The crowd got two right, the agronomist got two right, and nobody saw the durum coming.
Six markets, all settled on a published source, with the final percentages next to what actually happened. It is not a flattering table and that is what makes it worth keeping.
The six markets of 2025, settled.
| Market | The crowd’s favourite | What happened | Verdict |
|---|---|---|---|
| Spring barley yield | 5.5 to 6.5 t/ha, on 41 per cent | 4.1 t/ha | Wrong, and badly. Only 9 per cent picked the lowest band |
| Winter wheat yield | 7.5 to 8.5 t/ha, on 38 per cent | 5.9 t/ha | Wrong. Drought took about thirty per cent of it |
| Durum yield | Around 6 t/ha, on 52 per cent | Under 2 t/ha | Spectacularly wrong. 4 per cent picked under 3 t/ha |
| Lambing percentage | 170 to 190 per cent, on 44 per cent | 178 per cent | Right, comfortably |
| First combining date | 15 to 24 July, on 39 per cent | 19 July | Right, and the closest market of the year |
| June rainfall | 30 to 60mm, on 40 per cent | 21mm | Wrong, and the clue to everything above it |
The durum market, in detail.
Fifty-two per cent of nearly three thousand people put the durum at around six tonnes a hectare, which was the agronomist’s number and the number in the farm plan. Four per cent went under three.
It came in under two. Not because of anything anybody did, but because durum wants a dry ripening period after a wet enough spring, and 2025 delivered the driest spring in a very long time followed by exactly the wrong sort of June.
The interesting part is not that the crowd was wrong. It is that the crowd was wrong in the same direction as the expert, at the same time, for the same reason — which is the standard failure mode of a prediction market and the reason nobody should treat one as a forecast.
The four results worth saying out loud.
Four cards, four numbers, and one of them is an admission.
Share cards carry the year and the number, and nothing that could be mistaken for a tip.
What two seasons of this have taught us.
That people are good at dates and bad at yields. The combining date market has been called correctly twice running; no yield market has ever been called correctly.
That the crowd is systematically optimistic about arable and realistic about livestock, which is probably because more of the people guessing have kept sheep than have grown durum.
And that a market on the weather, settled early, is worth more than all the yield markets put together — because it is the only one that settles before the thing it predicts.
Last season, questioned.
One hundred and eleven people out of two thousand nine hundred and forty. That is the highest-scoring single pick in the history of this game.
All six. Weighbridge tickets and the lambing record are published here; the rainfall is on the weather page; the combining date is on camera.
Five of them. The regional ranking market replaces the durum one, because there is no durum in the 2027 plan.
This season’s markets.
Six of them, open now, and still free.
Elsewhere on the estate.
/estate/the-money/
/estate/recording/
/estate/sitemap/
