Test Run: Makushita Game

[Randomitsuki 2,843]

Here is Itachi's dream:

I suggest a test run for a Makushita Game starting this Nagoya. If there is a considerable number of players (which I somehow doubt), and if the game mechanics are realistic in terms of results (which I hope) I could turn it into a regular game starting Aki Basho.

The rules for the test run are as simple as can be.

1) Slice the Makushita division in 15 equal portions (i.e. ms1-ms4, ms5-ms8, etc.).

2) From each slice name exactly one rikishi who you think will have the best record as compared to the other rikishi in the same slice (ms15Td would be in the ms13-ms16 slice)

3) If no other rikishi in the same slice has a better record than the one you picked, you will receive one win. I.e., the picked rikishi must at least be equal to the top rikishi in that slice in order to hand you a win.

4) With 15 slices to guess, you can get a 15-0 score.

Please send your entries for this test run either via PM or e-mail.

Submit your 15 rikishi until shonichi 9AM JST.

[Zenjimoto 40]

I must admit, when I read the title, I was more than skeptical, but seeing the game idea. I like it (Applauding...) Count me in on this one (Clapping wildly...) (surprise!!) :-)

Cheers

Zenjimoto

[Asashosakari 19,648]

My first thought was, "no way is anybody going to get better than 10-5 in this", but let's give it a whirl. I'm in. :-)

Edit: Random guessing has an expected value of about 2.5 to 3 wins in this game, yay. For comparison, the notoriously difficult Seki-Quad has about 4 to 5 expected wins for random picks... (Clapping wildly...) (Applauding...)

Edited July 2, 2005 by Asashosakari

[Itachi 396]

Thanks Randomitsuki!

I can't wait to get started. I'll have an entry for you soon.

I hope other lower division followers will pick up on this.

Itachi

[Kintamayama 45,507]

My Makushita site is offered as an aid to those who do not follow this division. It supposedly has direct links to every rikishi's stats.

http://www.dichne.com/maku.htm

[Zeokage 0]

Rotosumo for Makushita (at least in the way of selecting)... definately give this one a try. Then again, I give any game a try don't I? (I am not worthy...)

Edited July 3, 2005 by Zeokage

[Itachi 396]

My picks are in!

Blocks 41-44, 53-56, and 57-60 were the toughest - as in no one particularly standing out to be picked.

I think almost everyone will have the same pick for the 5-8 block but there is an interesting alternate choice.

There are also two very tempting options in the 21-24 section.

I hope I can make kachikoshi.

Itachi

[Itachi 396]

I just wanted to keep this topic near the top so all would be participants get a reminder when they visit this sub-forum.

Also, I wonder if anyone has heard about Wakakirin. He miised all of last tournament with some sort of injury. What was it? Is he ok now?

How is Gojoro? He also dropped down. I wonder if he's in free fall or if he's going to bounce back to Juryo in a few tournaments again. It's been a while since his last comeback to Sekitoridon. Him and Wakanojo dropped way way down and Gojoro came all the way back to yoyo between Makuuchi and Juryo for a while as I recall....

I have selected too many foreigners in my Makushita picks. I tend to notice them a bit more as they move up to this division but at some point, many of them slow right down and drift around in Makushita/Sandanme for a while. Does Nakanokuni still qualify as a prospect or will he just yoyo around well below Juryo until giving up and returning to China? There are several Monglolians who don't seem to mind treading water at these ranks. At what point do we stop expecting them to make a run towards the salaried ranks?

There are also a few former Juryo guys who never did all that well as sekitori. Masutsuyoshi was never all that great but he did spend a lot of time in and out of Juryo. Is he done?

What about Kotonomine and guys like that? I never thought he'd get there in the first place but he did and he didn't last long. Should we expect to see him up there again?

As much as I'd like to pick my long time adoptee Nishikikaze, I can't say I expect much from him at his current level (in the 30s). Also there is a pretty good prospect in his bracket as I recall (it's not in front of me now).

