Active Members
  • Content Count

  • Joined

  • Last visited

  • Days Won


Posts posted by Doitsuyama

  1. 15 hours ago, Gurowake said:

    It wasn't too hard to fix once I started stepping through the routine in debug mode to see what it was doing.  But I never thought I would need to change anything.

    If you are not happy with my site, feel free to grab from the official NSK site. But be warned, you have absolutely no idea how often I had to change my code to adapt to changes on the NSK site.

    • Like 1
    • Haha 2

  2. 1 hour ago, Kaitetsu said:

    Aki 2022 was the first time since Hatsu 2018 (coincidentaly, Tochinoshin's maiden Yusho) that Kasugano stable had 4 sekitori.

    With the almost certain demotion of Tochimaru back to Makushita, I theorize that it will be a long while until Kasugano gets 4 sekitori again, not even counting that both Tochinoshin and Aoiyama may be looking at retirement in the not-so distant future...

    This seems like a dangerous bet with Tsukahara threatening to get to Juryo pretty soon.

  3. On 13/07/2022 at 20:10, Akinomaki said:

    Maezumo today: Sachinofuji-Hamasaki



    I think it's weird that there was only one bout on day 4, why would Okuyama and Otani sit the day out? They didn't meet on day 5 as well, and it doesn't make sense to not have them meet at all as both are 2-0 against the other two. They also aren't from the same heya. I'm going to assume that they met on day 4 and that Otani won... (Idunno...)

  4. 23 hours ago, Ganzohnesushi said:

    That's funny.Why on earth did I see it on the DB already at 1:00 p.m.?


    The database for most games is different from the Sumo Reference DB and has its own torikumi. Oh, and Golynohana's games have another database with their own torikumi. They all draw the torikumi from the main DB, but the process sometimes needs to be kicked.

  5. On 19/07/2022 at 00:25, Seiyashi said:

    While we have your attention, could I rather shamelessly check what the second column on the simple style torikumi list is for?

    What second column? (Whistling...)

    (It indeed used to be an order column over all bouts of the day, but it is bugged with the automated grabbing and I'm never using it anyway, so I removed it now)

    • Haha 2

  6. These ID numbers are not at all times present. For example if you just list the torikumi of a day, you don't see this ID (and sometimes there even isn't an hyperlink where you could see the ID in the url). For this reason it does make sense for a mainstream outlet like the NSK website to change the shikona from the standard romanization. The Sumo Reference is a niche website though, and so I don't see urge for me to make shikona "changes" to cater for the audience.

    I thought about using unicode characters for long vowels but decided against it, as unicode wasn't too widespread. By now it seems to be more usual, so I might make a change to unicode characters (ō and ū) with another checkbox option on the left not to use unicode. But I'm not sure if this would even be conceived as a positive change from the users of Sumo Reference.

  7. 14 hours ago, Tamanaogijima said:


    Funny that you mention that. Whenever I dig into their data I too get the nagging feeling that their rikishi data's key column is the shikona, not any id number.

    They do have an id number, it's in the url.

    • Like 1

  8. 4 hours ago, Tamanaogijima said:

    I did not change the birthdays, though. Since Doitsuyama has an offline file for biographical data, too, that one would become outdated then.

    That's right, but I'm happily correcting birthdays if needed. I didn't quite see what has to be fixed though?

  9. 23 hours ago, Tamanaogijima said:

    Do the rikishi yearbooks count? In the H18 book - the last one he featured as 駒乃富士克明 - there is かつあき furigana for his shikona given name.

    Small limitation: The H18 yearbook is the first one which has furigana for given names at all, before that it was only for the shikona. Now, the yearbooks had their flaws as well, but the idea that 克明 reads "Katsuaki" doesn't come completely out of the blue.

    I guess someone has to ask him via twitter...

    I'm pretty sure I got that reading from such a yearbook or another source and from that took it for his real name. But sure, if the Kyokai lists his name as Yoshiaki I will change this now, and also the old shikona names on the assumption that a mistake is more likely than a different reading (which would also not be impossible).


    • Like 2

  10. 5 hours ago, Seiyashi said:

    Was about to comment that after the implementation of the initial 6/10 rate limiting, the UX for the DB tended to be: load into DB landing page, immediately click onto link that the user wants to visit (e.g. banzuke, torikumi, kabu), then immediately get 403ed because it was too fast. That hasn't been a problem since about mid basho, so I assume that was corrected with the new 20/20 rate limit?

