Anatomy of a thin-book loss · BroCode Research
01The claim
Polymarket trades through a Central Limit Order Book (CLOB). In a thin market, very<br>few shares are available near the current price. Our system trades daily temperature<br>markets, and those books are thin by construction: each city and day is split into<br>narrow temperature ranges, attention is scattered across dozens of cities, and the<br>extreme ranges barely trade at all. Most writing about thin books stops at slippage,<br>the cost of pushing a large order through a shallow book. Our weather book found<br>three other failure modes: resting orders that fill exactly when you are wrong,<br>inventory your own ledger does not know about, and a measurement pipeline that a<br>single unquoted share can switch off.
02What the weather book lost
The weather book's losses were small in dollars and large in information. The<br>figures below come from completed trades and system logs on live temperature<br>markets, not a backtest.
Working notes · from the live weather book
The book's worst recorded day lost $14.80 : six stop-losses<br>totaling −$20.94 (average −$3.49) against eight winners totaling +$6.13 (average<br>+$0.77), a 4.5:1 loss-to-win ratio. Four winners that all eventually resolved to<br>$1 were cut early, leaving about $29 on the table ; holding them<br>would have flipped the day to roughly +$15.
The entry mix leaked from both ends. An expensive favourite, a Seoul range<br>entered at $0.82 that the model priced at 0.91, lost $6.66 ; a<br>near-coin-flip Istanbul range entered at $0.56 lost $13.09 . The<br>note in the log reads "pennies up, dollars down."
The largest weather inventory incident has no clean headline number because the<br>ledger could not see it: a duplicated entry left 51.7 shares invisible<br>to every exit decision , and they rode a collapsing range into<br>settlement (Case 1 below).
While all of this ran, the venue's account page showed a $37<br>loss when the account, across all books, had actually lost<br>$187 . The account P&L API lagged by as much as an hour, and our own ledger, the running record<br>of events that change the account balance, trusted the delayed figure.
The logs also contained 15,639 failed exit attempts , roughly<br>11,728 of which the remediation notes bucket as venue rejections for<br>insufficient balance or below-minimum order size. We changed cancellation and<br>exit handling so the same operation can be retried safely without duplicating an<br>action or crashing the strategy.
03The weather bleed: adverse selection in a thin book
The slow loss in a thin book comes from resting orders. An order that waits in the<br>book and adds liquidity is the maker; an incoming order that executes<br>immediately against it is the taker. A resting bid does not fill at a<br>random moment. It fills when someone chooses to sell into it, which happens more<br>often when the seller knows more or the price is about to fall. That is adverse<br>selection.
Our first maker model made the classic mistake of booking profit at posting time. It<br>treated the midpoint between the best buy and sell prices as fair value, our<br>estimate of what the share should be worth, and recorded<br>edge = 0.3 × halfSpread + rebate the moment an order was placed, where<br>the 0.3 multiplier was an unsupported guess and a rebate is a venue payment for<br>providing liquidity. With fair value pinned to the midpoint, the Kelly criterion, a<br>formula for choosing position size to maximize long-run capital growth, turns<br>positive for every non-zero spread, so the model saw opportunity everywhere. What it<br>never measured was what usually happened after a fill.
In the weather book this failure had a specific shape the logs call the<br>weather bleed. Early box construction bought the cheap side first: a $0.09<br>long-shot leg resting in the book, to be paired with a $0.83 favourite leg for a<br>guaranteed-profit box. The favourite leg often never filled. The cheap leg did,<br>precisely because selling into a resting bid on a long shot is attractive to whoever<br>knows the outcome is drifting away, and the system was left holding a naked,<br>decaying long shot.
The fixes were structural. Boxes are now built favourite-first, and a cheap leg may<br>only be bought to complete a box whose favourite side is already secured, never led<br>naked. An entry floor (no solo entry below $0.20, no directional entry below $0.50)<br>blocks the bleed at the gate. Section 07 covers the one deliberate, narrow exception<br>we later cut into that floor.
04Case studies from the weather book
The cases below all come from the live weather book during June 2026. Each one led<br>to a specific change in the system.
Case 1: The entry the ledger could not see
On an NYC temperature range, the engine bought 51.7 shares at $0.62, twice. The<br>second order was sized and sent before the first fill had been polled from the<br>venue, so at decision time the first purchase was invisible: not yet a position,<br>no longer a resting order. Both fills were real. The local book recorded roughly<br>half the true inventory.
Every...