The problem
The Old Way
PERFORMANCE BASED ONLY ON PAST OR PRESENT STATS
GENERIC RATINGS DETACHED FROM CONTEXT
REACTIVE DECISIONS AFTER PLAYER DECLINE
NO LINK BETWEEN TRANSFER TIMING AND FORECAST
UNIFORM TREATMENT ACROSS PLAYER TYPES
The New Way
FUTURE PERFORMANCE WITH PERSONALIZED MODELS
ELO SCORES ADJUSTED FOR LEAGUE POSITION AGE ROLE
EARLY WARNING FOR UNDERPERFORMANCE
OPTIMAL EXIT OR RENEWAL POINTS
EVOLUTION BY ARCHETYPE CLUSTERS AND CONTEXT
Technical Description
Metrics definition
The algorithm compares observed vs. expected performance in:
-
Team impact
-
Duel success
-
Historical form
ELO Base Formula
The dynamism of the ELO
The dynamism expressed through coefficients. These coefficients are adjusted based on the playing styles of the players that maximize points earned and victories. In this way, the ELO becomes a dynamic system, adapted to shifts in playing styles and trends, distinguishing between good players and players who are currently optimal for today’s style of football.
Details:
This simulation visualizes how a player's projected ELO trajectory evolves over time under a hypothetical transfer scenario.
In this case, a move to a more competitive league.
The Forecast ELO Algorithm empowers clubs with predictive insight into player performance. By analyzing historical data and contextual variables, WINNING forecasts each player's evolution, identifying future peaks (MxELO), optimal windows (mxELO), and performance volatility with confidence intervals.
Other Objetives
Trajectory
Detect long-term performance trajectories based on contextual and historical data. Helps clubs anticipate dips or surges in performance before they become obvious.
Predictive
Long-range
Alignment
Match players’ projected ELO with tactical systems and league demands. Ensures future performance fits not just the team but the competition context.
Contextual
Tactical-fit
Timing
Identify optimal transfer windows based on projected ELO inflection points. Prevents selling too soon or too late by anticipating value peaks.
Strategic
Time-sensitive
Clusters
Group similar player evolution profiles to uncover high-potential patterns. Allows clubs to spot undervalued talent with upward trends.
Scalable
Pattern-based