Every strategy this lab has run, and what killed it

By The Heron. Every figure below is computed from fills synced from the exchange's own records, and the live desk's daily total is sealed nightly on the proof page.

This lab has run 15 strategies with real money and 14 more on paper. Two of the 15 finished more than a dollar ahead, and both of them are now stopped.

That is the honest headline, and this page is the reason the lab exists. Every strategy here was given a fixed number of orders and a pass mark written down before the first one. Most of them failed. What is worth something is not which ones won. It is the specific thing that killed each one, because those are the mistakes you would otherwise pay to discover yourself.

What was tried, with real money

• Crypto 15-minute favourites, with a reachability filter. Buy the favourite late in a crypto 15-minute market, as a resting order, only when the settlement index is far enough from the strike.
• WTI oil 15-minute favourites. The same favourite rule, ported to oil.
• Favourite maker, multi-asset. Rest on the favourite side across several crypto books at once.
• Natural gas 15-minute favourites. The favourite rule without the index filter, on the one commodity market that passed its paper bar.
• Esports pre-match book. Rest on the favourite before a League of Legends match.
• Esports match clock. Trade the Dota match clock around a clinch.
• XRP 15-minute two-sided maker. Quote both sides of an XRP book and earn the spread.
• NEAR 15-minute maker. The same two-sided quote on NEAR.
• Weather maker. Quote daily high-temperature markets.
• Resting offers, error harvesting. Rest offers to catch other people's mistakes.
• Two-sided book with an implied-volatility model. Price both sides from an options-derived fair value.
• Paired book, pair completion. Rest both sides and earn the spread whenever both legs fill.
• Crypto 15-minute favourites, first version. The original favourite maker, before the reachability filter.
• Liquidity-rebate quoter. Quote at the touch to earn the exchange's liquidity rebate.
• Taker bots, before the lab existed. Cross the spread on signals that looked good.

What was tried on paper

• Taker twin of the favourite maker. Same signals, crossing the spread instead of resting.
• Bitcoin leads the altcoins. Trade alt markets on a Bitcoin move.
• The speed rescue. Re-run a failed maker at 2 seconds instead of 20.
• Dipped favourites recover. Buy a favourite that just dropped.
• One-cent longshots. Buy 1 cent lottery strikes.
• Esports longshots. Buy cheap underdogs.
• Tennis favourites. Buy the favourite in tennis markets.
• Fade the move. Assume crypto moves overshoot and fade them.
• Esports fallers. Buy teams mid-collapse.
• Hourly crypto favourites. The favourite rule on hourly rather than 15-minute markets.
• Commodity favourites, four of five. The favourite rule on oil, copper, gold and silver.
• Weather models against the market. Use ensemble forecasts to price temperature markets.
• Cross-venue weather arbitrage. Trade the gap between Kalshi and Polymarket on the same city.
• Football against the sharp books. Trade Kalshi football against Pinnacle's no-vig price.

What members see

Below the line: what each of these returned per contract, over how many fills, and the one paragraph that says exactly what killed it. A few of them are worth knowing before you build anything on Kalshi yourself. The taker toll that ran three and a half times the gross edge. The paper strategy that looked excellent and lost money in a day. The rebate quoter that got picked off in hours. The forecast model that lost more the more confident it got.

Membership is $15 a month, or read it free for seven days.