Trusted by 16 Champions & Europa League clubs, plus 200 more. View Pricing

Football Analytics AI Logo
Football Analytics AI Logo
Order a Report

Coverage & Credibility

How much can you trust the model in your league?

We don't claim one number for the whole world. Every league is validated on its own, scored against real transfer outcomes the model never saw in training. Search yours and see exactly how far ahead of guesswork we are.

89.1%
Directional accuracy
Across all tracked leagues
21
Leagues validated
Each calibrated locally
0.91
Model fit (R²)
Value variation explained
Out‑of‑sample
Every figure
Strict temporal holdout
All tracked leagues

Methodology: Figures are out-of-sample results for the xV valuation model on under-23 players, over a two-year horizon. “Model fit” is R² on log-value; “valuation precision” is the share of forecasts within ±30% of the realised value; “directional accuracy” is how often the model calls the direction of value movement correctly. Models are trained on data up to a fixed cutoff and scored only on transfer outcomes that occurred after it, with no lookahead and no backfit. Each league is calibrated on its own data rather than a single pooled model. “Validation depth” reflects the amount of validated history behind each league. All figures derive from proprietary data & models.

Where the signal comes from

No single source is enough.

Our models are trained on a proprietary pipeline that fuses several independent source categories, cross-validates between them, and normalises across leagues and time periods. The aggregation logic, feature engineering, and cleaning rules are proprietary.

01

Community-consensus market valuations. Longitudinal valuation histories across 50,000+ player profiles, the trajectory signal our value models are trained against.

02

Match-level performance data. Per-match event and rating data (passes, tackles, xG contributions, minutes, discipline) at the individual player level across every coverage league.

03

Professional football data feeds. Structured feeds for fixtures, squad composition, transfer records, and live standings across multiple competition tiers.

04

Official league and federation sources. Direct collection from national association and league-body publications not available through commercial providers.

05

Verified media and financial reporting. Transfer confirmations, contract details, and club financials, cross-referenced against official records.

Questions about the numbers

What does "directional accuracy" actually mean?+
It is how often the model gets the direction of a player's future value right. When we forecast a value will rise or fall over the next two years, directional accuracy is the share of those calls that matched what actually happened. It is measured on real transfer outcomes, not on other estimates.
How do I know these figures are not cherry-picked?+
Every number is out-of-sample. The model is trained on data up to a fixed cutoff and scored only on outcomes that happened after it, so there is no lookahead and no backfit. We also publish all 21 leagues, including the weaker ones, rather than only the flattering markets.
Why is my league not listed?+
We only publish a league once it has enough validated transfer history for a figure we trust. Leagues are added as that evidence builds. If you do not see yours, ask us about coverage for your market.
Why calibrate each league separately instead of one global model?+
Player development curves and market dynamics differ by league, so a model tuned to one competition does not transfer cleanly to another. A locally calibrated model beats a single pooled one, and we publish the per-league results so you can see exactly where we are strongest.
What is "validation depth"?+
It reflects how much validated transfer history sits behind a league's figure. A high-accuracy league with a shorter history is marked as still developing, so a strong headline number is never mistaken for deep certainty. We never quote a figure without it.
What data are these models built on?+
A proprietary pipeline that fuses several independent source categories, from longitudinal market valuations to match-level performance data and official league records, cross-validated and normalised across leagues. The aggregation, feature engineering, and cleaning logic are proprietary.

See the full track record.

Per-league confidence intervals, model changelogs, and the validation methodology behind every number.

View the track record