2019: UK wheat disaster year, 8 of 9 regions correct

Crop year 2018/19 · published 2026-05-02

In one paragraph
The 2018/19 UK wheat year combined three independent stress events: wet autumn drilling, cold tillering in the east, hot dry June at flowering. CropIntel's walk-forward ensemble (trained only on 1999–2017) called 8 of 9 wheat-growing regions correctly as below-average. UK average yield came in at 7.0 t/ha, the worst since 2012. High-confidence calls were 100% correct.

By the numbers

-0.73
Predicted anomaly (t/ha), below-average
-1.12
Actual anomaly (t/ha), below-average
8 of 9
Tradeable regions correct

Walk-forward, the model was trained only on years before this one. Every figure is a plotted point on the track-record scatter.

The macro story

Autumn 2018 was the wettest in years across much of the UK arable belt. Drilling windows shrunk; some growers wrote off fields they couldn't get into the ground at all. The east and north then took an unusually cold tillering period through January–February, slowing crop establishment further. By April the surviving wheats were noticeably backward.

June 2019 then delivered an unusually hot, dry spell during the critical flowering and grain-fill window. With shallow root systems from the wet autumn and weak tillering, the crop couldn't draw enough moisture to support kernel development. Yields fell across England.

What the model said

The walk-forward ensemble, trained only on data up to 2017, produced a low-yield call for 8 of 9 wheat-growing regions ahead of the May 2019 checkpoint. The compound stress score fired across drilling, tillering, and flowering stages simultaneously, which the model recognises as a high-confidence "bad year" pattern from the historical analogues.

The single region called wrong was Eastern, where the model's predicted anomaly was barely positive. Even there the call was within touching distance of the threshold.

What actually happened

Why it matters for the methodology

2019 is the strongest single-year vindication of the compound stress framing. Single-event weather models that trigger on heatwaves alone would have fired late (the heat was June, the harvest was August); compound models fired when the wet autumn had already locked in part of the deficit, months earlier. The 8/9 correct rate at the May checkpoint demonstrates how multi-stage stress aggregation produces earlier, more confident calls than any single-feature approach.

The 2019 call also stress-tested the consensus filter: all of the ensemble's complementary components agreed on the bearish direction, which is why the call was published as high-confidence. When the components disagree the model emits no tradeable call, by design.

The economic benefit, by user type

What the call was worth, by user type, the same signal, three different decisions:

Grain merchant. A months-early, regionally-resolved below-average call is time to secure supply and set basis before the market reprices the shortfall. On a 100,000 t book, a ~0.8 t/ha regional shortfall is tens of thousands of tonnes to re-source, moving early on that is the difference between paying the pre- and post-repricing price.
Crop insurer. A below-average signal a season ahead lets you reserve earlier and re-weight exposure away from the regions heading into compound stress, before the loss crystallises at harvest. One quarter of reserving lead time on a bad year is materially valuable.
Commodity desk. A leading (not coincident) directional signal with a published hit rate is a sized edge: hold or add to a long feed-wheat view through the deterioration, or use it as an independent confirmation layer on your own fundamentals.

Illustrative, the figures show the magnitude and direction of the decision, not a guaranteed return. The model is directional and probabilistic (62.3% walk-forward). See who this is for for per-segment detail.

Who this is relevant to

CropIntel's signal, and the underlying system (available to acquire), is relevant across the UK and European arable value chain:

Representative firms by segment, illustrating breadth, not claimed clients or relationships. If your firm is on this map and the track record is interesting, start a conversation.

Related: All case studies · Methodology · Track Record · 2023 wet harvest · 2012 washout