BlckKnght

While this is true for the regular matches, the 7-way playoff will include Ms3w Hitoshi (who may in fact need to fight several times). An unrelated Trivia bit: If Atamifuji wins the yusho, he'll be only the second rikishi to win the Juryo and Makuuchi titles back to back. The only time it was done before (in the SumoDB records) was by Ryogoku in January and May of 1914.

Seeing as we're headed for a seven-way playoff in Makushita and a 2-way playoff in Jonidan this basho (and the top two divisions might get playoffs too), I thought I'd check on historical records of big playoffs, to see how unusual the current situation is. Turns out, it's not that extreme. According to this query on SumoDB, the most playoff matches in a single basho is 19. That record was set in November 1970, where there were 9-way playoffs in both Makushita and Sandanme, as well as a Yokozuna vs Yokozuna playoff in the top division. The winner of the Makushita yusho had to fight 5 times! The second most matches, 17, also happened in a notable way. In May of 1982, every division had a playoff. From top division down, they were 2-way, 2-way, 7-way, 2-way, 3-way and 6-way. (After writing that, the word "way" has lost all meaning.)

I enjoyed all the Tochimaru inserts.

Yeah, my metric does reward long careers, a bit, though rare techniques are much more important than just stacking up wins with common techniques. I tried my own version of your scaling metric, though I scaled by wins only, not wins+losses (simply because I haven't written code to accumulate loss data). I also computed the KV50 value that SumoDB uses, for comparison. The scaled metric is quite interesting, as it highlights a few young technicians who are using rare techniques a lot more than you'd expect, given the length of their careers. KV50, by contrast, emphasizes diversity of techniques, without accounting for their rarity. Here's the top 30 active wrestlers, sorted by a win-scaled metric: Name Score Scaled KV50 Yamato 34% 15.36 11.3 Asakiryu 58% 14.04 13.3 Fujinoteru 146% 9.98 17.3 Ishii 49% 8.99 11.5 Kosei 4% 8.84 5.0 Aratakayama 52% 6.37 10.5 Matsuda 131% 6.35 18.4 Ishiura 220% 6.26 16.0 Satonofuji 326% 6.05 19.5 Ura 148% 5.35 16.2 Terutsuyoshi 218% 5.32 18.1 Kyokutaisei 210% 4.68 16.9 Kotodairyu 120% 4.11 14.9 Chiyotaiyo 36% 4.08 17.3 Wakatakakage 107% 4.06 11.8 Nishikiori 4% 3.88 7.0 Kainoshima 49% 3.85 16.4 Oyamada 8% 3.82 13.2 Urutora 55% 3.50 17.7 Sazanami 57% 3.42 16.5 Shinyashiki 56% 3.20 15.9 Midorifuji 71% 3.19 16.5 Sekizuka 50% 3.10 13.4 Raiho 12% 3.06 15.2 Hinataryu 33% 3.03 14.9 Gaia 8% 2.84 11.8 Mori 63% 2.78 14.8 Chiyokozan 18% 2.67 12.2 Dairinzan 4% 2.65 8.8 Kurokage 48% 2.64 12.6 And for reference, here's the top 30 by KV50: Name Score Scaled KV50 Satonofuji 326% 6.05 19.5 Motokiyama 63% 2.28 18.6 Hoshoryu 48% 2.27 18.5 Matsuda 131% 6.35 18.4 Terutsuyoshi 218% 5.32 18.1 Wakaonehara 10% 2.49 17.8 Urutora 55% 3.50 17.7 Kotoryusei 31% 1.39 17.6 Chiyotaiyo 36% 4.08 17.3 Fujinoteru 146% 9.98 17.3 Kiyonohana 6% 1.31 17.2 Chiyoshoma 62% 1.39 17.1 Wakatakamoto 32% 1.11 17.1 Kyokutaisei 210% 4.68 16.9 Kiribayama 45% 1.58 16.9 Kaiseijo 20% 1.07 16.6 Tsugaruumi 51% 2.39 16.6 Kotoegashira 12% 1.78 16.5 Midorifuji 71% 3.19 16.5 Sazanami 57% 3.42 16.5 Akinishiki 15% 1.77 16.4 Kainoshima 49% 3.85 16.4 Keitenkai 66% 2.06 16.3 Ura 148% 5.35 16.2 Dewataikai 31% 1.46 16.2 Nihonmatsu 22% 2.21 16.2 Miyagi 6% 1.25 16.2 Enho 46% 2.08 16.1 Ishiura 220% 6.26 16.0 Shinyashiki 56% 3.20 15.9 Satonofuji is back on top!

