Every number on this page comes from the Cap.Space production database
Sports analytics platform

Every contract.
Every cap. One space.

Cap.Space is my sports analytics platform. It tracks salary caps, contracts, and performance across the NFL, NBA, MLB, NHL, WNBA, MLS, and the top soccer leagues, then runs its own models to tell you what a player is actually worth against what he costs.

Role
Founder, designer, engineer
Type
Sports analytics platform
Stack
Supabase, JS, custom models
Status
Live at cap.space ↗
cap.space · cap space · 2026
Cap Space NBA ▾
BKNBrooklyn Nets$16.8M
MEMMemphis Grizzlies$3.7M
UTAUtah Jazz−$13.1M
MILMilwaukee Bucks−$19.0M
CHACharlotte Hornets−$19.5M
WASWashington Wizards−$23.3M
Live from the Cap.Space database7 leagues · 34M+ data points
01 — THE SCALE

What's in the database

34M+
Data points across 90+ tables
6,349
Players tracked across leagues
33,703
Model-generated player ratings
29,716
MLB contract-history records
02 — THE TOOLS

Built for people who argue about contracts

Analysts, agents, front offices, and the kind of fan who has opinions about dead cap. Every tool runs on the same live database.

Trade Machine

Build trades that actually clear the CBA. The cap math updates as you go, so you find out immediately whether a deal is legal or fantasy.

Cap Sheet Explorer

Five-year cap projections for every team. Dead money, active allocations, and the space a front office will really have when it matters.

Contract Value Index

Market comparison for any deal. See where a contract sits against comparable players at the same position, priced by the model.

Extension Simulator

Model an extension before it happens. Change the years and the guarantees and watch the cap consequences ripple forward.

Player Comparer

Head to head on salary and performance. The comparison people usually make with vibes, made with numbers instead.

Live Cap Tracking

Real-time space for every team in every league covered, updated as contracts move, so the sheet is never a week behind the news.

03 — UNDER THE HOOD

How the models actually work

The ratings on Cap.Space aren't vibes. They come from a pipeline that turns raw play data into player values, and every number below is real.

1

Ingest

34M+ data points across 90+ tables: play grades, box scores, contracts, wages, and market values from every league, cleaned and joined into one warehouse.

2

Engineer

Up to 40 engineered features per position group, standardized so players only get compared to the job they actually do. Nearly 34,000 player seasons scored.

3

Model

A separate cross-validated regression for each position: 11 NFL groups and 6 soccer groups, each with its own replacement-level baseline. Out-of-sample R² runs as high as 0.97.

4

Value

League-specific wage curves translate ratings into an implied salary. Set that against the real contract and you get surplus value: what a player is worth versus what he costs.

Live model output: 2026 NFL surplus value leaders (500+ snaps)
Player Pos Model rating Cap hit Implied value Surplus
Byron YoungED 86.0$6.0M $38.5M+$36.0M
Puka NacuaWR 96.0$5.8M $36.2M+$32.8M
Drake MayeQB 79.4$10.0M $34.7M+$30.6M
Bo NixQB 76.2$5.1M $27.4M+$24.9M
Matthew StaffordQB 87.7$48.3M $55.9M+$23.7M
Pulled from the Cap.Space production models, Aug 2026. Rookie-deal production is where surplus value lives, and the model finds it without being told.
Projected 2027 NFL cap space, top 6 teams ($M)
0 60 120 $121.9M $64.9M ARI ATL MIA NYJ IND LAR
Real projections from the Cap.Space database, Aug 2026. Hover any bar for the full team name.
04 — THE BUILD

Designed, developed, and launched end to end

Cap.Space is not a dashboard bolted onto someone else's data. The pipelines, the models, the interface, and the deployment are all mine, which is the same thing I offer clients.

Data engineering

Ingestion from many sources with entity resolution across leagues, so the same player lines up no matter which provider named him.

Modeling

Per-position regularized regressions with cross-validated tuning and replacement-level baselines, plus wage curves fit per league.

Product design

A data-dense interface that stays fast and readable, because the best model in the world is useless if nobody can navigate it.

Want this pointed at your data?

The pipeline that prices athletes can price talent, contracts, or inventory in your business. That work runs through TFC Business Solutions.

See TFC Business Solutions