    No, I just did that today. I think since about mid basho you hit the DB so frequently that ressources like images never left your cache and didn't need to be loaded.

    • Thanks 1

  11. On 21/04/2022 at 18:40, Akinomaki said:

    Not just bots, the browser can also only access 1 thing at a time. Opening a page I have no win/loss marks, they are images and not all are loaded. Reload after some time has something else that isn't loaded, looking at another rikishi page quickly if the first one is not what I wanted gives me the 403 error, and the 404 one also occurs frequently.

    Performance at least was much better now, pages are loading very fast for me. But I see that one page (especially the first after a while) may load several elements like images. So I changed it from 6 requests in 10 seconds to 20 requests in 20 seconds. I can change that of course again if needed, and if performance doesn't suffer.

    • Like 3
    • Thanks 7

  12. The server is actually pretty good. I indeed suppose there are too many bots overloading the site, so I added IP rate limiting now with no more than 6 page visits in 10 seconds. If you get a 403 error now (access denied) it's because of this minor inconvenience.

    • Like 7
    • Thanks 6

  13. Some news out of the useless trivia department:

    - Chiyotsurugi set a new record with 8 losses in the current mae-zumo system (at least within my collected data), beating Shunpo who went 0-7.

    - Chiyotsurugi is the first rikishi to lose 3 times against the same opponent in mae-zumo (again, within my collected data). Before this only Sakashita and Yamakawa met 3 times in Haru 2020 and they went 2-1.

    - In fact, only once before one rikishi had 3 wins against the same rikishi in the same basho with Nanzan against Fujitora in Nagoya 1990 (of course again, among my collected data). I have 4 more cases of rikishi meeting in hon-basho and then twice in playoff bouts (but all 2-1): Roho against Wakanoyama in Hatsu 2004, Shimamusashi against Haguronada in Kyushu 1994, Konishiki against Hokutoumi in Haru 1990 and Tochifuji against Yoshinohana in Aki 1972.

    Edit: I just realize that Shimamusashi and Haguronada had all 3 bouts in the playoff, making it the only case of the same match three times on the same day.

    • Like 2
    • Thanks 2

  14. On 07/01/2022 at 11:15, Asapedroryu said:

    Who is the point of contact for issues with the Juryu Game? (the email on the website gets returned).

    I would need to get my user/password sort out.




    I fixed the email address on the juryo games website and will write you in PM about your problem.

    • Thanks 1

  15. 1 hour ago, Hankegami said:

    I made a couple of line charts to better show the situation (see attachments). The first shows the BMI (Body Mass Index) of all Yokozuna from Tanikaze to Terunofuji. Although one could object that early data might be not completely realiable, the pattern still shows that every Yokozuna was at least within the Obesity I range and most were in Obesity II-III, that is grave obesity. Of course, we must include muscles and all, but there has always been an undeniable tendency to be as fat as possible in documented sumo history.


    Nice charts, but the highest BMI on the far right seems to be a serious mistake...

  16. 17 hours ago, Asashosakari said:

    Which NSK source are you using?

    In any case, maybe @Doitsuyama still has his original sources at hand and can confirm it as possibly a case of erroneous data entry. If it's outright wrong information, the rikishi should probably just have his heya data removed altogether (which wouldn't be unusual for that era).

    Hmm, since there are 3 height-weight entries I should have those details somewhere, but I will need some time to investigate.

    • Like 1

  17. On 29/09/2021 at 05:11, Gurowake said:

    So I guess this is very good evidence that they (at least, this basho) draw up the banzuke immediately, and it just doesn't get ratified until Wednesday.  There was certainly enough time for them to put together the banzuke if they were so inclined, and probably didn't want to throw it away when Hakuho suddenly retired.  My guess is that they will list Hakuho, since he's not having a disciplinary-related retirement.  At least, since we're going to have to submit 42 names for GTB, that's the obvious thing to do there.

    I rather thought it's not about disciplinary-related incidents here, but about not listing a person twice on the banzuke and he will be listed as Magaki. Obviously Sokokurai got listed twice but as was said already, Makushita is different from Makuuchi and especially Yokozuna.

  18. 40 minutes ago, yorikiried by fate said:

    It was definitely a memorable thing with e.g. Justine Hardenne (before ill conceived comebacks).

    Why "Hardenne"? At the time of retirement she wasn't married anymore and was back to Henin.