Well, I'd like to hear a bit more chatter about this division. I hope Randomitsuki's efforts won't be in vain - with no one participating.

Itachi

[Randomitsuki 2,843]

As of now I have received guesses from five players (including myself). It's nice to see some variability in the picks so far. There is no bracket where everybody agreed on a particular rikishi, and only few brackets with two different picks (one of them, you named it, is the ms5-ms8 portion). The largest number of different picks per bracket is four (there are several instances). I recall that the last bracket ms57-ms60 was quite variable. I for one did not really see anybody standing out from that banzuke region.

[Asashosakari 19,648]

  Itachi said:

I have selected too many foreigners in my Makushita picks. I tend to notice them a bit more as they move up to this division but at some point, many of them slow right down and drift around in Makushita/Sandanme for a while. Does Nakanokuni still qualify as a prospect or will he just yoyo around well below Juryo until giving up and returning to China? There are several Monglolians who don't seem to mind treading water at these ranks. At what point do we stop expecting them to make a run towards the salaried ranks?

My personal rule of thumb is that a rikishi ought to make it to the top of Makushita by age 25, or he probably won't amount to much. Might still make it to Juryo eventually, but an extended run as a sekitori or even a further promotion to Makuuchi are unlikely.

By that guideline, only two foreign toriteki are out of the running for future success, namely veteran and new Japanese citizen Kyokutenzan, and Taika (who turns 25 this month and has never been higher than Ms52). Nobody else is older than 23, and all could well mature into a future Makuuchi rikishi if things go their way...even somebody like Hakuba who's been bouncing around mid-Makushita for two and a half years shouldn't be counted out, since the rise to Juryo can sometimes happen surprisingly quickly.

Though, if I had to name some others who I think are lagging behind the necessary improvement curve for a future sekitori, I'd have to name Haku (Kokonoe's Chinese guy who's had injury problems, but still shouldn't have this much trouble to get out of Jonidan, considering he's 22), and the two Tongans at Musashigawa-beya...they're also only 22, but they really should be capable of hanging with the lower Makushita/top of Sandanme crowd at this point, and neither of them is. Still, maybe they're both late bloomers...we'll see in a couple of years. (I'm pretty sure Haku won't go anywhere though, except intai.)

[Zeokage 0]

  Asashosakari said:

By that guideline, only two foreign toriteki are out of the running for future success, namely veteran and new Japanese citizen Kyokutenzan, and Taika (who turns 25 this month and has never been higher than Ms52).

<{POST_SNAPBACK}>

Shikihide-beya's rikishi, right? If you're right it would be a shame, he's a nice guy. Even took the effort to speak with me in English when I visited the heya.

Then again, I suppose most rikishi would be nice guys.

[Feginowaka 53]

  Itachi said:

I just wanted to keep this topic near the top so all would be participants get a reminder when they visit this sub-forum.

...

Well, I'd like to hear a bit more chatter about this division. I hope Randomitsuki's efforts won't be in vain - with no one participating.

I used to make a Banzuke for Makushita until Kyushu 04, but now I don't have the time anymore ;-)

There are always many interesting Rikishi in that division, which makes it worth having a look at those guys.

  Itachi said:

Also, I wonder if anyone has heard about Wakakirin. He miised all of last tournament with some sort of injury. What was it? Is he ok now?

Normally he should easily win 6 bouts at his current rank, but he missed the last 2 basho and I don't know if and when he will return to the dohyo. :'-(

  Itachi said:

I have selected too many foreigners in my Makushita picks.

I only selected 5 of them :-)

There are some very interesting blocks in the Makushita-Game:

1-4 is always the most interesting as there are normally several promotion candidates at the top of Makushita.

5-8 has some interesting guys as well with 2 highschool-Yokozunas and 2 promising foreigners...

