Kozaru

Hey, I'll be playing Chain Gang for the first time this basho. I hope to crush you all.

Is the price of the rikishi just $1 for each point over the last 6 basho? Is that why Asashoryu is such a steal?

Hey all! I'll be playing Salarycap sumo for the first time this basho. Does anybody want to give me some newbie-pointers (Sign of approval...) , or are you all going to be selfish and keep those nice tips to yourself? :-(

Hey, for you guys who asked me about staggering start times of markets, I have a question. What times (please use Japan time, GMT+9) are best for you? I can't promise to accommodate you, but I'll see what I can do. I can't delay starts of "daily" markets, because they're so short, but anything that's longer than a day, I could maybe do that for you.

Hi, I'm Kozaru. I usually post on ummm... another.... sumo site, but I wanted to let all you wonderful Sumo Forum people know about a new kind of sumo game that I created. Just in case you haven't gotten enough of sumo-related games...... While looking for ways to bet on sumo, I made a new game which has just started for this tournament. It is modeled on the stock market. Players are given $5000 of virtual money, and they buy and sell "shares" in the outcome of a sumo-related event. It is different from the other games because you can bet on not only the outcome of matches, but also injuries, retirements, promotions, etc. It is also very hands-on, as trading can occur at any time, day or night, even between tournaments. I'll explain a bit about it. The URL is: http://sumo.inklingmarkets.com/ (The game uses an engine which was designed to make business decisions with market predictions, so some of the language used on the website might seem not suited for sumo, but oh well) As an example, there is a "market" set up for Goeido. You are asked if Goeido will receive a Sanyaku promotion between now and the July 2008 tournament. So you might buy a share in this stock at $50 (virtual money). As the tournament goes on, and Goeido has a 4-6 record, people will start to sell-off that share, and the price will go down. But if he then goes on to win the next 4 matches, and reaches 8-6, players might start to buy the stock again and the price rises. If Goeido does make it to Sanyaku by July, the stock closes at $100. If not, the stock closes at $0. You win or lose money depending on how many stocks you owned in Goeido. You can also "short" shares, meaning that you sell a share if you don't think the outcome will occur. For example, there is a poll for Ama vs. Hakuho. The title states that "Ama will beat Hakuho in March 2008". If you agree with this statement and you like the current price, you can buy. If you think the stock is over-valued, or that Hakuho will kick his little Mongolian butt, you can sell the share. Let's say you sell when the price is $45. If Hakuho beats Ama, you will win $45-$0=$45. If Ama wins, you will lose $100-$45=$55. It seems weird to "sell" something that you haven't bought, but that's how "shorts" in the stock market work. The website was just set-up, so right now there are only a couple of players, including myself. Come give it a look! The more players this kind of thing has, the more interesting it will be. There is no cost to play the game, and you can start trading immediately. The project is still in pilot mode, so I'd be interested to hear any feedback you might have. Also, if there's something you want to bet on which isn't there, please let me know by clicking the "Contact us" button on the page!

Ha! I was hoping nobody would call me on that one. Well, if no Makuuchi guys make 8-0, and a Juryo does, I'll probably I might (Weeping...) pay-off the stock. Call it a learning experience. I added a new rule in the rules section that all bets refer only to Makuuchi guys, unless I state otherwise. I'm hoping that either no Juryo guys will make it, or a Makuuchi guy does. Then the controversy will go away. (Applauding...)

I realise that I screwed-up in a way by making too many long-term bets. Everybody came in at the start, placed their bets for events that will be decided near the end of the tournament, and left. But I've started new "Daily" bets which payoff each day, and I invite people to come look at the site each day, and place some money on those markets. It's more fun that way, when the markets cash-out more frequently. Also, Martin has made a lot of money on some bets by putting all of his cash into one market. He's now got about $32,000 of worth, and when he bets on something, it completely crashes everything. So it appears that I will have to be resetting the accounts after each basho. Sorry if that's disappointing, but it seems that's a practical necessity.

