detail/70162027/

    Maybe someone who speaks Japanese can translate more of the article.

    • Great catch. I’m not surprised, though. He faced rather weak competition, didn’t beat Tochinoshin (winning a bout and defeating an opponent being distinct). Ryuden really won 4 of 6 sanyaku bouts. Asanoyama “won” 2 of 3 with a big fat asterisk on one of those.

      • Sure, but there’s a reason why I and most of the GTB players picked Asanoyama based on available precedents. So I’m not surprised politics played a role.

Leave a Reply to OneBigBuggaCancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.