Learn Kalshi
Learn Kalshi from a lab that publishes every trade: the glossary, the lessons, every strategy and the tools, in order.
Pondletter teaches you Kalshi using a real-money record anyone can check. A lab runs trading rules against Kalshi's real order books, on paper and with a small amount of real money, publishes every result including the losses, and turns what it learns into lessons, tools and plain explanations.
It is not a signal service and it will not tell you what to buy. It will show you how the exchange actually works, what a bot gets wrong, and which strategies died and why, on real data.
The path
The real money record, plainly. The live desk is down $76.11 since April (through 2026-09-29), over 6,160 real orders, sealed by machine every night on the proof page. It trades one or two contracts at a time on purpose. The record is why the lessons can be trusted. It is not the product.
Each Sunday the board goes up: what ran, what settled, what it lost. Nothing is tuned after the fact, and the pass mark for every live test is written down before the first order.
Opening a Kalshi account? Here is our referral link. Disclosure: Kalshi adds a bonus credit to new accounts that trade $25, in an amount Kalshi sets, and the lab receives a trading credit too. It changes nothing about what is published here, and none of this is advice.
What you get
| Free | Member, $15 | Kit, $40 | Desk, $99 | |
|---|---|---|---|---|
| The glossary, the lessons by email, the grade sheet, the Sunday story and every study's finding | yes | yes | yes | yes |
| Full weekly board and every study's full tables | yes | yes | yes | |
| The lab kit, current version, as source | yes | yes | ||
| Kit updates each month, per trade files, methodology notes, the Queue vote | yes |
Prices are per month. Annual plans are two months free. Details and sign up on the membership page.
Start with the graveyard
Every strategy this lab has run, and what killed it. Fifteen strategies with real money, fourteen more on paper, and two finished ahead. The list is public. What killed each one is the reason to subscribe.
Five studies to start with
- Does Kalshi charge maker fees? On about 130 markets, yes. The fee list nobody else has published.
- How much is actually resting on a Kalshi book? Depth by time to expiry, which decides what size a strategy can ever be.
- Paper trading vs real money: the same bot, run both ways. The gap that kills most published backtests.
- Three blocks at one lot made money. The two lot block did not. What happened when the one working strategy tried to grow.
- An edge in five markets, and we can only collect it in one. The commodity finding, and the two explanations our own data killed.
The whole archive is on the studies page. Weekly boards are under State of the Lab.
Five words the posts use
These five are specific to the lab. For the exchange's own vocabulary, see the glossary.
Seat. One trading rule running on one kind of market, with its own clock and its own money.
Desk. A group of seats trading the same family of markets, such as crypto or weather.
Lane. The code that runs a seat. Live lanes place real orders, paper lanes only record what would have filled against real prints.
Block. A fixed number of orders, with its pass mark written down before the first one. A block is judged once, at the end, and never tuned in the middle.
Lot. The number of contracts per order. The lab climbs from one lot to two to four only when a block passes, and steps back down when one fails.
Written by The Heron. The exchange reserves its own market data and this lab does not pass it on; what you get here is the analysis.