Thread #62120873
File: 1776352410422.jpg (30.4 KB)
30.4 KB JPG
Thread about strategies, good markets to invest in and machine prediction models
24 RepliesView Thread
>>
>>
>>
>>
>>62120873
Easy 14% if somebody wants big apy with little risk
https://polymarket.com/de/event/friedrich-merz-out-as-chancellor-of-ge rmany-before-2027
Betting no on Fc Bayern winning Cbhampions League could be a good gamble.
https://polymarket.com/de/event/uefa-champions-league-winner
>>
>>
>>
>>
>>62121120
You can if you find undervalued items to buy and sell them later at a bit higher price. I use this for scanning
[code]
function is_good(prices) {
if (len(prices) == 0) {
return false
}
let did_have3 = false
let did_have4 = false
let did_have3_second = false
let hi_price = maxn(prices, 10)
let lo_price = hi_price / 4 * 3
prices.each(price => {
if (price <= lo_price) {
if (did_have4) {
did_have3_second = true
} else {
did_have3 = true
}
}
if (price >= hi_price && did_have3) {
did_have4 = true
}
})
did_have3_second
}
function process(offset) {
events(100, offset).each(event => {
event.markets.each(market_info => {
try {
let market = get_market(market_info.slug)
let prices = price_history(market.Yes(), "1d").map(p => p.Price)
if (is_good(prices)) {
println(`Found good market: polymarket.com/market/${market_info.slug}`)
}
} catch(e) {
//println(e)
}
})
})
offset += 1
process(offset)
}
process(0)
[/code]
>>
Was just about to make a thread. I have a Kalshi script I've been working on and it's actually profitable during my testing of buying/selling 1 contract over the course of a few days, but I'm really worried about buying and selling multiple contracts. I think that's the next biggest bottleneck. I started at $10.17, got up to $11.56 yesterday, and now I'm at $10.74. It's a slow process, but at least it's working.
>>
>>62121374
I spent the last 2 days making a weather prediction model and it is just below the profitability line
(simulated on last 30 events with 1$ for each, got 10 wins and 28$)
The wins are pretty good on average. If I can somehow filter out the bad positions, maybe I can make it profitable.
>>
>>62120873
'Nothing Ever Happens': This Bot Always Bets 'No' on Polymarket, And It Has a Point
Sterling Crispin's 'Nothing Ever Happens' bot automatically buys "No" on every non-sports Polymarket it finds. It's not that crazy.
>>
>>
>>62121420
I trade on the 15 minute Bitcoin market, which is already full of bots. I can achieve a high win rate, but it's the losses that kill my profit. I'm buying at $0.50 just to sell at $0.75, but if it doesn't hit then I'm losing the majority of my money. Stop losses are difficult because the price of BTC can vary so much, so I set it to around 0.19 or 0.20 to retain at least a little money. I haven't tried other markets like weather, but I might start looking into sports since there's a lot of data available.
>>62121422
I would say 98% of the time my bot buys No, but I did have an occasional 'Yes' here and there.
>>
>>
>>
>>
>>
>>62121727
Also today the API sent fake order book prices for several hours. Some markets just say 404 if you try to visit the correct url in a browser. It's incredible. The worst coding of any online project in the history of the internet
>>
>>
>>
>>
>>
>>62122081
How do I even go about building a bot? Do I do weather? Or is bitcoin event easier? Or something else entirely.
I can't for the life of me get the weather model to work above 40%. Then again, I am doing a coastal city in Asia, maybe that is not the best idea since weather there is unstable
>>
>>62120873
>>62121374
WTF there was already a thread: >>62119145
Started 14 hours before this one.