FORECAST FORKThe instability layer of Weather Trader
FF

Forecast Fork

Is the forecast stable, or forking?live from weathertrader.com
A measure of forecast instability

Is the forecast stable,
or forking?

Most weather products tell you the forecast. Forecast Fork tells you whether to trust it — how much the ensembles disagree, which competing scenarios are still unresolved, and how likely the forecast is to change in the next few cycles. It is the open, continuously-verified instability layer of Weather Trader — and the numbers below are pulled from it live.

See it live on the Forecast desk → How it's verified

  The Fork, live & verified in the open

Spread predicts skill
correlation of issue-time ensemble spread with realized Day-5 ACC — negative is the whole point
Runs verified out-of-sample
every forecast scored against what actually happened, in the open
The latest run is forking toward
the pattern the current cycle is aiming at
Issue-time ensemble spread  →  realized Day-5 anomaly correlation · the whole archive● live

loading live from weathertrader.com…

This is not a mock-up. It is fetched, as you load the page, from Weather Trader's verification engine: bin every run by the ensemble spread it showed at issue time, plot the skill it actually verified. The down-slope is the proof that spread is a bust signal — known before the outcome. That is the Forecast Fork.

The idea

A forecast can be confident — every model and member converging on one outcome — or forked: split into distinct scenarios the atmosphere hasn't chosen between yet. The second kind busts. Forecast Fork is the probability that materially distinct scenarios remain unresolved — and, uniquely, that the forecast itself is about to move.

FFI = P( materially distinct forecast scenarios remain unresolved )  ·  from cluster divergence, spread growth, cross-model disagreement, run-to-run momentum & bimodality

Being precise about status: the Fork is really three quantities. FFspread — current uncertainty — is live and verified (the chart above). FFbust — a calibrated P(a forecast fails) — and FFrevision — P(the forecast itself changes next cycles) — are the models we are building next. Spread predicts uncertainty; it is not yet a calibrated bust probability. We combine them into one headline index only after each is independently, out-of-sample calibrated. Method & status on weathertrader.com/method.

The components — live now on Weather Trader

Index livethe fork number

One number per region & lead — how forked the forecast is — on the Forecast desk's agreement heat.

Scenarios liveEPS clusters

The 50 members clustered into competing patterns per region; when a minority scenario beats the mean, the consensus underweighted the truth. On every verify-time page.

Calibration liveproven, not claimed

Spread↔skill correlation, published and continuously updated — the chart above. See the Method.

Momentumrun-to-run

How the distribution moves between cycles — drift, spread acceleration, reversals — surfaced as the what's-changing strip.

Attribution livewhy it forks

The synoptic regime the flow is in — blocked, zonal, PNA− — and which model owns it. The named reason a forecast is unstable.

Archive liveissue-time record

Tamper-evident, content-addressed record of every forecast as issued — the sha256 is the identity, so any change is detectable. Hash-chained + externally-anchored roots are on the roadmap.

The method — a branching probability graph

Forecast Fork is not ensemble spread. For a fixed valid time the ensemble is a mixture of coherent scenarios, and we track those branches across successive model runs — not re-clustering and forgetting each cycle. The object is a branching graph: how forecast alternatives emerge, split, merge, and gain or lose probability as the forecast evolves toward reality.

Ft(x | v) = Σk πk,t · Gk,t(x)  ·  persistent branches · probability-mass flow Δπ · cross-model support · calibrated revision & bust

The system runs in both directions. Forward Fork: from the current state, which outcomes are possible and how are their probabilities evolving? Reverse Fork: given a specified outcome — a station's daily high above a threshold — which branches carry it, what must stay true (trigger), and what would kill it? P(E) = Σ qk(E)·πk, decomposed by branch via Bayes P(Bk|E).

The Fork family

Fork State livescenario structure

The current branches — how many, how separated, their probabilities and cross-model support.

Reverse Fork experimentaloutcome → pathways

Given an outcome: its probability, the branches that carry it, and the trigger / kill conditions.

Fork Momentum experimentalprobability-mass flow

Which branch is gaining, losing, stabilizing or reversing — Δπ tracked across runs.

Fork Bust experimentalcalibrated P(fail)

A fitted, out-of-sample-scored probability the forecast fails — not merely spread.

Fork Independence experimentalreal vs echoed consensus

The effective number of independent model families behind a scenario — AI models echo their IFS/ERA5 training.

Fork Revision buildingP(forecast changes)

The calibrated chance the distribution materially revises in the next cycles.

Fork Cascade buildingone branch → all targets

Propagate one atmospheric branch to every downstream outcome — temperature, HDD/CDD, load, wind, contracts — coherently.

Fork Autopsy buildingpost-mortem

After verification: when the winning branch appeared, how its probability evolved, what was knowable when.

Status is honest: live is running publicly, experimental is built and being out-of-sample verified, building is on the roadmap. All of it runs on our own multi-terabyte ensemble archive, computed near the data. Full method + component status: weathertrader.com/method.

Why it's different

Not "will it be hot"

  • the full outcome distribution, not a single number
  • the competing scenarios, named and tracked
  • whether the forecast is likely to change
  • where and when the models will most likely bust

Public & provable

  • continuously, out-of-sample verified — in the open
  • model-by-model reliability and error, published
  • honest probabilities (recalibrated, validated)
  • a recognizable measure anyone can cite

Forecast Fork is the transparent, auditable layer of Weather Trader — a prediction market for the atmosphere. The maps and verification are free and public; the goal is for "the Forecast Fork is high" to become how people say a forecast is unsettled. Start on the live Forecast desk or read the method.