To rehash: at November’s Kyushu Basho, Takakeisho lifted the Emperor’s Cup and picked up with it all of the appropriate speculation over where his career will take him. Will he take advantage of the flux in the upper ranks and follow Tochinoshin to a swift Ozeki promotion? Or will that opportunity fizzle, as we saw from Mitakeumi? It may come down to a question of fighting style.
In the NHK’s English-language preview special, Murray Johnson makes the case that Takakeisho needs to develop his yotsu-zumo in order to reach the pinnacle of the sport. Oshi-zumo is a pushing-thrusting style of sumo while Yotsu sumo is generally known for grappling and heavy use of grips on the opponent’s mawashi, or under the arms.
I had an interesting exchange with Herouth and Leonid about belt-battling versus pusher-thruster-style wrestlers. My contention was that specialization can get a wrestler into the upper echelons of the sport. What I didn’t realize until I dug into the data was just how committed to oshi-zumo Takakeisho is…perhaps to the point of “one-trick pony” status.
This chart (data from the amazing SumoDB) shows us how Takakeisho has won and lost bouts during his career. His bouts are almost exclusively fought to his style. Not only do his wins come largely from oshidashi, the majority of his losses do, too, with hatakikomi slap-downs. Tsukiotoshi is probably the most “grappling-style” that he uses to win, while his opponents have some success with yorikiri/yoritaoshi, as well.
I’m still in a bit of disbelief that Takakeisho is this allergic to yotsu-sumo. He has won two bouts by yorikiri…only one of those was a makuuchi bout. I had brought up the infamous Kotoshogiku as an example of a successful “one-trick pony,” this time of the yotsu-style. He has been able to reach Ozeki, nearly exclusively using his patented hug-and-chug.
As we can see, nearly all of Kotoshogiku’s bouts are fought in his favored, yotsu-style. Even when he loses, the winner generally wins with yorikiri or with throws. I’ve always been surprised that Kotoshogiku has never developed throwing skills of his own and I think that (and being injury free) could have taken him to Yokozuna, without needing to develop an oshi-style.
The two styles seem to be diametrically opposed, physically, with entirely different muscle development and training needed for each. Kotoshogiku clearly focused on legs. Others who follow that mold are Toshinoshin, Terunofuji and Ichinojo. Large men with massive, powerful legs. Tochinoshin adds another muscle group complementary to that yotsu-style: the trapezius (neck/back/shoulder) muscles and powerful biceps. Arawashi is another yotsu-style wrestler, though he’s not quite as one dimensional. He’s clearly been able to evolve throwing capabilities to make up for his smaller size but he’s not quite as capable of dictating a yotsu-style bout, as the others are, since more of his opponents are able to force an oshi-bout.
We love the dramatic Kaiju-mode of Terunofuji, Giku’s hug-and-chug, and Tochinoshin’s atomic-wedgies. But it comes at a cost. All of these guys put immense strain on their knees and backs. With so many of the top ranked yotsu guys struggling with injuries, I’m not surprised by the sudden surge of the largely oshi-style tadpoles.
Abi is a great example of an oshi-wrestler, hoping to cross-over into yotsu-dom. As Herouth has mentioned in her Jungyo reports, Abi is trying to battle more on the belt but we haven’t seen it much “in prime time”. What shocked me most is that he actually LOSES more to oshidashi than he wins. The big difference seems to be his ability to win by hatakikomi. Perhaps this is why he’s felt the need to diversify a bit more urgently than Takakeisho?
This said, I don’t think we will see Abi successfully shifting any time soon because it is such a dramatic change. In the NHK preview video, Onosho – another oshi tadpole mentioned how his training has changed. And he seems to already have a bit more of a yotsu foundation than Abi. Takakeisho would probably be more like Abi, requiring a drastic shift in weight training and tactics. But he’s also a much more successful oshi-artiste so he may be able to carry on longer and advance further. Tamawashi has been able to remain a sanyaku main-stay with an almost exclusively oshi-style and Takakeisho seems better because he is able (so far) to dictate the style of sumo.
None of the current Yokozuna, however, are oshi-specialists so my question about whether Takakeisho can be one is still not looking good. I had pointed to Harumafuji. As a small(er) champion he bled kinboshi as larger opponents could pick off wins. He had that henka-non-henka-Tazmanian-devil-death-spin thingy that he could resort to…which may have exacerbated those elbow issues?
