Kachikoshi

Oof. Seven falling out of Jonokuchi with MAYBE one going the other way. Add in only Mita joining and the next banzuke will have at least five fewer names on it. And that’s before counting any retirements.

In the spirit of the opening post, here's my attempt to rank the games based on how much advantage you get for checking the kyujo list every day before the deadline. This ranking is definitely more subjective than the one for missing a day. Here are the games in increasing order of advantage: Tipspiel offers no advantage whatsoever for checking the kyujo list because fusen bouts are automatically disregarded and point assignments automatically adjusted. Bench Sumo also offers no advantage assuming you check the "henka" box, and there's literally no reason not to check the henka box. In Chain Gang if your chosen rikishi is involved in a fusen match, the game automatically falls back to your default selection. It's not very often that this even occurs, and when it does, your default pick is probably pretty good anyway, so there is not much advantage to be gained here. In Odd Sumo if you pick someone involved in a fusen, that match automatically counts as a win but also the quota is set to one, meaning you earn no points from picking it. However, you're required to pick at least three winners each day you enter, and picking someone involved in a fusen effectively means you only need to pick two that day. Since the odds are stacked against you in this game, picking someone involved in a fusen means you can gamble a little less, which gives you a small increase to your expected final score. In ISP, if you know someone in the chosen bout is kyujo you get a guaranteed win. But because ISP only uses a single bout each day, this is a fairly rare occurrence (though it did happen this tournament). Sekitori-Quadrumvirate lets you pick fusen winners. It is only slightly discouraged by penalizing you in the sansho standings and way down the list of tiebreakers. However you will frequently be in the situation that you already chose the winner by fusen earlier in the tournament and thus can't take advantage, and even if you do take advantage, you still need 2/3 of your remaining picks to win in order to claim a victory that day. Sumo Game counts fusen wins as ordinary wins. If you're paying attention, you can not only claim a free point, but you can also place him at the top of your list, giving you a high chance of winning any potential tiebreaker. It's very difficult for a player who didn't react to the kuyjo announcement to beat an opponent who did. Sekitori-Toto also counts fusen wins as ordinary wins. Getting an extra point is frequently the difference between claiming a win or a loss on any given day, and it comes up more often than in Sumo Game because Sumo Game only cares about Makuuchi. In my opinion Turn the Tide offers the biggest advantage to paying attention. Turn the Tide lets you pick any sekitori, so fusens come up frequently, and when they do most of the time either the winner has a losing record or the loser has a winning record, so you'll be eligible to pick one of the two. Of course you still need to get your other pick of the day correct to earn a win (unless there are multiple fusens that day!).

I assure you I did not start this thread to get into a prolonged discussion on potential minor changes to Tipspiel. I’m well aware that a game which hasn’t changed in decades, which concerns a sport which hasn’t changed in centuries, and run by people from the world’s most prolific user of the fax machine, is not going to be very open to reform.

Actually I took a closer look and I believe I’ve understated the difference. In the Tipspiel Nagoya basho group 1, the average daily score per participating player was 7.764. So if you missed an average day you would be expected to lose out on 7.764 points. But you only needed 3.385 points to get an extra win, so missing a day actually would cost you about 2.3 wins (I originally estimated 1.5). If we estimate that droning in Bench gives you a 25% chance to win (compared to 50% if you played properly) then droning for a day penalizes you 0.25 wins, or about a ninth of the Tipspiel penalty. To be fair I believe this example presents an especially big penalty in Tipspiel and that other groups/bashos penalize absences less. So 6:1 is probably about right.

The obvious solution to this problem is that default points only count in calculating your final record, not in determining who wins the yusho. You could go even farther and add a rule that default points can only help get you to a 7-8 record but not beyond that, to eliminate any possibility of abusing them to ensure a kachikoshi.

The fact that you’re even entertaining droning as a reasonable strategy at any level of Bench shows how light the absence punishment is, relatively speaking. In a game like Quadumvirate or Turn the Tide “droning” isn’t even a thing and absences are treated as guaranteed losses. I’d estimate the negative effect on your record of missing one day of Tipspiel is about equal to droning six days of Bench. Giving guaranteed points definitely feels wrong, I agree with you there. But the default score is set at exactly the point that if you did it every day you would wind up 0-15; it’s not a very good score. I’d estimate the win penalty for autopicking for a day in Tipspiel is about half the penalty of my “default score” proposal. You could make the same argument about the kosho system. If I try my best at Tipspiel and do badly I’ll get demoted way down the banzuke. But if I don’t try at all and just declare myself kosho I get complete rank protection. No penalty whatsoever!