25-28 has no rikishi that will end up mk and has 2 future Makuuchi rikishi

33-36 also has no rikishi that will score less than 4 wins

37-40 was the easiest as there is only 1 rikishi who is able to win 5 or more and this block is also known as the "mongolian block" with 4 mongolians!

49-52 was a difficult one to choose from as almost everyone is able to win 5

[Asashosakari 19,648]

  Feginowaka said:

25-28 has no rikishi that will end up mk and has 2 future Makuuchi rikishi

33-36 also has no rikishi that will score less than 4 wins

<{POST_SNAPBACK}>

I dare say it's practically impossible that 8 consecutive Makushita rikishi are all kachi-koshi. (Let alone two times eight in the same basho.)

Edited July 6, 2005 by Asashosakari

[Randomitsuki 2,843]

  Feginowaka said:

There are some very interesting blocks in the Makushita-Game:

1-4 is always the most interesting as there are normally several promotion candidates at the top of Makushita.

5-8 has some interesting guys as well with 2 highschool-Yokozunas and 2 promising foreigners...

25-28 has no rikishi that will end up mk and has 2 future Makuuchi rikishi

33-36 also has no rikishi that will score less than 4 wins

37-40 was the easiest as there is only 1 rikishi who is able to win 5 or more and this block is also known as the "mongolian block" with 4 mongolians!

49-52 was a difficult one to choose from as almost everyone is able to win 5

<{POST_SNAPBACK}>

Interesting predictions. I certainly won't bet that you are wrong, but your predictions are not really in line with what the others thought.

From your observations one could expect that players picked many different rikishi in the blocks 25-28, 33-36, and 49-52, whereas picks should be quite focused in block 37-40. In fact, the opposite is currently true. In the "difficult" blocks we have some focused picks (e.g. 60% of players picked one rikishi in 49-52), and in the "easy" category 37-40 six different picks were already made.

As of now, we have ten players, and four more have announced their participation.

[Susanoo 389]

I think that I cannot fully enjoy Sumo, without turning one's eyes to lower division.

Although I watch only Torikumi of Makushita higher rank, the trend of whole lower division is always imposed on mind.

Therefore, it is time very significant for me, and pleasant to choose pic of this game.

About this simple clear rule, I can think that I do not need to test.

For example, as the 1st "幕下ランド" connected with your name, even if it holds suddenly, I think that I am satisfactory at all.

A difficult problem is deciding the standard of Sansho.

Although it is the problem which cannot be judged that a game does not finish this, the rule is once determined and changing it at the time of the next basho will be admitted.

[Yusho]

This does not have complications, either and is the quantity of a best choice, i.e., a gamer's number of victories.

Since a tie breaker corresponds to it, I think it desirable to make a gamer choose about three Makushita rikishis in advance regardless of classification(1-4 and 5-8 ...).

It is because the reason determines three more prizes in addition to this, so I want to save anticipation of classification.

[shukun-sho]

Selection of Makushita-Yusho-rikishi.

Tie-breaker 1: More selections of 7-0-rikishi.

Tie breaker 2: More selections of 6-1-rikishi.

When three or more gamers are in addition by this, "withhold Sansho become none of entries deserve(該当者無し)" is specified.

The gamer who gained Yusho is not elected by the tie breaker, but naturally he himself's award is possible.

[Kanto-sho]

The total number of victories of selected rikishis.

Tie-breaker : The number of selected rikishi's KK.

the gamer who gained Yusho only when it exceeds other gamer with the highest point in except 10% or more, a duplication award is possible with the gamer.

[Gino-sho]

Acquisition of the classification which had few hit gamers.

When three classification is compared and three or more gamers exists in addition, "該当者無し" is specified.

Both gamers can be awarded when two gamers is at the time.

The gamer who gained Yusho is not elected by the tie breaker, but naturally he himself's award is possible.

Naturally a gamer's KK are conditions all the Sanshoes.

You don't have to rush to determine all rules now.

A rule is completed being readjusted.

I am glad if this becomes your reference in that case.

