Weather Markets on Polymarket Just Grew 18× in a Year — and Our Traders Are Quietly Cashing In
Twelve months ago, trading the weather on Polymarket was a curiosity: a few dozen markets a month and under $4M changing hands. Today it is one of the fastest-growing corners of the entire platform — 9,472 daily city-temperature events, 98,061 individual markets and $827.7 million in lifetime volume. In the first eight months of 2026 alone, traders moved $741.6M through weather markets, against $39.4M in the same stretch of 2025. That is 18.8× growth in a single year. And while the market has been quietly exploding, so has WeatherBot.
We have always run this blog on one rule: show the numbers, not just the story. So before we say a single word about WeatherBot, here is the market itself — measured straight from Polymarket's own public API on August 23, 2026, not from a press release.
The market: what the data actually says
The shape of that growth is the part worth staring at. Weather trading did not drift upward — it went vertical in late 2025 and never came back down.
Read that chart slowly. In August 2025, weather traders moved about $127,000 a day. In August 2026 they are moving $3.18 million a day — and the month is not even finished. April 2026 remains the all-time record at $180.6M in thirty days, and the market has held above $120M every month since. This is no longer a niche. It is a market with real depth, real liquidity, and real money on both sides of every quote.
What the weather market looks like right now
And the people? We sampled 400 of those 1,331 open weather markets and counted 3,012 distinct wallets holding a position — in under a third of the board. The weather crowd is no longer a handful of meteorology nerds. It is thousands of traders, every single day.
While the market grew, WeatherBot grew with it
Here is the part that matters to us. A rising market is only good news if you are actually positioned in it — and our members have been, from the very first month of the run. These are our own numbers, the same ones you see on the front page:
Put our 30-day volume next to the market's current run rate and the picture gets interesting: weather markets are turning over roughly $95M a month at today's pace, and $15.5M of it flows through WeatherBot. That is close to one dollar in every six traded on Polymarket weather — from a single platform, executed by bots that never sleep through a forecast update.
The card we just put on the front page
We got tired of members asking us how everyone else was doing. So we put it on the homepage, in public, where anyone can see it: a live payout card that shows exactly what our traders won this week — the total, how many of them won, the biggest single win, the best trade of the week, and which cities the money came from. It re-rolls every Sunday. Here is the week of August 23:
Look at that +1,198% for a second. That is one member, one weather market, one afternoon — a position bought cheap because four forecast models disagreed with the crowd, and settled at a dollar when the thermometer agreed with the models instead. Those trades are not the norm, and we would never pretend otherwise. But they happen most weeks, and when they do it is almost always the same setup: a mispriced temperature bucket that nobody was watching, on a city nobody was talking about.
The quiet money
Here is what we love most about the weather crowd. Nobody posts about it.
There is no weather-trading Twitter. No influencers screaming about their temperature calls, no Discord full of rocket emojis. Our members are not loud — 2,249 of them won money this week and you would never know it, because the average winner did not hit a moonshot. They took a hundred small, boring, positive-expectation positions on whether Seoul closes above 26°C, and let the math do what math does.
That is the whole thesis. Weather is the rare market where the truth arrives on schedule. Every position resolves against a thermometer, usually within 24 hours — no CEO can talk it up, no narrative can save a bad entry, no whale can hold a losing bag until the story turns. Either the temperature hits or it does not. And because five professional forecast models publish their view of tomorrow hours before the market fully reprices, there is a window — small, repeatable, and open several times a day across 57 cities — where the odds on the screen are simply wrong.
Our engine does nothing more clever than that: read five models, find where the crowd disagrees with the consensus, size the position with the Kelly criterion, and hold to resolution. Do it a few thousand times and the result is the number on the front page.
Where these numbers come from
The Polymarket figures in this article are not estimates. They were pulled from Polymarket's public Gamma API on August 23, 2026, month by month, covering 10,994 weather events and 112,962 individual markets; the daily city-temperature subset used for the chart is 9,472 events and 98,061 markets. The live snapshot counts 125 open events and 1,331 open markets, and the trader count comes from querying holders on 400 of those 1,331 markets — so 3,012 is a floor, not a ceiling.
August 2026 is incomplete: it covers trading through the 23rd only, which is why its bar is a lighter shade. WeatherBot's own figures come from the same internal ledgers that power the live leaderboard. And the usual, honest caveat: none of this is a promise. Weather markets can and do run against a good forecast, past performance guarantees nothing, and you should never fund a bot with money you cannot afford to lose.
The market is 18× bigger. The edge is still there.
That is the thing we did not expect. Usually, when a market grows this fast, the easy money disappears — more traders means sharper prices means thinner edge. Weather has not worked that way yet, and we think we know why: the volume arrived faster than the expertise did. Thousands of new traders now click into a Seoul temperature market every day, and almost none of them have read the 06z ECMWF run. The crowd got bigger. It did not get better.
We do not know how long that lasts. Probably not forever. But right now there are 1,331 open weather markets, $3.18 million a day moving through them, and a forecast that publishes on a schedule anyone can trade against.
Market data sourced from Polymarket's public API on August 23, 2026. Platform figures are WeatherBot's own realized member results. Trading prediction markets involves risk of loss; nothing here is financial advice.