skepticalsports

Hi all. Thinking of preparing a (shorter) article for 538 (my tl;dr sumo feature is here) in the event that Kisenosato wins and is promoted. It seems like, up until this year, Kisenosato had been an above-average ozeki, but probably below-average for a yokozuna. E.g., since making ozeki he has won 63 percent of his matches against other ozeki, but 38 percent against yokozuna. Here's a quick breakdown: 2012 2013 2014-15 2016 M 77.78% 85.37% 79.76% 85.71% K 91.67% 81.82% 78.26% 85.71% S 75.00% 50.00% 57.14% 87.50% O 47.83% 78.57% 60.00% 87.50% Y 28.57% 58.33% 29.62% 44.44% In 2016 he has been especially consistent against non-yokozuna opponents, particularly in March and May bashos where he won all of his matches against them (29 match win streak against <Y opps is the longest of his career by a lot. Previous longest was 17 in 2013). So I guess what I'm wondering is: Is there any narrative about what happened between January and March to help explain why he made this sudden jump (at age 30)? Or put a bit differently: Is there any plausible explanation other than just "on a pretty good run right now"? As always, links to sources in English are always appreciated.

Newbie question: I see in 2004 Hakuho was M1w when won jun-yusho, shukun-sho, and had a kinboshi -- against Asashoryu -- and was still only promoted to K1w. Is it just literally impossible to skip ranks? No, certainly not. It just happened that both Sekiwake finished with kachikoshi and, as a result, neither man vacated his sekiwake slot. Though had Hak been a 12- or 11-win komusubi you might have seen the bigwigs open a haridashi-sekiwake (or S2e, whatever they want us to call it) slot. No less than 147 modern-era maegashira have been directly promoted to sekiwake: http://sumodb.sumogames.de/Query.aspx?show_form=0&form1_rank=M&form1_year=%3E1959&form2_rank=S Thanks, helpful and makes sense. Kiyokuni went from M13! On the other end, M9 Toyonoshima went 14-1 and was only promoted to M1. But definitely no one has gone from maegashira to ozeki (if I'm using that tool right).

Newbie question: I see in 2004 Hakuho was M1w when won jun-yusho, shukun-sho, and had a kinboshi -- against Asashoryu -- and was still only promoted to K1w. Is it just literally impossible to skip ranks?

The data for both the article and the interactive is through the March basho and is unlikely to be updated unless we do some kind of follow-up. The shikona thing I noticed as well when I saw Tanikaze (who is mentioned in the article) listed under a different name. I discussed it with the guy who programmed the interactive and there's a chance we'll switch to last shikona used next week or when he has time. Speaking of which, are there any rikishi who are best known for a shikona other than their last? The interactive has been updated to use last shikona. http://projects.fivethirtyeight.com/sumo/

I don't know of any historically that made sanyaku and would be covered by the data set. Currently there's Yamaguchi who is using his family name after having used Daikiho as a sekitori and for a few tournaments after he fell a lot in the rankings and having used Yamaguchi originally when entering. He never made it beyond the bottom of Makuuchi, but if you discuss him now you'd probably be understood more if you used Daikiho than the very common surname Yamaguchi who sounds like a new recruit given how obvious it is his surname. There's also a guy that just retired as Wakakeisho who is known for going 7-0 as a Ms15TD debut and not getting promoted to Juryo despite that normally be a "guaranteed" promotion, so he's sometimes known by his original name, Shimoda, which he used then. He switched it a few times afterward, but since he never came close to making Juryo again no one really cared all that much. On preview Asashosakari knows a whole lot more about that than I do having followed the sport longer. -- The chart specifically says published May 13, so I would expect any data that is presented to me would be verified through the publish date. I would add "Data through(/as of) March 2016" for accuracy. If you don't care, well, I doubt anyone else cares either. But I'm the kind of person who "likes" pointing out technical mistakes such as this one. (I don't really like it, I just have a compulsion to do so that makes me seem as though I like it. I really don't. It's more like being OCD; I see something wrong, I have to point it out.) Thanks. Agree there should be a "data through" line somewhere, I'll mention it.

The data for both the article and the interactive is through the March basho and is unlikely to be updated unless we do some kind of follow-up. The shikona thing I noticed as well when I saw Tanikaze (who is mentioned in the article) listed under a different name. I discussed it with the guy who programmed the interactive and there's a chance we'll switch to last shikona used next week or when he has time. Speaking of which, are there any rikishi who are best known for a shikona other than their last?

Article is out: http://fivethirtyeight.com/features/the-sumo-matchup-centuries-in-the-making/ Thanks for the helpful emails and PMs.

