FairValue

Checking model

Esc
  • EnvironmentOverviewWhat FairValue measures, what it found, where to go next.
  • EnvironmentTerminalExplore, compare and track players; market, league and model views.
  • EnvironmentStoriesValuations, head-to-heads and research findings, told as stories.
  • TerminalPlayer ExplorerSearch, filter and sort the player export.
  • TerminalComparison StudioTwo to four players on one scale.
  • TerminalWatchlistsNamed collections, kept in this browser.
  • TerminalMarket LandscapeHow observed value is distributed, and where the model drifts.

↑ ↓ to move · Enter to open · Esc to closeLoading player index…

Terminal Models

Valuation Studio

Build a profile or load a research player. The served model returns a value, its tier, Model A against Model B and the inputs that moved it. Every result says whether it is live, unverified or cached.

Profile

Checking the model service

Research players are real; others illustrative
Model

There is no nationality field. The model never sees nationality, citizenship, confederation or birthplace, and the API rejects any request that includes them.

Player
yrs
Optional
cm
Position
Preferred foot
Club and leagueUsed by Model A only
League, last 365 days. Optional
Playing timeClub matches in the two years to the valuation date
min
%
min
days
%
%
Output and disciplinePer 90 minutes played

Ready to value this profile goes to the served model

Age
24.5
Role
Centre-Forward
League
LaLiga
Minutes
4,200
Goals / 90
0.45
Assists / 90
0.15
Model
Model A

The value scale log scale, €20k to €250M

EmergingEstablishedEliteWorld Class€2M€10M€30M€100k€1M€100M
  1. Emergingunder €2M
  2. Established€2M – €10M
  3. Elite€10M – €30M
  4. World Class€30M and above

Boundaries at €2M, €10M, €30M: the 50th, 80th and 95th percentiles of market value in the training population. Fixed definitions, not model outputs.

What comes back

  • Estimate in euros from the selected model, with its typical miss on unseen players
  • Tier from the value, and a separate classifier's probabilities
  • Model A vs B the same profile with and without league and club
  • Why SHAP contributions of each input to this estimate