Ha ha... I just made a test market, started it at 50% odds, gave myself $50,000 and put it all into that stock. I was able to buy 655 shares of the stock. When I was done, the price was $93.22. (Clapping wildly...) So the poor guy who is the next buyer has to buy it at a minimum price of $93.22. When the test event "came true", the shares were worth $65,571.81. So I now have $15,571.81 MORE to use to exert my influence. That's why I can't let people gain too much money. They can't be allowed to exert too much of an influence on the stocks. The game will fall apart if I don't occasionally reset the account.

Wait a minute. I remember something now. The game is set to default at $5,000 for new accounts. That default can't be changed. To make it $10,000, I had to manually add the funds. There is a reason why it can't be changed. They said that if one person has too much money, then that person can exert too much of an influence over the price of stocks. Imagine that after a few basho, there are or more people who have done very well (it's highly likely). They can then find a stock that they're interested in, and pump a lot of money into or out of it, dramatically increasing or decreasing the price. The next buyer then is kinda screwed, which is exactly the reason why some of you asked for staggered start-times above. One person could make or break a stock. (It can also be used to cheat, if a user has multiple accounts). The point is that the algorithm for determining the price of a stock is tuned to a $5000 starting value. If one user eventually has $100,000, then the algorithm is in big trouble. Can we figure out a way around that one? One way might be to cap how much one user can put into a stock, but the system has no option for that. It isn't practical to just ASK people to limit their spending, either. And if we cap how much can go into each stock, then that is really the same thing as resetting the accounts occasionally. It would result in account money which can't be used, because you've hit the limit on all available markets. For example, a cap of $10,000 per market x 5 active markets means that $50,000 would be a cap. You'd be able to see the money but not use it. But back to the point, the system has no option for a cap. One way might be to reset the value each Hatsu basho? Treat each year separately? That would allow new players a chance to not have to "fight" their way up the banzuke. Let's say that everything is reset in January each year, and the banzuke is completely re-done, based on the performance in November. So if I was a new user at Juryo 14 in November, but had the highest percent gain that basho, I would then be the top-ranked player for the new year. (with the exception of Yokozuna and Ozeki, who don't get reset?) Or, I could just go with my original plan and reset each basho. I can't allow a user to have $50,000 in one stock. What do you think? Damn, this is tough.

Itachi, so you're suggesting that I measure the percentage gain from one tournament to the next? That might work. One problem that I don't know how to resolve is that if I restart people when they lose their money, then they might put lots of money into long-shot bets. For example, say the chance of a rikishi besides Asa or Hakuho winning is 5%. If they put all their money into that bet and lose, then I restart them for the next basho. But what if it wins? They'll make 20x their money(!) and crush all opposition, having $200,000 at the end of the tournament. I'm afraid that people will be encouraged to make irresponsible bets if it will just be replenished next time. If we can resolve that problem, then I'll give that system some serious thought. One idea is to demote them to the absolute bottom of a banzuke if they lose their money. Comment? As for markets carrying over, I'm currently negotiating with the site Admin (Inkling) to not charge me for the off-months (non-basho months), in exchange for shutting down the markets during those months. So all markets would have to close at the end of a month, unless I can convince them to merely suspend all TRADING during those times. If they tell me to go F-myself and that I have to pay for all months, then I guess we can have markets always running (Clapping wildly...) So what we might be looking at is a system where money carries over, but all markets are closed during the off-months. I could measure the percentage gain from each player during that time, then move them up or down the banzuke based on that. Sounds like a decent idea. We're making progress. Oh, by the way, I figured out a way to stagger the market opening times, even if the site doesn't add an automated start feature. The system lets me nominate users who can make markets. They don't have admin power, only market-making power (they're called "market makers"). If I nominate a couple of people in Europe and a couple in North America, then they can open markets at times of day when I'm sleeping or whatever, and we can stagger the times. But my hope is that they'll make an automated feature.

Oh, and I like the idea of staggering start times as well. The system lets me set and automated end time, but not a start time. I'm going to mention it to the website dudes, or try to figure out another way to make it work without myself having to get up in the middle of the night to create markets (Clapping wildly...)