Hi sumo fans, I've been working on a sumo feature for FiveThirtyEight -- a data journalism site owned by ESPN. Thank you to everyone who responded to my previous query about henkas. This is a fairly sizeable project that includes a lot of stuff on sumo history including a big interactive timeline. It should be ready to come out soon, and I hope you all enjoy. But meanwhile I'm chasing down a few loose ends, so if any of you can shed any light on a few questions, it would be much appreciated: 1. On the wikipedia page for Raiden Tameemon, it says: After 1800, he remained dominant, and sumo officials even disallowed him to use his favourite techniques in order to keep his matches interesting. But this doesn't have any citation. Do any of you know any details about this bit of history? In particular, a link to any source discussion of the subject would be most helpful. 2. In general, historical sumo ranking have been passed down through time via the banzuke. But overall tournament results dating back to the 1700s are largely available as well. Do any of you have specific details about how these older results were recorded, preserved, and digitized? I've heard suggestions that most results probably came from newspapers or books, but again looking for specific detail. Did the sumo association take over official record-keeping at some point? Etc. Again, links to sources in English preferred. 3. A bit more hazy, but I recall reading somewhere that Hakuho has been studying some new sumo style that may be more defensive than offensive (or not), but again I have been unable to find any solid sources discussing the topic. 4. Again relating to sourcing, this very common woodblock drawing of Raiden seems to be ubiquitous on the internet, but I'm unable to find the original source. Do any of you know where it comes from? 5. Additionally, if any of you happen to be verifiable journalists or academics with professional knowledge of sumo and wouldn't mind being interviewed for the story, please email me at bmorris at fivethirtyeight dot com (and if I don't respond, post here to let me know). Thanks so much for you help!

Yeah, this pretty much confirms my fears. "the phenomenon of a legal and seemingly game-theoretically important strategy" = Why don't they jump out of the way more? "foreign to American sports fan" = Hey look at the weird foreign stuff! "game-theoretically" "whether it provides an "unfair" advantage" = Let's throw a bit of math at it and pretend we understood something! Of course henkas work. That's why they happen. Now, I'd be fascinated to read a good statistical analysis of henka usage and a good game theoretical analysis of tachiai tactics, but quantifying the efficacy of henkas (which is what all your original questions were about) is totally irrelevant to why they're frowned upon. That is "a weird tradition", or to say it in a less demeaning way, it's cultural. There have already been some great replies on this, so I'll just throw in my 2 yen. 1. Winning isn't everything. Of course it's important for ranking (and therefore salary) and prize money, and winning tournaments is naturally a big deal. But there's a very strong culture in Sumo of doing things the right way (and in Japan generally, I'd say--you see all sorts of signs about correct behavior on the train, for example, or the proper way of taking a bath in a hot spring). And the "right way" for sumo is all about strength, power, moving forward, and humility. It goes well beyond the actual bouts, extending to how rikishiki behave, dress, etc. off the dohyo. It's a way of life. And the furore around Hakuho is probably more for late extra shoves than for the henka. There's even an official advisory committee which regularly nags the top guys on their sumo style and behavior. And bear in mind that sumo has its origins in shinto ritual and is still quasi-religious (but not in the western sense!). 2. Henkas aren't usually loved by the fans (you'll find exceptions on this forum, though), especially in major bouts. I want to see great bouts where both guys really give it their all, lots of attacks and counter attacks, defense at the edge, a real battle. A henka feels like I'm cheated out of a bout. And let's face it, sumo is a business and needs bums on zabutons and eyeballs on TVs, so entertaining sumo is in the association's interest. The sumo association is also part of the ministry of eduction and has a duty of evangelize the sport--meaning that standards of behavior appropriate for a responsible position in society need to be maintained (see point #1). So why are henkas tolerated at all, then? 1. They're a corrective and equalizer. Kotoshogiku (much as I like him) probably should never be a Yokozuna because he has an easily exploited weakness, which Yokozunas shouldn't have. Same goes for guys that get too big and have no agility. 2. They're not that common. Checking the old SumoFanMag data upthread, only seven guys pulled henkas more than once in 20 bouts; 12 had none at all for a year. I don't particularly like henkas, but there just aren't enough to bother me too much. (Unless it's in a bout I'm really looking forward to...) 3. Making a rule against them wouldn't be very practical (because of the definition problems discussed above), and it wouldn't be a very Japanese way of dealing with the problem, anyway. Social pressure can be used to deal with problems using more finesse. Kintamayama's translation from the ample men thread: So to summarize, henkas are fair sumo, but they're not good sumo, and that's for cultural rather that strategic reasons. There are a few here and there to keep guys honest, but not enough (usually) to spoil the enjoyment of a day of sumo. If you only try to optimize wins using all legal tactics, you're going to completely miss the point. This is an utterly bizarre thing for a journalist to write. The only way to be responsible to your audience is to be fair to and accurate about your subjects. Compromising one compromises both. I apologize if that sentence is clumsy. I mean my responsibility is to my audience rather than die-hard sumo fans, not that I would sacrifice accuracy or fairness (nb you didn't quote the previous sentence which I thought made this more clear). Again, the reason I came here is to try to get things right.