Moreover, you do not need to judge objectively about all the rules.

I welcome also making your subjectivity reflect.

I will continue participating in this game.

As a result, I will love Sumo more deeply.

(Clapping wildly...)

Edited July 7, 2005 by Susanoo

[Randomitsuki 2,843]

Great suggestions by Susanoo! Thank you! They will certainly be considered for implementation.

While the sansho are added value to the game concept, they could be discussed during the basho. However, we should agree on a yusho tie-breaker beforehand.

Susanoo's idea was to name three rikishi in advance. Given that I have already received many picks, it would be a little inconvenient to ask players again to submit tie-breaker picks. Therefore I would suggest that in case of a tie for the yusho the order of wins by descending banzuke bracket will decide.

Example:

Yusho candidate A had wins in the first two brackets (1-4, 5-8) and a loss in the third bracket (9-12).

Yusho candidate B had wins in the first and third brackets and a loss in the second bracket (5-8).

Then Player A would receive the yusho because of the second bracket.

Edited July 7, 2005 by Randomitsuki

[Doitsuyama 1,192]

I have sent my entry with PM and my observation is: It should be REALLY hard to get a KK. Maybe the rule for getting a shiroboshi in a bracket should be somewhat easier like 5-2 is enough even if not best in the bracket. Or another suggestion: KK gives 0.5 points, bracket win a full point (can lead to awkward final scores of course). Or: Being among best TWO in a bracket gives a point.

Well, the test run should provide good material to experiment with the scoring. I like the format of selecting one rikishi out of 15 brackets.

Oh, one last point: Scoring could be made just by number of wins of course (like in PTYW). Maybe the most fair way of scoring.

[Asashosakari 19,648]

  Doitsuyama said:

Oh, one last point: Scoring could be made just by number of wins of course (like in PTYW). Maybe the most fair way of scoring.

<{POST_SNAPBACK}>

Well, as long as we're offering scoring ideas...I've been wondering how scores would look if players only need to be within one win of the bracket's top results, not exactly match it.

Checking the first three basho this year, random picking would have scored an average expected 6.148 wins, which may be a bit high, considering that informed picks are going to be better than random ones. But it looks more viable than the current rules (average expected: 2.699 wins).

FWIW, one could use a mix of both regimes, e.g. if only one rikishi scored the bracket's top result, it's enough to be within one of his win count, but if two or more got the top record, you need to have picked one of them. Three-basho average expected for that has been 4.991 wins.

[Randomitsuki 2,843]

Houston, we have a problem. Going by gut feeling I clearly overestimated the frequency for more than one rikishi winning in a bracket (that frequency is actually quite low). Testing for the last three basho I came up with similar results as Asashosakari (the difference probably being that I discounted kyujo rikishi).

Question to the experienced gamemasters: At what average score will banzuke making become difficult? For instance, what is the average score in S4?

My gut feeling is that an average score of 6.0 would be acceptable. The question is if we are good enough to get into that area.

So it looks as if we'll have to use some of the scoring tricks that Doitsuyama and Asashosakari have suggested. Similar tricks could involve standard deviations from bracket mean etc..

Anyway, it would take away from the simplicity of the game which is a pity.

Edited July 8, 2005 by Randomitsuki

[Asashosakari 19,648]

  Randomitsuki said:

Houston, we have a problem. Going by gut feeling I clearly overestimated the frequency for more than one rikishi winning in a bracket (that frequency is actually quite low). Testing for the last three basho I came up with similar results as Asashosakari (the difference probably being that I discounted kyujo rikishi).

Yes, I just did a quick-and-dirty check to come up with those numbers, your methodology should be more accurate. (I can hear the voice of my high school math teacher now: "Why the heck did you pretend for accuracy then by giving the results with three digits?" Er, umm...)

Edit: I do think the game can be made viable with few modifications (we'll just have to see what they will be)...fortunately, the game's format makes it possible to apply lots of different scoring ideas retroactively, once the first run has given some real-world picks to work with. (Showing respect...)