Gusoyama and Ganzohnesushi, Thanks again for the input. Unfortunately, due to practical considerations, we can't always get exactly what we want...... There are some problems with not having a start/stop time and a banzuke. If I don't reset the money, what do I do when people lose theirs? Tell them to leave the game (terrible idea)? Create a new account (bad idea)? Give them extra money, which is unfair to the others and provides no incentive to make wise bets (bad idea, and is actually a system which would encourage people to make the riskiest long-shot bets possible)? Resetting the money keeps everybody in the game. Maybe I could provide bonuses in the starting accounts for people who did really well the last time. Also, I thought having definite goals (stop times, banzuke, etc) would keep people coming back. If the game is continuous, then I can't really see the incentive to play past a couple of basho. As for the number of markets and the types of bets..... To be blunt, this service costs a lot of money. It was designed for businesses. I'm willing to pay for it out of my own pocket, to a point, because I like it. As the prices stand right now, to go from 5 markets to 10 will double my cost, which is past the threshold of what is reasonable. I'm going to try to negotiate the price down based on several factors, such as "We only need it active for 6 months a year". By having markets active between basho, I lose the ability to make such a negotiation. Therefore, long-term bets would be out. Also, the reasons I stated in the last paragraph make another argument for having a definite stop time. For the cost, I've thought about advertising, etc. However, advertising to the same group of only 50 or so people who come back every day is probably not going to work. It won't be a large enough number to attract banner ads, and it won't provide enough clicks for Google Ads to provide much revenue. I don't mind paying for it myself, if I can keep the cost low enough. Either that, or I can charge you all a fee, which is a terrible idea. Ganzohnesushi, don't worry about having not enough markets to bet on. Even with 5 markets, it should be OK. By making more daily markets, I can have a lot of markets to bet on, even more than now. 2 daily markets each day, plus some 2-3 day markets, and maybe 1 long-term market (yusho winner), I can probably have over 40 markets in each basho (roughly as many as I'll have this time with 15 markets)! Plus, having the markets cashing-out more frequently will make things more exciting. After day 2, I haven't cashed-out anything yet. That will change on Day 3 with the daily bets. Anyways, to have 10 markets active at a time will probably be way too expensive (depending on negotiations), so I'll do my best to work with 5, and still get over 40 markets into a basho. Also, if you want to make the game interesting by really seeing the prices move, it's good to have people's bets concentrated into fewer markets. The website's research recommends that 10 simultaneous markets is too many if you have 100 players. Right now, we have a bit less than 20 players, and 15 markets! So the key is to limit the number of markets which are running at the same time. Comments?

Here are some more of my thoughts after a couple of days. 1. In order to have a banzuke system, with an "opening" and a "closing" for each basho, I'm going to have to reset each user's cash accounts at the start of each basho. I'll make the banzuke and move people up and down based on their performance in the last basho. It wouldn't be fair to have people who lost all their money in the previous basho competing against those who made a lot, so all will get reset. 2. Because I'm resetting the account, the long-term bets (such as "Ama will be an Ozeki in March 2009") will have to go. All bets will have to be contained between Day 1 of the current basho, and the release of the real banzuke for the next basho. Otherwise, people can shelter their money in those investments to protect against the reset. I'm not happy about this one, but it's the only way that we can have a "start" and a "conclusion". Otherwise it would just go on forever, and it would be impossible to ever fairly make a banzuke. Unless of course, you have any other suggestions on how it might work.... 3. I think when the real thing starts, I might have to cut it down to 5 markets active at a time. The price of the service is quite a lot if I want more markets (going to try to negotiate). With 5 markets at one time, in order to keep things interesting, I'll dedicate 1 or 2 of them to "daily bets", (ie. bets for which the outcome will be decided during the action on that day). I'll also make some bets which last 3 days or so (ie. "Kotomitsuki will win just one match from Day 3 to Day 5"). Using this method, there can be a total of more than 30 markets run during the course of a basho. It also keeps the markets changing, giving people an incentive to come bet each and every day, while providing profits and/or losses each day. Feedback?

