← All projects

Personal project

Scorigami

A scorigami is a final score that has never happened before. I built interactive maps of every score in MLB and Big East basketball history, then designed a jackpot game around the idea using NFL scores.

Try it

Both maps are single self-contained pages. Each cell is a final score, colored by how often it has happened.

MLB: every score since 1871

The MLB map is built from Retrosheet game logs going back to 1871 and covers all 30 teams. A Python script reads every season file, counts each final score, and writes one HTML page with the data inside it.

A second script answers one question after a Cincinnati Reds game: was that score a scorigami? It accepts the score in any format, looks it up, reports how many times it has happened and when it last did, and drafts a post ready to share.

Big East basketball

"Big Eastagami" does the same for Big East men's basketball from the 2014 season through 2026. It adds a team and season filter, an overtime marker on each cell, a banner for the most recent scorigami, and a panel that estimates the most likely next one.

Scorigami Jackpot: a game concept

The third piece is a working demo of a game. Every NFL score that has never happened carries its own jackpot, and players buy tickets on the scores they think will hit.

  1. Board1,074+ unseen scores, each with a jackpot
  2. TicketA player picks a score
  3. Eligibility lockTied to games not yet kicked off
  4. ResultsA week's games are finalized
  5. PayoutOnly eligible tickets win

The hard design problem was fairness. Without a rule, a player could wait until a game was nearly over and buy the score on the board. So each ticket is permanently linked, at purchase, to the games that hadn't kicked off yet. A ticket bought after kickoff can't win on that game.

The demo is a Python web server and a SQLite database with no external packages. It simulates three NFL weeks, so the eligibility rule can be shown end to end. It's a concept demo with play money, not a live product.

What it shows

  • Turning a large public dataset into one clear picture.
  • Shipping a finished, shareable result as a single file with no server.
  • Working out a rule system and proving it with a database and a simulation.

MLB data: the information used here was obtained free of charge from and is copyrighted by Retrosheet. Interested parties may contact Retrosheet at www.retrosheet.org.