Anyway, I guess we’ll have to wait-and-see where this goes. But before I close, I want to mention Yago, Gokushindo, and Enho. These guys are interesting to me because they’re 1) successful, 2) young, 3) diversified. Yago and Gokushindo are very interesting as balanced oshi/yotsu guys. While Enho is just totally different, characterized by shitatenage throws.
I hope you enjoy playing with the new visualization. I’ve sure enjoyed building it. Keep in mind this is ACTIVE wrestlers only, for now. I hope to add some of the recently retired soon. Again, a huge thanks to the SumoDB for such amazing data.
I think that Takakeisho’s reliance on oshi is more a question of necessity rather than preference. With his short arms and huge circumference he can’t really reach his opponent’s belt, and it’s very difficult for a rival to get to his. He will continue to play to his strength. Old proverb says “the tiger who mocked the elephant for having neither claws nor fangs became a very flat cat”.
My thoughts also. Takakeisho is limited by his stature but nonetheless remains very dangerous indeed. I suspect very intense video scrutiny of his “style” is on going.
Fascinating! More analysis like this please!
Is there a clear example of a Yokozuna in recent memory who was a clear oshi-zumo specialist?
Takakeisho says very clearly that he can’t do mawashi sumo. When he interviews, instead of the usual “I would like to keep doing my own brand of sumo”, he says “I would like to keep doing my tsuki-oshi brand of sumo”. In one of the post-yusho interviews he said explicitly: “I can’t work on the mawashi, so all I can do is polish my tsuki-oshi”.
He has smaller hands than most other sekitori, and shorter arms which are pushed to the sides by his own bulk. Abi is completely different. He has the reach for yotsu, so it’s just a question of will and training. He simply belongs to a heya whose master was known for his quick tsuppari, so it’s no wonder oshi is his style of choice. Aoiyama is an oshi wrestler, but has the reach for a mawashi fight. If you have long arms, you can potentially do either, it’s just a matter of developing your skills.
As I said in a comment a while back, it’s not that there weren’t Yokozuna who specialize in oshi. Akebono was one and Demon Kakka mentioned him as a role model for Abi (they have a similar body shape, though not the same size obviously). Hakkaku, then Hokutoumi, was an oshi specialist. But they all had the skills to go into a yotsu battle when they felt the need.
When Hakuho was promoted to Yokozuna, Hiro Morita mentioned that he used to be a one-dimensional mawashi man in the past, and needed to develop his tsuki-oshi to become Yokozuna. Thinking of Hakuho as one-dimensional boggles the mind, but that was 12 years ago.
I think injuries have more to do with the rising average weight more than they have to do with the type of sumo practiced. Yotsu was the reigning style for many years. Suddenly wrestlers decided it was not good for their health? Yutakayama and Ura are oshi wrestlers.
It would be interesting to see what the styles of the current stablemasters are. If there happens to be a large percentage of stablemasters who used to be oshi specialist, that could be the reason for the transition in styles. Also interesting to check the percentages among university wrestlers who entered by tsuke-dashi or had a very fast promotion. This would mean they got their style at University rather than at their heya. So it would be interesting to see if the style distribution is different than the general rikishi population.
Overall Yokozuna win stats by kimarite (top 4): 1. yorikiri 3460 2. uwatenage 983 3. oshidashi 941 4. yoritaoshi 672
all i can say is Yago Yago Yago! u just watch my little dragonfly hover!
Is there an API on SumoDB? I’d love to play with this data myself.
I was about to ask Andy what his secret was… So far the only thing I could do was write scripts for downloading and parsing pages (ugh).
Basically that. I used Spyder to crawl through the torikumi from 1985. I went back that far because I wanted all active wrestlers and that meant including Hanakaze. I need a more efficient process so I can update the old career visualizer and this one. Since the government is shut down, I may have the time over the next couple of days. Also, the kimarite data for ranks below Juryo seems to disappear around 1987.
Maybe we could try and convince the SumoDB guys to open up their data through an API? That would be very handy for a lot of stuff. I’m building a sumo-related game and I could really use something like that.
