Hakke yoi!

[Kashunowaka 300]

:-)

You are welcome to visit my new sumo site at http://hakkeyoi.net!

Besides some of the usual stuff (banzuke, sumo links), the gist of the site is the ability to make custom queries to the database. I hope this will be interesting for all you stats junkies and/or GTB players out there. :-)

(Beta, thread carefully...)

(At last an opportunity to use that smiley!)

Please let me know when you encounter bugs and errors (which you undoubtedly will).

[furanohana 3]

(Applauding...) :-) (Applauding...) :-) (Applauding...)

I think this is a tool I'm going to use very often... (Dribbling...)

[Doitsuyama 1,259]

Excellent! Gives me some ideas how to make my database online when I'm ready with the data collecting stuff... :-) Your site is closest to what I imagined to do myself.

[Asashosakari 20,220]

Neat!! :-) :-)

Do you accept ideas for query additions as well, besides bug reports (haven't encountered any of those yet, btw)?

[Doitsuyama 1,259]

Oh, bugs and errors? The site seems to be flawlessly. It would be nice if the time of shikona change could be seen somehow in the rikishi details... And the missing records 1973 Hatsu of Asaarashi and Katsuhikari are 3-12 and 6-9 respectively. And Daimonji and Aobajo in Haru 1973 are 5-10 and 2-13... Hmm, on closer look quite some early lower Juryo records seem to be missing. How come?

Edit: Maybe you guessed the lower Juryo results from the changes in the banzuke? I saw Chiyozakura in Natsu 1973 listed as 9-6, but I have him 8-7, which probably is correct. Then I can see why the missing results are missing, because they got demoted to Makushita. Now where did I find the complete records again? Escapes me now, sorry. :-)

Edit again: Ah yes, they are here: http://www.asahi-net.or.jp/~AN4S-OKD/private/sumou/sma.htm back to 1971.

Edited December 8, 2003 by Doitsuyama

[Kashunowaka 300]

Excellent! Gives me some ideas how to make my database online when I'm ready with the data collecting stuff...  :-) Your site is closest to what I imagined to do myself.

I hope you won't hold that against me. :-)

Do you accept ideas for query additions as well, besides bug reports (haven't encountered any of those yet, btw)?

Of course!

[Kintamayama 47,174]

Awesome. Awesome. And some.

[Yoavoshimaru 0]

Great site, thanks :-) (Although I would have started this thread in another area of the forum) :-)

Edited December 8, 2003 by Yoavoshimaru

[Kashunowaka 300]

Oh, bugs and errors? The site seems to be flawlessly. It would be nice if the time of shikona change could be seen somehow in the rikishi details...

Yes, I agree about that. Yubiquitoyama's site has information about shikona changes for all rikishi who ever visited makuuchi. I managed to find a lot of changed shikona that way, but for rikishi who never got past juryo there are probably several that have change shikona while in makushita (or juryo) that I don't know about.

I obviously know the shikona for rikishi while in juryo or makuuchi, so I contemplated a couple of solutions on how to best display that information on the rikishi page. I wasn't quite happy with any of the solutions I came up with, so I left that out for the time being.

And the missing records 1973 Hatsu of Asaarashi and Katsuhikari are 3-12 and 6-9 respectively. And Daimonji and Aobajo in Haru 1973 are 5-10 and 2-13... Hmm, on closer look quite some early lower Juryo records seem to be missing. How come?

The data are from Chijanofuji's site, and some juryo records are missing there. I would be happy to eliminate any question marks in the current data.

[Kashunowaka 300]

Great site, thanks  :-)  (Although I would have started this thread in another area of the forum)  :-)

Hm... perhaps you are right. But then, which one?

Edit: personally, I would have liked to post in "Sumo Related Subjects", but none of the subforums apply. And the post is not off-topic either. That leaves me with this subforum. (Applauding...)

Edited December 8, 2003 by Kashunowaka

[Kashunowaka 300]

Edit again: Ah yes, they are here: http://www.asahi-net.or.jp/~AN4S-OKD/private/sumou/sma.htm back to 1971.

Thank you very much, Doitsuyama-zeki. The data will be updated shortly.

Only one question mark:

Mitsuruyama J10E 1978 Kyushu 5-9

Should that be 5-10, 6-9 or 5-9-1?

[Asashosakari 20,220]

Do you accept ideas for query additions as well, besides bug reports (haven't encountered any of those yet, btw)?

Of course!

Okay, here goes. Judging from the already-existing complexity of the form, I'm guessing that the following wouldn't be too difficult to implement, but if I'm way off base with that, just forget I said anything. I love the queries possibilities plenty enough, as is. :-)

Anyway...what I wouldn't mind seeing is, in addition to the constraining criteria, also the possibility of a "highlighting" criterion that only marks a certain subset of the results, but doesn't actually filter them. To take the "Ozeki run" query mentioned on your Examples page, one could then make a similar query, but for 4 basho, with the highlighting set for Basho #4 = rank Ozeki, to make it immediately visible who got promoted and who didn't. Or one could make a query that is limited to, say, Kotomitsuki's career results, and highlight all his make-koshi. Or a query limited to all M2e who got a 8-7, and highlight those who were promoted to sanyaku. Etc. etc.

