秋田预测市场 · Discord 审计

用户bot在15分钟合约中频繁止损与策略效率争议

2026-07-02 · Kalshi · 14 条相关讨论

Can anyone give advice? My bot has like a +30% success rate when it comes to trading but my algo is hit +70-80% in predicting the prices throughout the contract length. The bot seems to be doing good in the one hour contracts but tanks horribly in the 15m

Can anyone give advice? My bot has like a +30% success rate when it comes to trading but my algo is hit +70-80% in predicting the prices throughout the contract length. The bot seems to be doing good in the one hour contracts but tanks horribly in the 15m

skywurz 2026-07-02 06:38:14

so this is the part people miss . if you are entering in at that 70-80% range .70-.80 you have to be correct 85%+ of the time to make like 2% and the omega is not going to be great. 15m is a 100% different beast than hourly and if you use the same algo for both your gonna have a bad time

so this is the part people miss . if you are entering in at that 70-80% range .70-.80 you have to be correct 85%+ of the time to make like 2% and the omega is not going to be great. 15m is a 100% different beast than hourly and if you use the same algo for both your gonna have a bad time

Nah I have different algos for 15m, 1h, and daily Same with each currency everything has its own historic data and has different methods, weights, etc

Nah I have different algos for 15m, 1h, and daily Same with each currency everything has its own historic data and has different methods, weights, etc

skywurz 2026-07-02 06:55:28

also last night to now has all been up markove has an up bias so i have found it to work ok during a steady up chug . but then weekend chop market changes sudden collapse all sucks

also last night to now has all been up markove has an up bias so i have found it to work ok during a steady up chug . but then weekend chop market changes sudden collapse all sucks

A-Rod/Breakdancingfool 2026-07-02 07:28:20

This is great and for under 24 hours. I say you're good. As SKYWURZ said you gotta make some chamges. Im trying to create a bot just for the 15min but life happens and before I know it im knocked out lol. Its tough for i had one made but it was slowly bleeding my paper account. Now im trying to do a different strategy but still not finished. Which one is more successful the hourly or daily?

This is great and for under 24 hours. I say you're good. As SKYWURZ said you gotta make some chamges. Im trying to create a bot just for the 15min but life happens and before I know it im knocked out lol. Its tough for i had one made but it was slowly bleeding my paper account. Now im trying to do a different strategy but still not finished. Which one is more successful the hourly or daily?

Goat 2026-07-02 08:25:31

No the 15 min markets are ridiculously efficient. You can’t calculate fair value better than the market implied probability of the yes/no prices. You are essentially along for the ride with taker Strats. Maker Strats are a better chance but they are priced appropriately as well. I’d suggest going to a different category like elections

No the 15 min markets are ridiculously efficient. You can’t calculate fair value better than the market implied probability of the yes/no prices. You are essentially along for the ride with taker Strats. Maker Strats are a better chance but they are priced appropriately as well. I’d suggest going to a different category like elections

Goat 2026-07-02 08:26:40

I said better I should have said *faster* and on small time scales like 15 mins the price is a random walk

I said better I should have said *faster* and on small time scales like 15 mins the price is a random walk

The hourly is doing the best so far It’s the only one with gains but when the 15m hits, it wins big but bleeds so much the daily is just meh and the yearly scalps a lot Also thank you for the feedback as well. I’m sorry about ur situation. By different strategy are you still trying to do the bot for prediction markets or something else entirely

The hourly is doing the best so far It’s the only one with gains but when the 15m hits, it wins big but bleeds so much the daily is just meh and the yearly scalps a lot Also thank you for the feedback as well. I’m sorry about ur situation. By different strategy are you still trying to do the bot for prediction markets or something else entirely

skywurz 2026-07-02 09:02:50

i ended up getting like 2k from a programming accident that happened to scoop up multiple reversals . i bled 600 of that trying to chase that dragon and have use the rest to pay mayself back for startup fees and the left over has just kept this as a really really strange hobby..

i ended up getting like 2k from a programming accident that happened to scoop up multiple reversals . i bled 600 of that trying to chase that dragon and have use the rest to pay mayself back for startup fees and the left over has just kept this as a really really strange hobby..