I would be concerned about load on their servers for anything that intense. I space queries out with a low load because I don’t want to stress those servers.
It’s not going to be intense, it’ll just read the data once every day, to get the newest results of the basho currently going on. So not more than a person getting all the data through the site. Heck I’d host the data with an API on it if I had access to the db.
You can contact Doitsuyama via the feedback form on the database website if you want to help out with anything. I can’t say he’ll be particularly interested in doing much on his end in terms of new features since he barely seems to have time to keep everything together as it is now. For people that aren’t long-time contributors to the Sumo Forum community it might also be hard to get the level of trust needed to get special access, but that’s just my impression of things. Regardless, you would need to contact him; he’s not going to be reading any of this.
Instead of going by date, get the active rikishi page (“rikishi” then on the left, from the “intai in basho” select “active”), then crawl from there… Sigh.
In the torikumi list, there are anchor tags with the date of intai. So I parsed that, shikona, birth, height, weight, heya…I’m eager to get to use the height and weight data but I’m still toying around with it right now.
The heights and weights are not really up to the quality of the rest of the data in that database. They have Ichinojo at 183kg…
You called it. It’s like there haven’t been updates in the last 3-4 years. I pulled the data in, put it into a new frame on the dashboard but for many wrestlers, it’s a straight line.
Yeah, the height and weight haven’t been updated except for new rikishi in a very long time, since before I started following sumo. I don’t know if that’s because of a lack of available data, lack of ability for Doitsuyama to easily give volunteers access to upload the data, or something else.
John Gunning and Herouth have posted sekitori weights before and put it on Twitter. I’ll do my best to see where I can find the data and could hopefully update it on the SumoDB.
That’s a good idea. I just used a script that kind of used the site as an API…I just had to parse the HTML.
Same here – would love to have go at this and maybe come up with some nice visualizations. Hm.
Thanks for this incredibly interesting piece. I am a big Takakeisho fan for a number of reasons – I strongly admire his fighting spirit, he doesn’t seem phased by opponents of greater rank than him, and he was also very nice to me when I met him on my first time attending sumo haha (back when he was fighting as Sato in Juryo).
I totally understand that diversification of sumo technique certainly gives rikishi a greater edge, particularly when they can implement a number of them very well. Being focused exclusively on one kimarite offers fewer routes to victory.
Firstly I would say that ‘Oshi zumo’ is clearly not a homgenous concept. There are many minute details in the Tachiai and the offensive charge (in the way the pushing is implemented, in the way the centre of gravity is kept, in the way the legs are used etc etc), meaning that there can be a wide repertoire of ways that a rikishi can achieve that otsu win. Ten consecutive oshidashi wins could all offer a great amount of variation in approach – these could draw on a number of Takakeisho’s strengths and exploit different forms of the opponents’ weaknesses.
Secondly, Takakeisho can also work on his defence to develop a wide variety of ways that he can overcome the yotsu zumo of his opponents, if indeed they are the sort of rikishi that use that brand of sumo particularly well.
Thirdly, Takakeisho is very young and hard working. He will continue learning and improving for a while.
Fourthly, Takakeisho rules. Lol.
I agree… he is amazing. Although I didn’t like the “slap” bout against Mitakeumi in Aki (2018, day 7) – but luckily Mitakeumi wasn’t impressed :-) …I’ll keep my fingers crossed that Takakeisho can beat Hakuho this time.
BEST English-speaking Sumo site out there! Way to go Andy. I like what you did there: the conversation with other contributors and readers led you to produce this interactive chart for the debate on specialization or not. You guys and gals need to be PAID for all the hard work! Curious: are you able to make a lil’ change here and then?
And yes, I have added the analysis visualization to my bookmarks. Outstanding work indeed, sir! Now, I have a ready-to-view break down of strengths and weaknesses, wins-to-loses ratio to go by.
And yes, I have visited SumoDB — it is a very good sumo site for information on the rikishi, and tip-of-the-hat to all the volunteers that keep it going, but it does need some updating. But who knows, maybe Doitsuyama’ll reached out to Gert, Andy, C. Drake, Herouth for ideas — one of these days. It has been a while, but I could give a few design/database schemes, too. Oh well…
I reached out to them, but have not received a reply yet. Still hoping we can do something with the SumoDB data.