So, to give Satonofuji his full due, I scraped a couple more years of data, so that his matches from 1996-1999 are all included. His score goes up to the incridible 326%. None of the guys who started before him are anywhere near 50%, so I don't think I need to bring in more 90s data. I'm going to edit the post above to have the updated data. I don't believe any of the other scores changed much, though there will be a notable change after I add results from the current basho. Midorifuji's win on day 10 was by waridashi, which had been used by only twice before by currently active rikishi. The recalculation of its value will knock Hidenoumi well down the chart, and Sakura to less than 50%.

Hmm, the guys who have it in their list have actually got an unusual number of them. Midorifuji has three, Ishiura has two. There are only 55 in the dataset. I didn't put any effort in to sanitizing the data. It's using results scraped from SumoDB (in an off-basho time of year to avoid beating the server too hard when it's busy). Now that I think of it, my dataset probably sells Satonofuji's accomplishments a little bit short, as it only contains data starting from the January 2000 basho to the present (which is long enough to cover the whole careers of all the top division men), and he'd been fighting for several years at that point.

I threw together some data today to see which active rikishi is the most capable technician. Its fun stats, but probably doesn't deserve its own thread, so I'll throw it in here. My metric: Of all wins by all active rikishi, who has the greatest cumulative share of all of the winning techniques? That is, if you are the only one to have ever used a technique, you get 100% added to your total. If you've used a move 3 times and somebody else has used it only once, you'd get 75% added. Each yorikiri win is worth 0.001126%. Here's the list of everyone with more than a 50% score, together with the moves they have more than 3% of: Jd57w Satonofuji 326% tasukizori: 64%, zubuneri: 63%, izori: 62%, uchigake: 23%, kainahineri: 17%, kawazugake: 12%, kakenage: 12%, ushiromotare: 11%, ipponzeoi: 10%, kotehineri: 6%, sotogake: 6%, kirikaeshi: 6%, watashikomi: 5%, kubinage: 5%, susoharai: 5%, utchari: 4% Sd90w Ishiura 220% mitokorozeme: 100%, susotori: 50%, komatasukui: 23%, shitatehineri: 9%, susoharai: 5%, harimanage: 4%, hansoku: 4%, shitatedashinage: 4%, ashitori: 3% Ms1w Terutsuyoshi 218% kozumatori: 50%, sotokomata: 33%, okurigake: 25%, koshinage: 20%, sakatottari: 12%, ashitori: 10%, zubuneri: 9%, komatasukui: 9%, tottari: 7%, kainahineri: 6%, ketaguri: 4% Sd55w Kyokutaisei 210% chongake: 67%, ketaguri: 52%, susoharai: 37%, nichonage: 13%, kawazugake: 12%, tsukite: 5%, okurinage: 4% M8w Ura 148% okuritsuridashi: 50%, koshinage: 20%, tsutaezori: 17%, ashitori: 10%, tasukizori: 9%, kubihineri: 7%, tottari: 6%, ushiromotare: 6%, okurihikiotoshi: 5%, izori: 4%, hikkake: 4% Jd64w Fujinoteru 146% tsukaminage: 100%, tsutaezori: 33% Sd48w Matsuda 131% nimaigeri: 54%, nichonage: 31%, ketaguri: 9%, sakatottari: 6%, kakenage: 5%, ipponzeoi: 5%, uchigake: 5%, kekaeshi: 3% Ms44w Kotodairyu 120% yaguranage: 100%, komatasukui: 5% Sd86w Asatenmai 108% okuritsuridashi: 50%, ipponzeoi: 14%, izori: 8%, ushiromotare: 6%, watashikomi: 5%, kubihineri: 3% S1e Wakatakakage 107% okuritsuriotoshi: 100% J8e Hidenoumi 75% waridashi: 50%, ushiromotare: 11% M5w Midorifuji 71% okurigake: 