Phil's Data
Log inGet Started
77 seasons · 1950–2026 · ML-ready feature vectors

Quantitative analysis forevery NFL playsince 1950.

Statistical dashboards, percentile rankings, power ratings, and position-specific feature vectors built directly into your browser — and exposed as a JSON API for ML pipelines.

Start Analyzing FreeExplore Features

0

Seasons

0

NFL Teams

0

Players

0

Stat Lines

0

Data points

0

Years of history

0

Legendary players

0

ML feature dims (max)

The Problem

Surface-level stats don't make you smarter.

Casual fans squint at box scores. Serious analysts need rigorous quantitative tools — and a unified dataset.

Scattered Data

Stats are spread across dozens of sites with no unified view. You waste time switching between tabs instead of analyzing.

No Quantitative Depth

Most sites give you raw box scores. We give you z-scores, percentile ranks, power ratings, and ML-ready feature vectors.

Hard to Compare

Comparing players and teams across decades, conferences, and contexts is tedious without proper normalization tools.

Everything In One Place

Built for analysts, not casual fans.

A complete quantitative toolkit — every feature engineered for real statistical work.

Statistical Dashboards

League-wide trends, division leaders, and top performers in one unified view.

Power Rankings

Composite 0–100 rating combining win%, Pythagorean expectation, and point differential.

Cross-Player Comparison

Up to 4 players side-by-side with cosine similarity matrix on normalized feature vectors.

77 Years of Data

Complete records from 1950 through 2026. Era-accurate scoring environments and team founding dates.

ML-Ready Vectors

Position-specific feature vectors as JSON via /api/vectors. Built for direct input to ML models.

Percentile Ranks

Every metric ranked against position peers. Z-scores show how many std devs above/below average.

Searchable Database

Filter players by name, team, position, era. All 32 NFL teams linkable from any view.

Legendary Players

30+ Hall of Famers from Brady, Manning, Rice, Sanders, Montana — all with bios and career arcs.

Season Selector

Pivot any view to any year, 1950–2026. Decade-grouped dropdown for fast navigation.

Real Analytics, Right Here

Every stat ranked, every player profiled.

We don't just show numbers. Every metric is contextualized with percentile ranks, position peer comparison, and consistency scores. Hover any chart to see the underlying data.

  • Percentile rank against position peers (live)
  • Z-score: std devs above or below average
  • Power rating: composite 0–100 team strength
  • Pythagorean expectation: skill vs. luck
  • 5-year career trend lines
Start Analyzing
Top QBs · 2024 Season · live data preview
KCP. Mahomes
4,839 yds104.2
96th
CINJ. Burrow
4,918 yds108.2
98th
BUFJ. Allen
4,544 yds101.8
92th
BALL. Jackson
3,678 yds99.6
84th
PHIJ. Hurts
3,858 yds96.4
78th
GET /api/vectors/players/[id]?season=2024
{
  "player": {
    "id": "abc-123-...",
    "name": "Patrick Mahomes",
    "position": "QB",
    "team": "..."
  },
  "vector": {
    "features": [
      "pass_ypg", "pass_td_rate", "int_rate",
      "rush_ypg", "td_int_ratio", "passer_rating",
      "pass_consistency", "pass_trend_slope",
      "best_yards", "worst_yards",
      "fumble_rate", "sack_rate", "games_played"
    ],
    "values": [
      0.815, 0.733, 0.433, 0.512, 0.692,
      0.658, 0.834, 0.012, 0.978, 0.234,
      0.176, 0.211, 1.000
    ],
    "raw": [
      285.3, 2.2, 0.65, 25.0,
      3.38, 104.2, 0.834, ...
    ],
    "meta": {
      "subject": "Patrick Mahomes",
      "season": 2024,
      "position": "QB",
      "sampleSize": 17
    }
  }
}

For Data Scientists

ML-ready feature vectors,exposed as a JSON API.

Position-specific feature vectors for every player and team, normalized to [0, 1] and accompanied by raw values + metadata. Plug directly into your ML pipeline.

Position-aware features

QBs get 13-dim vectors with passer rating + consistency. RBs get 12-dim with YPC and reception rate. WR/TE get 11-dim with catch rate and yards/target.

Cosine similarity in-app

Compare any 2–4 players to see how stylistically similar they are based on their full feature vector — color-coded similarity matrix in the Compare view.

REST endpoints for pipelines

GET /api/vectors/players/{id} and /api/vectors/teams/{id}. Iterate over seasons via ?season=YYYY query param. JSON output, ready for fetch().

How It Works

Three steps to real analysis.

01

Sign Up Free

Create your account in seconds. No credit card. Instant access to the full platform.

02

Explore the Data

All 32 teams, 78 players, 77 seasons. Filter by era, position, conference, season.

03

Build Your Edge

Compare players, query vector APIs, export to your models. Find the patterns nobody else sees.

"Phil's Data is the first sports analytics platform I've used that actually gives me feature vectors I can drop into a model. The 77-year history is ridiculous — and the cosine similarity matrix in compare is a feature I didn't know I needed until I had it."
AD

Alex D.

Data Scientist · Sports Analytics

FAQ

Common questions.

Free forever · No credit card

Stop guessing.Start analyzing.

77 years of NFL data, ML-ready feature vectors, percentile ranks, and power ratings — all in one platform.

Create Free AccountI have an account