01
Data Collection
We pull and cross-reference hundreds of data points before every game: lineups, referees, weather, injuries, travel fatigue, and line movement. We ensure the tightest possible context.
EdgeDesk Methodology
How We Grade Projections
Inside The EdgeDesk research engine. No guesswork, no emotional bias. Just pure data, disciplined modeling, and transparency.
EdgeDesk Engine Flow
01
Data
02
Normalize
03
Features
04
Score
05
Gate
06
Record
The 6-Step Process
Every projection passes through a rigorous six-stage pipeline before publication. Here is how we ensure only the best opportunities reach your dashboard.
02
Normalization
The system normalizes all inputs without bias. It aligns historical and recent statistics into a standardized format so the engine can process them cleanly.
03
Feature Building
We transform raw data into predictive features. For instance, in baseball, we cross bullpen pressure with opposing swing-and-miss tendencies. In basketball, we evaluate pace efficiency.
04
Scoring Algorithm
The model assigns probability and margin of victory to every matchup. Without emotional interference, it spots discrepancies between our mathematical projection and the market odds.
05
Quality Gate
Not every model output becomes a published projection. The quality gate strips out selections with extreme volatility or thin margins, publishing only the opportunities with true value.
06
Reporting & Grading Loop
Every projection is published with detailed notes and recorded into the public ledger. We rigorously grade every outcome, comparing against closing lines to calibrate future accuracy.
Philosophy
Core Principles
These four principles govern every decision the engine makes. They are not aspirational ideals — they are rules coded into the system.
01
No emotion
Every variable is processed mathematically. Feelings do not enter the model.
02
No forced projections
If the engine finds no value, we do not publish. We prefer no projection over a bad projection.
03
Public ledger
Every win, loss, and push is logged. We do not edit or hide outcomes.
04
Continuous calibration
We compare our confidence bands against real outcomes to constantly improve.
Transparency Without Excuses
Unlike many services that hide their losses, we log every single projection publicly. We believe in our model long-term, and we let the ledger speak for itself.
Verifiable public record
Separate product record buckets
Ready to research with real data?
Join free and access the dashboard preview, how we grade docs, and the public archive of graded projections.