State of the Lab — week ending Nov 1, 2026

This week the lab ran 21 seats across 3 desks and settled 36,956 trades, nearly all simulated, 41 of them real money. Best seat: The laggard coin. Roughest seat: Early leaning favourite.

The live bot took its first real exam this week and failed it. After 60 contracts, the last minute rule stood at +2.95 cents a contract after fees, up $1.77. The pass mark, written before the first trade, was +5 cents, with both halves positive and no single day over 40 percent of the profit. The first half made +8.50 cents and the second lost 2.97. One day made more than the whole profit. That is a miss on all three counts, and it stays on the record.

Most of the swing came from the trades priced 70 to 90 cents. In the first half they won 20 of 21, which nobody keeps doing at those prices. In the second half they won 13 of 19. Paper fell at the same point, including the same rule on eight other coins, so the market got harder as well. And on matching trades, real money went from paying slightly less than paper to paying about 4 cents more. With only 31 trades on each side, some of this can still be noise, which is why the test gets more time instead of a conclusion.

So the bot keeps running for a second block of 60 on the same pass mark, one contract at a time. A pass there would not erase this result. There is also a new live test, started today. A small group of the bot's cheapest trades made more than the whole block did, and that was spotted after the fact, which is exactly the kind of finding that deserves a fresh test before anyone believes it. It runs on its own record with its own stops, and at about six trades a week it will take around ten weeks to judge.

Three more verdicts came in. The late favourite sniper failed on every coin. Perp pressure failed, with every coin negative and the strongest signals paying worst. The sniper on the four best coins finished at +1.27 cents over 687 trades, positive and still not enough: its loss rate was 6.55 percent against a limit of 6. Those closed. The paper retest of the last minute rule passed its bar on 30 October, and the numbers on that one are for members. Its live twin's record above is why a paper pass is only where the work begins.

The live desk lost $1.26 on 41 trades this week, 76 percent won, after $0.32 in fees. The crypto desk settled 32,703 fifteen-minute trades for minus $761.99, and $500.95 of that was fees. The Bitcoin hedge is down $1.67 on $747.83 after 14 days, with $3.10 of funding collected from the short side. Bitcoin's fall moved the two sides by about the same amount, as designed, and Kalshi's first 7 percent payment is due by about 13 November.

Also published this week: paper against real money on the same 53 decisions, which found that the trades you miss are the good ones, and the test of picking winners after the fact. We are still hunting, and every one of these results narrows the search.

Each desk name links to The Seats, which gives every strategy's real trigger and threshold plus the Kalshi tickers it trades, enough to run any of them yourself. The one exception is market making: which markets it quotes is listed, how it quotes them is not, and the About page explains why.

All 187 seats behind these lines, with win rates, fee drag, all-time results and the per-trade CSV, are in the members edition. 13 seats retired to date; their records stay on the board.

The market-making seats are simulated fills, and a simulation cannot model adverse selection: the tendency of a resting order to be taken exactly when someone better informed wants the other side. Where this lab has run the same strategy with real money, the live result has come in materially below the paper one. Read the maker desk as an upper bound on what the idea could earn, not as what it would have earned. The live desk below it is that check running in public: same ideas, one contract at a time, settled by the exchange.

Fill verification: 37,036 of 37,036 trades (100%) were checked against real order-book depth.

The rest are entered at the ask the exchange reported, but the lab does not record the book for every market. Challenger tennis, MLB, soccer, basketball and football have no depth archive, so those fills cannot be re-checked after the fact. Trades whose price level had no size behind it are discarded, not counted.

The archive. The lab records and permanently keeps the raw market data behind every number it publishes: trade tape from 2026-08-15 (77 days); order book depth from 2026-07-28 (96 days); top of book ticks from 2026-07-27 (97 days); settlement index from 2026-08-16 (77 days); 15-minute market quotes from 2026-07-21 (103 days). 13.7 GB and growing nightly. None of it can be recreated after the fact, which is why it gets recorded. It stays in the lab: the exchange reserves its own market data and this lab does not pass it on. What gets published is what the lab finds in it.

Every trade above was logged live, then sealed: each night the lab hashes its raw market archive and trade ledgers into an append-only chain. Rewriting any past trade would change every head after it, so check this one against last week's issue.

Chain head as of 2026-11-01T04:51Z: adc604d7b63a056dfc59d7b3323f27f76ac26634d49fc03dc62cca4dfc43606c

The full board for this week, every seat with its win rate and fee drag, is in the full board for members.