Edited July 8, 2005 by Asashosakari

[sekihiryu 51]

  Feginowaka said:

33-36 also has no rikishi that will score less than 4 wins

37-40 was the easiest as there is only 1 rikishi who is able to win 5 or more and this block is also known as the "mongolian block" with 4 mongolians!

<{POST_SNAPBACK}>

33-36 was rubbish agreed, but dont be so hasty with 37 - 40 call there are 2 guys capable there not just one. (Clapping wildly...)

[Yubiquitoyama 4]

  Asashosakari said:

  Feginowaka said:

25-28 has no rikishi that will end up mk and has 2 future Makuuchi rikishi

33-36 also has no rikishi that will score less than 4 wins

<{POST_SNAPBACK}>

I dare say it's practically impossible that 8 consecutive Makushita rikishi are all kachi-koshi. (Let alone two times eight in the same basho.)

<{POST_SNAPBACK}>

Of course such a statement begs for a bit of checking ;-)

Actually, I can't see two 8KK-rikishi streaks happening ever, but the odd single one probably will happen occasionally. I did a semi-manual check of 12 basho (2001.09 to 2003.07) and it turns out there actually was one 8-rikishi KK-streak (2002.03: Ms47e to Ms50w) plus one 7 KK-rikishi streak. It won't happen often but it actually happens at least...

[Randomitsuki 2,843]

Here are some selection stats for the Makushita Game.

First of all, we have 23 players which is much more than I would have expected. So thanks for participating.

From 120 Makushita rikishi 93 were picked which is quite a lot.

The most popular rikishi per bracket were:

ms1-4: Shiraishi

ms5-8: Baruto

ms9-12: Bushuyama

ms13-16: Dewaotori

ms17-20: Kotokanyu

ms21-24: Tamarikido

ms25-28: Takanoyama

ms29-32: Nakanokuni

ms33-36: Gokenzan

ms37-40: Koryu

ms41-44: Kasugakuni

ms45-48: Saita

ms49-52: Wakakirin

ms53-56: Kotonomine

ms57-60: Shibuya

It is an interesting question if successful sumo gaming means going with average picks. In other words, do players who pick rikishi that many other players pick are better than players with unusual picks (at least in games that do not rely on single bouts like ISP)?

If sending an average pick is any indication of good play then we have a clear favorite for this inaugural basho, viz. Feginowaka. Each of his 15 guys was picked by almost seven other players on average. Other players with exceptional picking overlap are Doitsuyama, Itachi, Kintamayama, Takamueda, sekihiryu, Yubiquitoyama, and Zenjimoto. Soon we'll see how this turns out to be.

I will post updates on the standings every two days during the basho.

Good luck, and lots of fun!

[Kintamayama 45,507]

  Randomitsuki said:

It is an interesting question if successful sumo gaming means going with average picks. In other words, do players who pick rikishi that many other players pick are better than players with unusual picks ?

<{POST_SNAPBACK}>

I tested this theory a few bashos ago as I was curious. Paper Oyakata used to allow you to see others' picks, so I went for the pick with the most votes, i.e the most picked rikishi for each bracket. I would have gone 7-8.

[Itachi 396]

I am thrilled with the response to this game. 23 is more than I had hoped for too.

I enjoyed browsing through the Makushita results last night while the Juryo bouts went on. Usually I end up crashing before the days matches are over and have to wait until morning to check my game results but now I had 15 little races to check on in the third division. Good fun.

As for the average pick thing, I think it's important not to miss the 'obvious' good picks that most players will know about so there will be a few that all good players will have. Then, there will be some dark horses - rikishi who have gone unnoticed by most but are in a great position to win. It's better not to be with the crowd in those brackets.

I never played Paper Oyakata so I can't respond to that comment but I feel I'm in good company when I look at the names of players who also often picked with the popular choices.