Weather 80 %, switch on a flip
by PConsensus · Oct 25 2026 · Site export · done
Study rule: Weather markets (99 % temperature ladders): buy the side 80 %+ of the 5+ tracked holders agree on, any money, any price; switch when a 60 %+ consensus forms on the other side. Study verdict (Oct 22 2026, categories): speculative: no edge found (+1.4 %, n 526); only to collect live data on the category the owner asked for
- Closed trades
- 706
- Mean return / $1
- −5.7 %
- −11.7 % to +1.3 %
- Hit rate vs break-even
- 68 % vs 64 %
- In / out of sample
- +0.2 % / −6.3 %
- 54 / 651 trades
Buy the side 80 %+ of 5+ tracked wallets agree on, in Weather markets, at any price (0.05–0.95), bought as soon as the rules hold, switched to the other side when the money flips (up to 3 times); after a 1¢ half-spread and Polymarket's taker fee, $1 a trade, over the dataset's whole window. Over 706 closed trades between Mar 10 2026 and Oct 22 2026 it lost 5.7 % per $1 staked (95 % interval −11.7 % to +1.3 %). It won 68 % of its trades and needed 64 % to break even at its entry prices.
The mean return is at or below zero and the interval includes zero: no edge after costs. Past results, not advice.
Plain consensus, all markets on the same data and period: −2.4 % per $1 over 11,908 trades (−4.1 % to −0.7 %). This rule did 3.2 points worse per trade, intervals overlapping.
Plain consensus, Weather only on the same data and period: +0.8 % per $1 over 958 trades (−4.4 % to +6.1 %). This rule did 6.5 points worse per trade, intervals overlapping.
The numbers
| Per $1 staked, after costs | All | In-sample (to Aug 10 2026) | Out-of-sample (after) |
|---|---|---|---|
| Closed trades (events) | 706 (391) | 54 (27) | 651 (364) |
| Mean return per trade | −5.7 % | +0.2 % | −6.3 % |
| 95 % interval | −11.7 % to +1.3 % | −26.0 % to +26.3 % | −13.0 % to +0.3 % |
| Hit rate | 68.4 % | 68.5 % | 68.4 % |
| Break-even hit rate | 63.7 % | 49.6 % | 65.3 % |
| P/L at $1 stakes | −$40 | +$0 | −$41 |
| Largest drawdown | $55 | $4 | $53 |
| Most stakes open at once | 219 | 22 | 219 |
| Average entry price | 0.72 | 0.68 | 0.73 |
| Average hold | 2.8 d | 13.4 d | 41 h |
| p (mean > 0, one-sided) | 0.958 | 0.495 | 0.969 |
Period Mar 10 2026 – Oct 22 2026; markets ending before Oct 15 2026 count; split at Aug 10 2026. Interval: cluster bootstrap (events resampled).
Cumulative P/L at $1 stakes
By topic (the 10 most traded)
| Topic | Trades | Hit | Mean / $1 | P/L |
|---|---|---|---|---|
| Weather | 706 | 68 % | −5.7 % | −$40 |
By entry price
| Price band | Trades | Hit | Mean / $1 | P/L |
|---|---|---|---|---|
| 0.05-0.30 | 22 | 23 % | +50.8 % | +$11 |
| 0.30-0.50 | 57 | 39 % | −19.4 % | −$11 |
| 0.50-0.70 | 191 | 52 % | −13.7 % | −$26 |
| 0.70-0.95 | 429 | 82 % | −3.3 % | −$14 |
| 0.95-1.00 | 7 | 100 % | +4.0 % | +$0 |
By month of entry
| Month | Trades | Hit | Mean / $1 | P/L |
|---|---|---|---|---|
| 2026-03 | 36 | — | +7.4 % | +$3 |
| 2026-04 | 17 | — | −3.8 % | −$1 |
| 2026-05 | 2 | — | −47.1 % | −$1 |
| 2026-08 | 2 | — | −47.9 % | −$1 |
| 2026-10 | 649 | — | −6.2 % | −$40 |
Best trades
- Will the highest temperature in Seoul (Incheon) be 27°C on October 12?No at 0.07 · entered Oct 12 2026 · won+1141 %
- Will the highest temperature in Manila be 33°C on October 7?No at 0.12 · entered Oct 7 2026 · won+637 %
- Will the highest temperature in Seoul be 17°C or higher on March 29?No at 0.13 · entered Mar 29 2026 · won+580 %
- Will the highest temperature in Cape Town be 20°C on October 13?No at 0.70 · entered Oct 12 2026 · won+482 %
- Will the highest temperature in Chongqing be 23°C on October 12?No at 0.28 · entered Oct 12 2026 · won+233 %
Worst trades
- Will the highest temperature in San Francisco be between 64-65°F on October 11?No at 0.79 · entered Oct 11 2026 · lost−100 %
- Will the highest temperature in Wellington be 17°C on October 11?No at 0.56 · entered Oct 10 2026 · lost−100 %
- Will the highest temperature in Ankara be 24°C on October 10?No at 0.65 · entered Oct 9 2026 · lost−100 %
- Will the highest temperature in Miami be between 86-87°F on October 3?No at 0.87 · entered Oct 3 2026 · lost−100 %
- Will the highest temperature in Seoul (Incheon) be 26°C on October 11?No at 0.18 · entered Oct 10 2026 · lost−100 %
Read before believing it
- Trying many variants finds lucky ones: the more rules you test, the more of them will look good by chance. A single Promising run is a hypothesis for a paper bot, not a proven edge.
- Returns are per $1 staked, equal stakes, prices from the tracked wallets' own fills (median error about half a cent, more in fast markets), entries and exits at hour boundaries. The live order book will differ.
- The in-sample half is the first 70 % of the period by time, the out-of-sample half the last 30 %; a trade belongs to a half when it was entered in it and its market ended before the half closed.
- Past results, not advice.
- Only the fills of wallets the site tracked, from the day it started tracking them: curated wallets from Mar 10 2026, the leaderboard set from Oct 2 2026. A position a wallet opened before it was tracked is invisible.
- Holders are rebuilt from on-chain fills; the live site also reads Polymarket's position answers, so the live consensus can differ (85–92 % side agreement on tradable markets).
Weather 80 %, switch on a flip
Study rule: Weather markets (99 % temperature ladders): buy the side 80 %+ of the 5+ tracked holders agree on, any money, any price; switch when a 60 %+ consensus forms on the other side. Study verdict (Oct 22 2026, categories): speculative: no edge found (+1.4 %, n 526); only to collect live data on the category the owner asked for
Past results, not advice. Numbers are per $1 staked after the run's costs on the tracked wallets' own fills; the live order book will differ. Clone this run and change one thing.