TECHNICAL · MODEL OVERVIEW
ENSEMBLE MODELMONTE CARLO BRACKETPUBLIC RECEIPTS

How our World Cup model works

A calibrated, transparently graded probability model and a Monte Carlo bracket simulator for FIFA World Cup 2026 — built on the principles below and held to the public receipt on the Model Record page.

We publish every prediction and grade each one as the tournament plays out. The internal weighting is proprietary; the outputs are not.

1 · What the model does

Every World Cup 2026 match probability on Onside is the output of an ensemble that blends multiple independent indicators of team strength, calibrated against the long-run historical baseline for international football. The output is a win / draw / loss probability vector for every fixture — and, by running the full bracket forward many thousand times, a per-team probability of reaching each knockout round and lifting the trophy.

2 · The principles we held to

Calibrated, not optimistic. Where the football itself doesn't give you a clear favourite, the model shouldn't pretend otherwise — draw rates and tight-fixture splits hew to the historical baseline rather than the consensus narrative. Transparent on receipts, private on internals. We publish every prediction the model has ever made and grade each one as the tournament plays out; what we don't publish is the exact weighting recipe, because that's the work. Stable under live results. As real outcomes land, completed matches lock in as certainties and propagate through the simulator — the bracket-wide probabilities sharpen naturally without us moving the goalposts.

3 · The simulator

Tournament-wide probabilities — who reaches the round of 32, the quarter-finals, the final, the trophy — come from a Monte Carlo simulator that runs the full bracket forward repeatedly, sampling each remaining fixture from the model and respecting completed results. Aggregated across many thousand runs, the per-team counts give the round-by-round probabilities you see on the simulator and winner pages. The number of runs is well past convergence; the meaningful uncertainty sits in the per-match model, not the sim.

4 · Live result integration

Live results come through a 4-second-timeout client with Redis caching, so a slow upstream never hangs a page. Every completed group fixture is treated as a certain prior in the simulator; remaining group fixtures and every knockout are sampled from the model. By the semi-final round only two matches remain to sample, so the champion probabilities you see become very precise. The pages refresh on an hourly cadence to balance freshness against compute.

5 · What the model deliberately does not do

It is not a tipster. We do not model injury news after the squad-submission deadline, individual key-player absence, in-tournament tactical change, weather, late substitutions, or stadium effects beyond the basic host bonus the historical record supports. The model is calibrated for the long run; on any single match, variance is high. Treat the percentages as probabilities, not predictions.

6 · How to verify our numbers

The Model Record page is the receipt — every match the model called, the live grading, biggest correct call, biggest miss, and a calibration grid that shows whether 60%-favourite calls actually win 60% of the time. A full post-tournament reconciliation, including Brier score by round, will be published within 14 days of the final.

RECEIPT

The receipt is on the Model Record page

We don't publish the exact weighting recipe, but we do publish every prediction the model has made. The Model Record page shows the headline favourite hit-rate, a calibration grid (does a 60%-favourite call actually win 60% of the time?), the biggest correct call so far, and the biggest miss — updated live as matches finish.

Backtested across the last seven World Cup tournaments (1998–2022), the model was competitive with bookmaker baselines on draw rate and group-stage favourite accuracy, with the expected miss profile on knockout coin-flips and structural shocks. A full reconciliation including Brier score by round will be published within 14 days of the 2026 final.

See the Model Record →
FAQ

Frequently asked

Why don't you publish the exact model weights?

Because the recipe is the work. We publish every prediction and grade it transparently — that's the trust signal that matters. The internal weighting is proprietary. We give journalists and researchers the per-match probability CSV on request so claims can still be verified end-to-end.

How accurate is the simulator?

The Monte Carlo simulation is run well past the convergence threshold for a tournament of this size — the per-team trophy probability stabilises to within a fraction of a percent. The bigger uncertainty is in the per-match model itself, which is why the model-record page exists.

Why don't you just use bookmaker odds?

Two reasons. First, bookmaker prices bake in the bookmaker margin and bettor sentiment — they're not pure probability estimates. Second, our value is "here's an independent model, transparently graded", not "we reflect the market". A consensus comparison view may appear post-tournament if there's demand.

How does the simulator handle the knockout bracket?

Until FIFA publishes the exact 2026 R32 seeding map, we use a re-pair-per-round abstraction — surviving teams are matched fairly at each knockout round. This is statistically valid for aggregate champion probability but slightly off on specific path lengths. We will swap to the actual bracket structure as soon as FIFA fixes it.

Can I cite your numbers?

Yes — please link back to onsidearena.com/world-cup-2026/methodology so readers can see the trust framing and the live model-record receipt. Journalists and researchers who need the underlying per-match probability CSV can email [email protected].

How often does the model update?

The model re-runs hourly via ISR; the per-fixture predictions and the simulator inherit the same cadence. As real results land, completed fixtures lock in as certainties and the remaining probabilities sharpen.

CITE THIS PAGE

Reference

For citations, please use: Onside. (2026). Onside World Cup 2026 Model — Methodology. Retrieved from https://onsidearena.com/world-cup-2026/methodology.

Journalists, researchers and FPL writers: email [email protected] for the underlying per-match probability CSV or for comment on any specific fixture.

All 72 match predictions →Champion simulator →Predicted upsets →
WC MATCHDAY BRIEF · ONE LIST, TWO SEASONS

Get the World Cup brief in your inbox — plus early FPL 2026/27 picks

Match-by-match predictions, captain picks for World Cup Fantasy, likely scorers, and the first FPL 2026/27 differentials to watch. Same engine that powers our Premier League tools, pointed at the world's biggest pre-season.

Weekly digest

Captain picks, transfer tips & differentials — every week.

One email per gameweek. Unsubscribe any time.

SAME ENGINE · POINTED AT FPL

Tracking the World Cup? Use the same engine for FPL 2026/27.

v5 depth-2 GBT stacker · 51k+ OOS predictions · MAE 0.86. Captain picks, MILP squad optimiser, AI Coach, set-piece intel — the production Premier League tools.