The five day limit barely even matters in Chain Gang, since most people are knocked out by then. As for ISP and Tipspiel, let’s compare them to a real life sekitori. If he missed seven days, he would be expected to win about half of his remaining bouts and finish with four wins. In ISP, you would auto pick for five of the seven days and get about two wins, then get four wins from the days you participated, for a total of six. In Tipspiel, if you miss seven days (and don’t use the non-obvious generator feature), you’re going to finish with zero wins. In the unlikely event there’s ever an appetite to amend the Tipspiel rules, here’s a suggestion: if you’re absent for a day, you get a “default score” equal to the average score for your group that day minus half the “step size” between each win threshold at the end of the tournament. For example, if you miss day 6, and the average score for your group that day is 8 points, and then the step size at the end of the tournament for your group is 6 points (such that every 6 points would have earned you an extra win), then your default score for that day is 5 points. That might not seem like a big penalty but it’s equivalent to half a win, which would still make Tipspiel tied for the most punishing game on this list!

Well at least in comparison to Sumo Game. In Bench Sumo everyone has a Yokozuna, Ozeki, and Sekiwake and you basically want all three in your lineup every day. That just leaves two of the slots, and if you should auto pick someone who withdraws you’ll be saved by the henka feature, where if you did that in Sumo Game you’ll just get dinged with a missed pick day after day. But if anyone has hard numbers on the relative success of auto pickers in Sumo Game vs. Bench Sumo I’d love to see them.

For the first time this basho, I started playing in all nine daily Superbanzuke games. Some of the games I only joined in mid-basho and I found it interesting how some games punish missing a day much more than others. I've attempted to order them below from most to least punishing. I have no real point to this post other than that I thought someone might find it interesting. Tipspiel punishes absences the most by far. Because W-L records are based on comparing your total score over the basho to the average of all players who participated, and even a poor showing on any given day is bound to net you at least some points, each day you miss translates to roughly 1.5 more loses than if you had played and performed in an average way. If you miss half the basho you're basically guaranteed to finish 0-15. Sekitori-Toto is the most straightforward. Not playing a day nets you a loss-equivalent, and exactly half the players win/lose each day, so missing a day gets you an expected 0.5 more losses than if you had played. This is the most similar punishment to an absence in real-life Ozumo. Turn the Tide is similar to Sekitori-Toto, except that in general fewer than half the players win on any given day, so even though missing a day gets you a guaranteed loss-equivalent, you had more than a 50% chance of losing anyway. In Sumo Game if you don't send in picks your default lineup gets used. You'll probably lose but you have a reasonable chance to get lucky and win. Sekitori-Quadumvirate delivers you an automatic loss-equivalent for not showing up, but because of the unique rule that you can only choose each rikishi once, an absence means you don't use up any of your pool of rikishi which makes the rest of the tournament easier. Bench Sumo lets you fall back on a default lineup like in Sumo Game, but the nature of the game means you don't have very many choices to make each day and you wind up trotting out pretty much the same team every day regardless. Plus the "henka" feature which automatically moves rikishi involved in fusens means that not paying attention is punished even less. In Chain Gang, assuming you made decent default picks, you're going to end up with someone pretty solid regardless of whether you study the particular day or not. This is less true the later it gets in the tournament, but most people are knocked out by then anyway. With ISP, you're just picking the winner of one evenly-matched bout which is pretty much a coin toss anyway. So long as you have some kind of auto-entry formula set up your chances of winning are almost as good as anyone else's. Finally Odd Sumo is the kindest to absent players. The nature of the game has you gambling points at unfavorable odds. The more you play the more points you should expect to lose. So long as you make the required number of bets, which can be done in as little as five days, you're actually rewarded for not playing. As a relative newcomer to these wonderful games which have been running smoothly for decades, I don't expect anyone to care much about my opinion. But if I were to offer it anyway, I'd say that Tipspiel is too harsh on absent players while Odd Sumo is too generous, and that modest changes to how the games operate could fix this "problem."

I wonder if this is a strategic move. With so many rikishi in Jonokuchi now using the strategy of entering the tournament for only the final bout, if Sananojo does the same he will likely have to face someone who is fairly strong. Whereas if he instead joins for the sixth bout, he can face a bonafide 0-5 opponent.

I’ve been trying to register an account on sumofan.net in order to play Tipspiel, but I keep getting rejected for having given an “incorrect answer.” The security question asks for “Hakuho’s rank” and I’m not sure what answer it’s looking for but it doesn’t appear to be Yokozuna.

I’ve been trying to register on account on sumofan.net in order to play Tipspiel, but I keep getting rejected for having given an “incorrect answer.” The security question asks for “Hakuho’s rank” and I’m not sure what answer it’s looking for but it doesn’t appear to be Yokozuna.

I'm very curious to know how you did this. I assume given the vast amount of data that you did it programatically. How did you decide how many to people to put in each division?

