Membership

Free gets you every finding the lab makes. Member gets you the bot itself: Ping installs on your computer with one line and runs eight of the lab's strategies on paper with your own Kalshi key, each with its real record next to it, with the code open to read and change. Plus a seat at our live trading, every real trade our bot makes on Kalshi, updated every night; and the tools to build your own: a step by step course, a tested prompt for every strategy we cover, and the harness the lab itself runs on, with every number behind every claim.

Ping, in one minute

Ping is the paper-trading bot the lab runs its strategies on, packaged so you can run it yourself. See the public demo first, no sign-up. Once you join, the install page gives you a personal one-line command; about ten seconds later Ping opens in your browser. Paste your Kalshi key, pick a strategy, press Run on paper, and the panel tells you what it is watching and what it is waiting for.

• Eight strategies, eight records, losses included. Buy the favourite, chase the jump, fade the spike, the 99 cent contract, Sam's week, the maker ladder, follow the big trade, the long shot. Three of them are losers and are in the box on purpose.
• Honest fills. A paper fill counts only when a real trade prints through your resting price, with Kalshi's real fees. That is the rule that has kept the lab's paper numbers close to its real-money numbers.
• Your own rules. Press Modify on any strategy to see its code next to its settings; Customise makes a copy you can edit, and Save checks it before anything is written. Or describe what you want and your own Claude or OpenAI key writes the rule for you to read first.
• Yours, on your machine. Ping listens only on your own computer, keeps keys in your keychain, has no order-sending code, and confirms your membership with the site once a day. How it is built and how to check it.
• What is next. A live-orders version after a month of beta: per strategy only after a paper record, one contract to start, a daily loss limit and a kill switch, every fill reconciled against the exchange. Members on paper now get it first.

The course

The course is the build the lab wishes it had followed: what to set up, in what order, and the checks that stop a paper result from lying to you. Each lesson ends with something running, and comes with the prompt that builds it with Claude, ChatGPT or Cursor. Lessons release weekly.

1. Read a Kalshi market: prices, the book, fees by series, and how it settles. Out now
2. Pull free history and test a rule at prices you could really have traded. Out now
3. Turn the test into a paper bot that records every decision. Out now
4. Honest fills: count a trade only when the market proves it would have filled. Oct 27
5. Grade it: when a result is real, and when it is one lucky week. Nov 3
6. Run it in small blocks with written stops, before any real money. Nov 10

Real money is never required. Nothing in the course is a strategy that makes money; it is the machine that tells you honestly whether yours does.

Three honest notes

Ping is paper only, today. It cannot place a real order; there is no code in it that could. The live version is the next build and it ships behind gates, not before. Paper results are a best case, which is exactly why Ping counts them the strict way.

One hold-back. Every real-money fill is in the ledgers, with prices, except for two favourite-maker strategies. Their prices show where our resting bid sat, and we're testing a night-only version of that rule. They'll be added when that test ends, by 7 November. Their results are in full on the proof page.

None of this is advice. No funds are managed, no trades are placed for anyone, and nothing here is a recommendation. It is a research publication about a market, written by someone running experiments on it with their own money, as stated plainly on the About page.

The data archive

What members get, stated precisely so nobody is oversold. Every per trade file behind every board the lab has published, week by week back to its first week, served at /data/ and gated by your login. Access follows your subscription rather than your join date. These are the lab's own records: which seat traded, at what price the book was showing, what fee applied, and how it settled. They are what you need to recompute any figure the lab prints. What members do not get is Kalshi's market data itself. The lab records the raw order book continuously for its own research, but Kalshi's data terms reserve redistribution, so those recordings stay in the lab and what reaches you is what the lab found in them.