I know a similar effect can, depending on the nature of the query, be achieved by re-sorting the output by one of the columns, but I usually find the chronological order more natural and more informative, so highlighting seems like a solution to that.

Maybe I'm the only one who'd find that useful though. :-)

Edited December 8, 2003 by Asashosakari

[Kashunowaka 300]

Do you accept ideas for query additions as well, besides bug reports (haven't encountered any of those yet, btw)?

Of course!

Okay, here goes. Judging from the already-existing complexity of the form, I'm guessing that the following wouldn't be too difficult to implement, but if I'm way off base with that, just forget I said anything. I love the queries possibilities plenty enough, as is. :-)

Anyway...what I wouldn't mind seeing is, in addition to the constraining criteria, also the possibility of a "highlighting" criterion that only marks a certain subset of the results, but doesn't actually filter them. To take the "Ozeki run" query mentioned on your Examples page, one could then make a similar query, but for 4 basho, with the highlighting set for Basho #4 = rank Ozeki, to make it immediately visible who got promoted and who didn't. Or one could make a query that is limited to, say, Kotomitsuki's career results, and highlight all his make-koshi. Or a query limited to all M2e who got a 8-7, and highlight those who were promoted to sanyaku. Etc. etc.

I know a similar effect can, depending on the nature of the query, be achieved by re-sorting the output by one of the columns, but I usually find the chronological order more natural and more informative, so highlighting seems like a solution to that.

Maybe I'm the only one who'd find that useful though. :-)

No, that shouldn't be too difficult. Good idea. I have actually missed something like that myself.

I'll give it a thought.

[Zenjimoto 40]

Well done! Hey, do all you stats database guys realize that if you all had worked together we'd have a KILLER comprehensive database with super-query functions already since 2002? :-)

Since we are talking about wishful thinking in ideas for database queries... (Applauding...) Weeding thru various historical banzuke in preparation to fight in GTB, I would immeasurably invaluable find a function that allows you to find "any rikishi ranked at XX who scored YY in any catalogued basho. Would be even better to cross reference to find cases where there is a banzuke where rikishi A is at XX and scores YY while rikishi B is at ZZ and scores QQ (you know, that whole precedence research thing) :-) Anyways, one can dream...

Cheers

Zenjimoto

Edited December 9, 2003 by Zenjimoto

[Kashunowaka 300]

Since we are talking about wishful thinking in ideas for database queries... :-) Weeding thru various historical banzuke in preparation to fight in GTB, I would immeasurably invaluable find a function that allows you to find "any rikishi ranked at XX who scored YY in any catalogued basho.

You got it, back to 1973. (Applauding...)

Would be even better to cross reference to find cases where there is a banzuke where rikishi A is at XX and scores YY while rikishi B is at ZZ and scores QQ (you know, that whole precedence research thing) :-)  Anyways, one can dream...

That is on my "To do"-list.

[Asashosakari 20,220]

Would be even better to cross reference to find cases where there is a banzuke where rikishi A is at XX and scores YY while rikishi B is at ZZ and scores QQ (you know, that whole precedence research thing) :-)

[Zenjimoto 40]

You got it, back to 1973. :-)

:-)

[QttP 0]

Great site, Kashunowaka-zeki! (Sign of approval)

[Kotoseiya Yuichi 3]

:-) :-)

Oh, and I'd say this is probably the least bad subforum to post about this... (Applauding...)

Edited December 9, 2003 by Kotoseiya Yuichi

[Naganoyama 6,024]

Awesome :-)

[Kashunowaka 300]

Anyway...what I wouldn't mind seeing is, in addition to the constraining criteria, also the possibility of a "highlighting" criterion that only marks a certain subset of the results, but doesn't actually filter them. To take the "Ozeki run" query mentioned on your Examples page, one could then make a similar query, but for 4 basho, with the highlighting set for Basho #4 = rank Ozeki, to make it immediately visible who got promoted and who didn't. Or one could make a query that is limited to, say, Kotomitsuki's career results, and highlight all his make-koshi. Or a query limited to all M2e who got a 8-7, and highlight those who were promoted to sanyaku. Etc. etc.

Highlighting implemented. I used two of your examples for the examples page.

New shorthand notation under Basho constraints: you can enter kk or mk in the Wins field.

[Kaikitsune Makoto 210]

[Doitsuyama 1,259]

Only one question mark:

Mitsuruyama J10E 1978 Kyushu 5-9

Should that be 5-10, 6-9 or 5-9-1?

I have him as 5-10, as their was another source (not really sure?!?) But of course the second source could just be based on the first, you know... 5-10 or 5-9-1 are most likely anyway.

[Yoavoshimaru 0]

I just spent another half hour or so just browsing around, answering questions I've had for a long time... What an awesome site :-D

I didn't know Asashoryu only has one MK,, and a 7-8 one at that, as a sekitori, wow...

[Asashosakari 20,220]

Highlighting implemented.

Thank you once again. :-D

I used two of your examples for the examples page.

Hmm, are there any other Easter Eggs besides the possibility to enter 'Mickey' as a valid shikona? :-S

New shorthand notation under Basho constraints: you can enter kk or mk in the Wins field.

And this is a neat little improvement, too. (Blushing...)