I have an outrageously inconsequential topic to discuss concerning what happens if exactly one person enters maezumo. For background, to those who are unaware, here’s how the Jonokuchi banzuke is ordered 1. Demoted Jonidan rikishi and Jonokuchi rikishi with at least one win 2. Maezumo rikishi with at least one win 3. Jonokuchi rikishi with zero wins 4. Maezumo rikishi with zero wins If only one person enters maezumo, he won’t have anyone to fight, and presumably would go in category four i.e. he would be dead last on the banzuke. This seems sort of unfair as there could be several zero win rikishi he will be behind but could have been ahead of had there been someone for him to fight. I’ve felt for some time that maezumo is an unnecessary relic and simply slows down the rise of promising newcomers.

I suppose I could offer some help here. I’m a professional software engineer. I used to run a soccer/football game for five years, until factors outside my control forced me to shut it down this year. So now I have a void that needs to be filled. I know first hand how tedious running these sorts of games can be if you don’t do a ton of automation. If anyone is running a sumo game and would like some help with the programming side, please reach out. I’m going on paternity leave in October so I’ll have time for a personal project.

I don’t play Kachi Clash, so correct me if I’m wrong, but I thought their “banzuke” is not really a banzuke in the sense that you move up and down based on your rank and record. Rather it’s just a table of high scorers over the last six basho, i.e. the same thing as GTB’s “top ten” they just call whoever is in first place at the time “Yokozuna East.”

Hello! I made an account and would like to join for the next basho.

No, but I bear many similarities to Morimoto Yusuke, the main difference being athleticism.

I’ve been playing GTB under the name “Midoriyama.” Is it possible to change it to Kachikoshi to match my forum name (+ my name in all other games)?

I’m in.

Because Daieisho is ranked higher, he would only need to equal, not exceed Hiradoumi's record

I forgot to enter this tournament. Is it possible for me to be marked kosho?

Terunofuji (Ye 0-2-13) Y --- Kotozakura (O1w 11-4) O1 Hoshoryu (O1e 10-5) --- O2 Takakeisho (O2e 0-2-13) Abi (Sw 10-5) S1 Onosato (Kw 12-3 Y) Kirishima (O2w 1-6-8) S2 --- Daieisho (M1w 11-4) K Hiradoumi (M2e 9-6) Meisei (M5w 10-5) M1 Atamifuji (M1e 7-8) Wakamotoharu (Se 4-8-3) M2 Mitakeumi (M7w 8-7) Takayasu (M3e 7-3-5) M3 Gonoyama (M2w 6-9) Tobizaru (M3w 6-9) M4 Ura (M4w 7-8) Onosho (M5e 7-8) M5 Oho (M4e 6-9) Shonannoumi (M10e 9-6) M6 Takanosho (M8e 8-7) Kotoshoho (M8w 8-7) M7 Sadanoumi (M11e 9-6) Ryuden (M14e 10-5) M8 Oshoma (M14w 10-5) Kinbozan (M10w 8-7) M9 Tamawashi (M9e 7-8) Shodai (M9w 7-8) M10 Ichiyamamoto (M12e 8-7) Midorifuji (M6w 5-10) M11 Nishikigi (M7e 5-10) Churanoumi (M13w 8-7) M12 Hokutofuji (M11w 7-8) Asanoyama (Ke 0-0-15) M13 Takarafuji (M16w 9-6) Wakatakakage (J6w 14-1 Y) M14 Endo (J3e 12-3) Chiyoshoma (J3w 12-3) M15 Kagayaki (J5e 11-4) Roga (M15w 7-8) M16 Bushozan (J2e 9-6) Nishikifuji (M12w 5-10) M17 --- I may have demoted Asanoyama too much, but the banzuke committee has not been showing much leniency with absent rikishi lately, and to place him any higher would require an overdemotion of Hokutofuji. I realize last basho Nishikigi went ahead of Tamawashi, casuing him to get an overdemotion, but Nishikigi was only one rank beyond Tamawashi by the numbers (as opposed to Asanoyama being 2.5 ranks behind Hokutofuji) and Nishikigi fought all 15 days. Also I didn't go for it, but I wouldn't be totally surprised if Gonoyama maintained his rank of M2w on a 6-9. I realize no demotion for a 6-9 is unprecedented in Makuuchi, but they've done it several times the last few bashos in Juryo, even though they could have easily avoided it.

This was not "bleedingly obvious" and was, in fact, rather observant and informative. Thanks!

This is a bit off topic, but is there a “formula” for the Makushita joi, like how in Makuuchi/Juryo you move up/down by wins minus losses? The Makushita joi rankings have always seemed way more arbitrary to me.