90
75k
resolved Feb 13
100%50%
d5
11%
Nc6
0.6%
Ke7
23%
Nf6
0.6%
d6
0.6%
f6
0.8%
Qh4
0.4%
Ba3
0.7%
a6
0.5%
f5
0.0%
Bd7
0.5%
h6
0.7%
Be7
0.5%
Na6
0.5%
Ne7
0.5%
Qg5
0.5%
Bd6
0.7%
a5
0.7%
b6
0.7%
b5

I'm playing a chess game against manifold where the answer with the most probability will be your move, if there are multiple responses of the same answer then the one with the highest probability will be chosen, note that invalid moves will be ignored and the next highest one will be chosen.

Moves so far: 1. e4 e5 2. Nf3

Manifold: Black

Me: White

Game so far:


Get Ṁ600 play money

🏅 Top traders

#NameTotal profit
1Ṁ7,784
2Ṁ4,717
3Ṁ2,686
4Ṁ1,627
5Ṁ423
Sort by:

My takeaway from this is:

1) I now know the definition of "whalebait"

2) Unless I have a hell of a lot of mana, I am no longer going to participate in markets where the resolution depends on votes in the market itself.

@Tumbles I confess to being a user that didn't really understand what was going on, but my investment was minimal. I definitely do not feel like I deserve a refund, but I do agree the question (which to the naive user seems to be about chess) ends up being deceptive. I still do not understand how Bayesian (I am guessing this is a bot) made money while placing tiny bets on really bad choices (just throwing some extra money away? confusing other bots?), but I don't actually regret losing 10 mana to get a little exposure to what is going on.

@JensFiederer Bayesian isn't a bot! He was placing all those small bets in the hopes that one of them would get bet on and made to win by someone else. Most of them didn't make him mana

@Tumbles Then @Bayesian definitely has a lot more time for this than I do! Or is PARTIALLY a bot, a human using maybe a program on the side to read current standings and make plays that calculate the arbitrage? I STILL don't understand the wisdom of all the little losses, unless they were hoping somebody else would bid some of the little choices up a bit and they could sell out at an optimum time, with the assistance of something that could do the calculations for them.

@JensFiederer It only takes a few seconds to make the little bets once you decide to do them. The thing is that if an option is down to like 0.2 percent, a very tiny amount of mana can get you an enormous return if it works out. And you never know when somebody with a 300k balance will come along and make a random option win

@Tumbles Then part of what I did NOT understand was the way payoffs are calculated. Never really read up on that, did not realize what a big win it would be.

The market on which I REALLY lost some mana was the one about the Academy Awards, where I made what seemed like an utterly safe bet and still lost money even though I was right. So it looks like I have to look at TWO things: the actual truth and the current odds.

@JensFiederer ok so, I am not a bot and haven't placed any bot bets on this account. I can summarize how i understand payouts to work briefly.
Essentially, shares cost % cents, so if an option is at 0.5%, it means it costs 0.005 mana, which is very very little. if it resolves YES, from someone else betting it up at the last instant, it returns to YES holders 1 mana per share, so 200x of the original investment. If there are 20 of these "low probability options", all of which are very unlikely to win, if one of them wins, you still made 10 times your original investment by betting on all of them. So it's a risk with little downside and a fair amount of upside. Secondly, I had a lot of shares on NO for Nf6, which ended up resolving NO, which is where most of my profits came from.

In the end, in the option that resolved YES, I spent a net of 23 mana if I counted it right, for 2000 shares, so I made a ~1980M profit on that option, by identifying that I could just bet up a bit all the low probability options and hope one of them would turn a profit.

@JensFiederer That sounded strange to me so I looked up the market to see what happened. That market was kind of strange, in that the description said the winners would all resolve to 10%, and the losers would all resolve to 0%. So in that market it was never profitable to bet an option above 10% no matter what, because you know the highest thing they could resolve to is 10%.

I would definitely say the Academy Award market was strange and possibly confusing. They had a good reason to do it that way, but I can definitely see how you got confused. One thing to look out for in the future is that a linked multiple choice market like that can normally only pick one winner, so you know there must be something important in the description to explain how they will be able to choose ten winners.