25%, uchimuso: 11%, katasukashi: 7%, hansoku: 5%, koshikudake: 4% J2w Tochinoshin 69% tsuridashi: 16%, tsuriotoshi: 11%, kubihineri: 7%, uchimuso: 5%, harimanage: 4%, koshikudake: 4% Sd13e Keitenkai 66% kozumatori: 25%, kirikaeshi: 7%, chongake: 4%, shitatehineri: 4%, kekaeshi: 3% M1e Tamawashi 65% gasshohineri: 33%, hansoku: 4%, kubihineri: 3% Jd53w Mori 63% nichonage: 31%, nimaigeri: 8%, sotogake: 6% Jk10e Motokiyama 63% tsuriotoshi: 11%, izori: 8%, kainahineri: 6%, ipponzeoi: 5%, shitatehineri: 4%, kubihineri: 3%, tokkurinage: 3% M16e Chiyoshoma 62% susoharai: 7%, okurihikiotoshi: 5%, chongake: 4%, harimanage: 4%, tottari: 4%, uchigake: 4%, kekaeshi: 3% M13w Kotoeko 59% amiuchi: 18%, uchimuso: 11%, kakenage: 9%, ipponzeoi: 5% M7w Takayasu 59% komatasukui: 14%, uwatehineri: 6%, ushiromotare: 6% K2w Tobizaru 58% kekaeshi: 18%, sakatottari: 6%, okurihikiotoshi: 5%, komatasukui: 5%, ketaguri: 4%, fumidashi: 4%, kubihineri: 3% Jd76w Sakura 58% waridashi: 50% Jd13w Asakiryu 58% tsutaezori: 33%, tasukizori: 9%, ashitori: 7%, uchimuso: 5% M11e Azumaryu 57% kozumatori: 25%, uwatedashinage: 4%, kirikaeshi: 4% Sd25w Sazanami 57% gasshohineri: 33%, ipponzeoi: 5%, harimanage: 4%, sotogake: 4% Sd61e Shinyashiki 56% okurigake: 25%, amiuchi: 8%, kubihineri: 7%, kekaeshi: 3% Sd35e Shosei 56% zubuneri: 16%, kainahineri: 10%, watashikomi: 5%, makiotoshi: 4%, izori: 4% Jd95e Urutora 55% ashitori: 10%, uchigake: 8%, kirikaeshi: 5%, ipponzeoi: 5%, shitatehineri: 5%, izori: 4% Sd62e Asanojo 53% koshinage: 20%, sokubiotoshi: 7%, okurinage: 4% Sd32w Aratakayama 52% susotori: 50% Sd17w Katsunofuji 51% sabaori: 25%, tsuriotoshi: 11%, koshikudake: 4% Jd10e Tsugaruumi 51% okurigake: 25%, uchimuso: 5% Jd5e Sekizuka 50% osakate: 29%, tokkurinage: 10% Satonofuji, perhaps best known for the bow twirling he does at the end of each tournament day, is clearly sumo's supreme technician. I guess that makes sense, given that he was chosen to be one of NHK's technique demonstrators. Four techniques have only been used by a single active rikishi: Ishiura's triple attack force out Fujinoteru's lifting throw Kotodairyu's inner thigh throw (which doesn't seem to have any video, it was back in 2009, before he started using his current shikona) Wakatakakage's rear lifting body slam from last year.

There was another tsutaezori win today, in Sandanme. Here's a clip. I wonder if Asakiryu tried out Ura's move, for fun, in practice sometime since day 4, and it just happened to work in his match today. He did go in pretty low, so maybe he was even planning it from the start.

I spotted this thread today and wanted to do some analysis of the data in it. So I threw all 15 days of stats into a Google Sheet and did some rearranging (removing forfeited bouts, mostly). Here's the pivot table with the main takeaways: I'm not sure there's much that's surprising to anyone who watched the tournament, but it's interesting to see the numbers back up subjective feelings. For example, Ichinojo fights for more than three times as long as the division average. At the quicker end of the division, Abi was fastest of those who had 15 matches, though Onosho and Takanosho were not far behind. Note that the grand totals at the bottom of the table for Total Time and # of bouts are double what you might expect, since the table counts them for both of the rikishi participating in each bout. All told, we had 52.1 minutes of fighting.