Hashira, good question. Today for example, I suspended trading during the Yokozuna's matches. That allows me to make sure that nobody can bet on the "First Yokozuna to lose" market with the outcome already known. If both Yokozuna win, I re-start trading. If one loses, I cash-out the market. It should not be possible for anybody to ever place a bet on such a market and be guaranteed to win. Plus, I'd like to ask people not to bet if I do miss shutting down a market (Laughing...) One thing that I'm going to have to do is stop bets like "Baruto will withdraw". That announcement could come as a surprise any day, or it might not come at all. I can't protect against it unless I suspend trading in the mornings before the kyujo are announced. Too much of a pain in the butt. That's why I'm running a pilot, to figure these things out. Jejima, the Admin decides the starting odds. The Admin does not have a trading account, although his best friend, Kozaru, does. :-) Now, I use some common sense when deciding the odds. I would probably not start a "Kaiho 12 wins" market at 50-50, more like 10%. Also, I notice that when I do buy, the market doesn't move much. The first buyer doesn't change the price all that much. It's only when people start piling on that the prices really shift. That's determined by an algorithm in the system that I have no access to and know nothing about. In response to your concern, during the "real deal", I will be pre-announcing the start of betting time for each basho, through a mailing list. I don't think I can stagger live times, except for manually opening them, which would probably be a pain. Other logistics have not quite been determined yet (which, again, is why we do the pilot first). Did you notice that I got a sample banzuke set-up on the homepage? :-) Please continue to give me feedback! I need to figure out the logistics of this thing once it's running "for real" (probably for July tournament).

Thanks for the feedback. I'm considering upping the starting amount of money for the next basho. Maybe we can try $8,000-$10,000 with 12 markets or so. In fact, why don't I just do that now? I'm going to add $3000 $5000 to all users' accounts right now. So spend some more! By the way, if you bought stocks and want to buy something else later, you know you can dump them at any time, right? You don't have to wait until the final result. For example, once I saw that Chiyo seems to be somewhat healthy on Day 1, I immediately dumped my position that he is going to retire. That's one of the best things about the game. It's dynamic, and you can change your decisions as new information appears, or as things start to look more or less likely as time goes on.

Well, that's a secret.... but I'll give you a hint. The name ends with "umoTalk". I'm the top poster over there, and they give me a lot of crap about it sometimes :-) So far, the game is getting a lot more attention here on SumoForum than it is over there. Actually, in the user profiles of the Sumo Market Game, I added a place to give your heya affiliation. You can fill it out to show your SumoForum pride. Then I can use it to shame those "umoTalk" people into joining to fight for their heya! (Welcome...)

That's funny, because after visiting TradeSports.com myself (I'm currently betting real money on its sister site, Intrade.com), I decided it was about time that somebody set this up for sumo. And you're right, the more players we have, the better it will be. 15 active markets might be too high for the number of players we have, though. We need lots of trading in each market for it to be worth while. If it doesn't work out as I'd hoped, I'll cut it down to 10 markets for the next basho.

OK, I'll add Kotomitsuki's final record if a slot opens up soon. Otherwise, I can do it next tournament.

It is only the admin who can make new markets. There is a check box for me to let others make them also, but at least for right now, I'm going to keep that control to myself. I want to limit it to 15 active markets at a time, and we've already reached that limit. However, I'm open to suggestions once some of the markets are decided and cashed out. You can make suggestions by using the "contact us" button on the page, or I suppose you can ask here, as you already have. What the hell is "guess the Kotomitsuki"? (Edit) I found the "Guess the Kotomitsuki" game on the website.... Wow, that would be tough to implement in the markets, and it would take up 7 of the 15 markets that I want to have active. :-)

Jeima, it is possible that the site does things that we haven't discovered yet. I'm still figuring it out myself. The game is based on a pre-made engine, so I don't have much control over the functions or appearance at the moment. I think I get some options for customization later on, but not sure. Thank you everybody for joining! Tell your friends! :-)