They're not independent, but a simple look at kimarite decisions would be meaningless. Many henka lead to a finish by hikiotoshi (pull down), but loads of non-henka bouts do as well, simply because pulling down your opponent is a very common move, which can be set up at nearly any stage of a bout.Other henka might lead to something completely different, because the henka "victim" didn't lose right away, but his opponent managed to get an advantageous position to his side or in his back, and initiates a different coup de grace (throw, push-out from behind, etc.). Then there are bouts like Hakuho-Harumafuji which have a finish that is basically indescribable in kimarite terms (there's none for "disoriented rikishi couldn't control his momentum and pretty much ran out of the dohyo on his own"), and they just slap something on it that - at best - is loosely related to the initial move. Yesterday, they went with tsukiotoshi (thrust-down), but you're not going to find much resemblance between Hakuho's tsukiotoshi and either this one or this one, both from the day before. Edit: And apropos that Daishomaru-Sadanoumi bout, we see two added complications right away. Did Sadanoumi henka? It wasn't an outright side-step, but also (much) less than full contact. And if it was a henka, was it instrumental in his victory? It didn't really work and he almost got pushed out before his last-ditch slapdown at the edge still gave him the win. Thanks for your helpful responses. I tried DMing you and it didn't work, but if you'd like to be quoted in my article, email me morrisb at fivethirtyeight.

I think you've got a few problems here. 1. Henka is poorly defined. For example, does Harumafuji's thing count? Where do you draw the line? 2. Success is poorly defined. If a rikishi more or less recovers from a henka but is in a disadvantageous position and later goes on to lose (or even win!) does that count as a success? How do you measure "disadvantageous position"? 3. The notion of "reckless charging" is a rather misguided. Kotoshogiku is great when he can come in hard under his opponent and stand them up. It's a deliberate strategy which plays to his strengths--he wouldn't be as good without it. But yes, it makes him very prone to henkas. 4. It's not a binary choice between blindly charging in and jumping out of the way. There's an entire continuum between, and that continuum is way more interesting than the extremes. Where do you make contact? How hard? How high? Do you go for a grip? What kind? How about some kind of slap or shove? On a higher level, I think this article is in danger of falling into two traps that are really worrying. 1. The "weird Japan" narrative. Everyone seems to think Japan is very strange, and so an awful lot articles about Japan focus on the weirdest bits, so people think Japan is strange, so... But let's face it, every country has its share of weirdness. And I have to say, having lived in Japan a third of my life, the US is at least as bizarre. (American football? WTF? And don't get me started on politics...) If you're going to write about sumo, please write about it as you would any other major sport and not as some sort of bizarre spectacle. 2. The "data journalism" thing. I've read 538 since the 2008 election cycle. I get what you do. And yes, more data is generally better than less, and a lot of journalism is really, really terrible when it comes to dealing with data and statistics. But for the data and statistics to be helpful, you need context and understanding. I'm sure a lot of the people here would be very interested in henka statistics. But if your sumo article is driven by the question of "Why don't they jump out of the way more?" then you don't have context and understanding, and a few pretty graphs aren't a substitute. This is a trap far too many 538 articles fall into. Imagine an audience whose idea of baseball was "men in pajamas hitting balls with sticks" and a writer who asked "Why don't they hit the ball where the other guys aren't standing?" and gave a few stats about defensive shifts--and you'll understand my concern. Unless your article is very long, or part of a series, or written for sumo enthusiasts, there are probably much better things to write about than henkas. (Edit: Thanks for the new thread!) Thanks for your response. My article is not "about" henka, but the phenomenon of a legal and seemingly game-theoretically important strategy attracting so much derision is a bit foreign to American sports fans. That the most successful wrestler of the present (and arguably ever) is sort of "in trouble" for them makes it that much more interesting. Hence the investigation. And if you notice the pattern in my questions, I'm particularly interested in reasons why the move is frowned upon other than just "it's a weird tradition" -- e.g., whether it provides an "unfair" advantage. As to whether it will satisfy sumo fans, who knows. I try to be as fair and accurate as possible (which is why I'm consulting with more knowledgeable people) but obv my first responsibility is to my audience.

Indeed. ESPN owns us. You can find my contributor page and posts here. I write about sports (mostly).

Thanks that is interesting, though it's kind of hard to interpret without the data. He mentions that more "brazen" henka appear to work more often, though I'd guess more successful henka may be likely to appear more brazen. Obv need to compare to rikishi records, plus there's a big difference between 62% and 80%. If it's attempted only in the cases where it's most likely to work -- like you either see your opponent overcommitting or predict that they will for some reason (like match histories) -- working some reasonable amount above 50% would likely be expected. But if it works 80% of the time then it's probably winning a large percentage of the time that it's completely premeditated, or being used by a weaker or injured opponent, or for "payback" or "to embarass" or for whatever other motives as well. Do any of you know either if that data is public or where I can find Lon Howard's contact info?

Hi there. I'm working on an article about sumo for FiveThirtyEight and have a few questions I hope you all can shed some light on for me: 1. Does anyone know the success rate of "attempted" henka? 2. Are there certain kimarite that pretty much mean “henka," and vice versa? or are they independent? 3. Does anyone track which rikishi attempt henkas the most often, and which have them attempted against them the most often? 4. If a one competitor "overcommits" by charging recklessly at their opponent, is there an option for how to exploit it other than the henka? 5. Say a henka is X% likely to win against an overcommitted, recklessly charging opponent. What is the probability of winning without it, relative to X? Thanks for your help.