@predyx_markets please resolve, thanks.

There’s still a few hours left

ooo, did someone make an arbitrage bot?! I was thinking about doing that! That's exciting!

Totally fine to keep it as trade secrets, but I'd love to know:

- How do you specify the arbitrages? Can you point to various markets and say "this should go up when that goes down"?
- Does it have any complicated management (eg, spend more on these opportunities than those, based on time to resolution or expected gain per mana or something) or do you just have enough mana that you're not worried about that?
- Does it handle partial arbitrages (eg, I want to buy so that I am over-exposed to this side of the arbitrage) or only exacts (buy exactly this fraction of yes on this market and no on this market, whenever the prices diverge)?

I'd love to chat about if you want!

it is 73k euro

@1000mana title and description says $

@predyx_markets also UTC vs. PT close, thx!

@deagol done

For people who like to arbitrage: https://manifold.markets/DannyqnOht/btc-first-falls-below-80k-in-decemb

Whoever that was that just bid up to 80, want to go buy more over here? You'll get more 'yes' shares for your buck than you can here! And it'll help me out too :D

https://manifold.markets/Soli/will-bitcoin-go-below-80k-in-2025

Wow, people here are REALLY convinced that they know better than to call BTC price brownian motion....

I'm betting as if it's brownian & I'm treating all the other markets on BTC prices as inputs to decide what the sigma/mu should be. That math is telling me that this is insanely overpriced compared to the other bets out there that it'll be lower than 80K by the end of the year. So any yes holders here who do think it's brownian, should be able to make mana by selling this and buying those. You pay a bit of mana now, but in exchange, you pick up "yes" for all the chances it goes lower than 80K in december, at the cost of the difference (76% or 73% minus 69%). So, it's like paying 5-10m for a "yes" in "BTC first falls below 80K in december".

https://manifold.markets/polymathematic/how-low-will-bitcoin-fall-by-the-en
https://manifold.markets/Soli/will-bitcoin-go-below-80k-in-2025

Any yes bettors here want to help me understand why they're so confident those imaginary 'first dip in december' contracts are worth so little?

@DannyqnOht i'd be careful about assuming the sigma and mu on markets expiring in a week or in 6 weeks should be the same. Also, not all of these markets have enough traders/liquidity to form a strong opinion on those parameters. In any case, it's a very interesting approach. Best of luck!

@deagol Yeah, even just the brownian motion assumption is probably flat wrong. My main assumption right now is more along the lines of:

- There's lots of arbitrage opportunities even between identical contracts (below 70K in nov & below 70K in nov)
- Any kind of model that can find arbitrage between different contracts at all, even dumb ones that make bad assumptions like sigma/mu stable, will provide lots of free mana.

So far it seems to be kinda right ish! We'll see! I have a big spreadsheet! I need to actually use reasonable ways to decide where to put my mana, instead of just looping "put some mana on the biggest disagreement with the aggregate model". I built that for myself already, I just need to plug them into each other :P If anyone wants to collaborate, especially to make a better model, I'd love to!

@DannyqnOht When the quotient of the size of the sampling window to the amount of time sampled is small enough that most market participants can't judge the autocorrelation of a price move, non-linear effects dominate as the system meanders in an uncertainty region sorting complex variables from noise and guesswork, as outside driving factors begin to dominate. Down markets bring down coins. The japan move didn't help, reverse-FOMO also didn't help. It's already well into November and we're very close to repeat previous months lows.

@DavidAttenborough I don't think I understood this. But the gist I got was "my understanding of the actual stuff going on here says brownian motion is too predictable - in reality there's a ton of unpredictability still left in this, and until that shakes out assuming brownian is a bad idea." Is that kinda what you were saying?

If so, I think I'm fine with that - the idea is to take advantage of the market's over-uncertainty via an assumption like "this is gonna be the sum of a bunch of random variables", and I'm hoping brownian (which is based on normal curves in the definition, and normal curves are the sum of a bunch of random variables of equal size) is close enough. I'm explicitly ignoring the things that could affect the market and calling them all random coinflips, as a way to get some kind of signal out of the mess right now; not for the purpose of calling price moves, but for the purpose of weighing other people's calls against each other (deciding if I should buy market 1 and sell market 2).

@DannyqnOht I was basically saying uncertainty was too high, and the way people respond to uncertainty is to look for things that confirm their fears.

So rather than look at what people are doing. Look at what people are saying about what others are doing.

I just seen it it now, but it amounts to market timing in disguise, which frequently a losing proposition. And case in point, I lost this prediction.

But it was a good enough approximation of the dynamic to say the call would be close, and it was by some measure. So in the future, if you, like myself, find yourself following this line of reasoning, a good idea is to say "I'm probably being too pessimistic", and "time in market beats market timing."

Rules of thumb don't always hold in all situations, or forever, but I think it's a weak signal worth aggregating. Essentially amounts to "the market is gonna go down somewhat, but probably less than I think."