@JensFiederer Maybe it would be a good idea for Manifold to have a flag for "advanced structure" for weird markets, something to warn inexperienced users to stop and think carefully before participating

@Bayesian I am convinced you are not a bot. Thank you very much for the explanation. Now, you don't have to respond to this question, and I don't think there is anything wrong with using computers to help (I certainly would if I used this site a lot), but you use some tools, right? Spreadsheets, maybe a python program or two? Or do you just click on that mouse really fast to make a lot of bets manually?

@JensFiederer for bets, I just click with the mouse; it's fairly quick. it would almost certainly be more work to use a python program. but also, if i did use one extensively for betting, i would need to change my account to be a bot account, which i'd rather not do. however I do use tools for getting alerted when a bad bet is made by someone else, to react quickly and exploit that price inaccuracy, so to speak. people also create fully automated bots to do bets like that. I don't really use spreadsheets, some people find use in them but i haven't so far (except maybe once or twice)

If there are new users here who have made a significant loss in this market due to its poor description and construction, you can contact me, and I'll take a look and do my best to reimburse you.

[Active = under 1 month, net worth under M$2,000 | the extent of the significant loss depends, I'll take a look and decide on a case-by-case basis | max reimbursement / user = M$1,000]

[This offer is only applicable to this market and is funded by my own Mana. I won't extend this in future markets, and it's just a gesture, as I personally believe new users might struggle to comprehend this market and that it wasn't adequately explained.]

@DracksathAudio

[I'm not a mod. I didn't bet in this market. I'm just writing this because I often find myself getting involved in things and I don't know why.]

This market got attention and sparked a heated discussion on Discord because new users clearly did not know how these markets work. In my opinion, this market doesn't explain the market very well. Essentially, it seems like a trap for inexperienced users, portraying itself as part of a chess game when, in reality, it functions more like a whalebait. This can be misleading, especially for those who want to engage in serious chess, leading to losses for new users and, in my view, unjustifiably sidelining them. This issue is not your fault or that of any whales. It is a problem with the site that has been criticized frequently.

As of now, mods won't intervene in this market unless requested by you. Therefore, it is up to you to decide how to address this situation.

Personally, I am inclined to suggest marking this market as N/A, given that some new users clearly misunderstood its nature. However, I also acknowledge the frustration of users like Tumbles, who played by the rules and may face disappointment in not receiving their deserved winnings. While @Tumbles might find a way to endure this situation the extreme frustration this scenario causes for new users could potentially drive them away from Manifold.

Regardless of your decision, I kindly request that you include in future markets a clear description explaining how these markets work, emphasizing their high risk and unsuitability for new users.

@Lion I am not aware of a single user that didn't understand how the market works when they made their bet. Will anyone step forward? The only bettor who I'm aware of complaining is Alyx, but Alyx 100% understood how the market works from the very beginning and said as much multiple times.

There is no basis to resolve this NA. The description is extremely clear, it says three times that the highest percentage will be selected. And once again, I am not aware of any traders that didn't understand this.

This is purely a case of sour grapes. Alyx understood how the market resolves when they bet, confirmed their understanding when people talked to them about it for the last few days, and now they are upset that the didn't win. It's completely inappropriate.

bought Ṁ1 Na6 YES

@DracksathAudio resolves to d5

opened a Ṁ1,500 d5 YES at 50% order

@Bayesian d5 :P

@Tumbles LOL my bad

@Tumbles You basically just stole everyone’s mana through market manipulation lol

@snarkyalyx

I spent 1k, not like I forced it

Once again, you seem to not understand the situation. From the very beginning, this market was about "which option will have the most mana spent on it". If you didn't want to lose from somebody spending more mana than you, you shouldn't have participated. It's as simple as that

@Tumbles You basically just controlled the outcome last second for your own advantage, I don’t see how that’s in good faith / fun. You don’t have to be a prick in a play money market lol

(I got a "error sending message on mobile, went to desktop, and the comments duplicated lmao)

(I got a "error sending message on mobile, went to desktop, and the comments duplicated lmao)

@snarkyalyx Alyx, you misunderstand how these markets work, almost the whole point of them is making a bet at the last second to outbid other people and make profits. people that bet based on what the best move was made a strategic mistake, this market was not set up to incentivize making good moves in any way, this was mentioned many times

@snarkyalyx In the future, make sure you don't participate in markets that resolve based on the market percentage. This is how all of them work. Stick to markets that predict things, or markets based on which option gets the most bettors regardless of size

resolve