Four vectors · one puzzle a day · free · no account
- 4 models in rotation
- 12,966 NBA + 3371 tickers
- Era-honest comparisons
It's a dumb model.
Beat it anyway.
Take everything public about an athlete — the box score, the tracking data, the draft slot, the salary, the awards nobody remembers — and squash it down into a single list of numbers. That list is the whole player, as far as the model is concerned.
Then we hide the names and hand you the numbers. The model is wrong all the time — that's the fun part.
One puzzle a day per model. Each runs a real model on real public data — no staged scores, no decorative math. Three sports + one market.
Vector Hoops
Game 01Two real NBA seasons got fused into one impossible player. Name both.
12,966 player-seasons, each one a point in a 48-dimensional space the model learned on its own. Guesses are scored by how close you land — so a wrong answer that feels right usually is.
Vector Gridiron
Game 02Who should you actually start this week?
One net predicts a player's fantasy points and his stat line at the same time. Learning both makes it better at each — and the 32-dimensional trunk it builds along the way doubles as the map.
Vector Pitch
Game 03The same trick, played on the World Cup.
Built from StatsBomb's open event data for 2018 and 2022, normalized inside each tournament so a 2018 workhorse isn't judged against 2022 averages. The map here is still classical — honest PCA, no neural net yet.
Vector Equities
Game 04Public companies as vectors — SEC EDGAR XBRL + market tape.
3,371 tickers × 33,710 filing-years. 248 feats across 29 towers — Altman Z, Piotroski F, Beneish M, Sloan accruals, QMJ quality, plus 20-market microstructure. One 96-d MTNN has to predict sector, next-year profile, distress and payout at once. Guess the ticker from its vector.
4
Models in rotation
16,337
Hoops + Equities entities
48 / 32 / 3 / 96
Embedding dims · NBA / NFL / WC / EQ
Free
No account · no ads
A player isn't one kind of thing. He's a body, a shot chart, a contract, a draft night, a postseason. Most models pick one of those and throw the rest away.
-
01 · Towers
Seventeen towers, fused (Hoops)
Vector Hoops runs seventeen separate towers — one each for volume, playmaking, rebounding, defense, efficiency, shot mix, biometrics, tracking, form, market value, roster context, career arc, strength of schedule, team, draft pedigree, playoffs, and honors — then fuses them into a single 48-number embedding. Equities runs 29 towers — Altman Z distress, Piotroski F quality, Beneish M earnings manipulation, Sloan accrual quality, QMJ quality, investment, payout, growth, operational efficiency, market microstructure, and more — into a 96-d company vector.
-
02 · Multi-task
One vector, many jobs
That embedding gets graded on many jobs at once: cluster the archetypes, name the position/sector, rebuild the box score / financial profile, guess the salary, predict who rises in the playoffs, predict distress, payout, next-year stats. A vector that can do all of that simultaneously has nowhere left to hide a lie. That's the MTNN — a multi-tower, multi-task net.
-
03 · Per domain
Shared trunk, classical fallback
Vector Gridiron runs the same idea with a shared trunk and multiple heads. Vector Pitch now ships a true MTNN (24-d) that beats the old PCA baseline on role recovery; Equities ships a 96-d MTNN (29 towers) trained SOTA with CQS_v2 = 0.4*recall_no_wiki + 0.25*purity + 0.2*next_R2 + 0.15*sector.
Where this is going
The per-sport games still live in their own spaces (48-d hoops, 32-d gridiron, 24-d pitch MTNN, 96-d equities MTNN). On top of that, a 64-d joint embedding now folds ~20.7k player-seasons across the three sports into one shared role geometry — so you can ask what a power forward and a strong safety have in common and get a real neighbour.
Shipped with caveats: sport identity is still partly recoverable from the joint vector, and there is no joint daily puzzle yet. Equities is its own market puzzle — guess the ticker from the vector. The four games are the product you play.
- Every number is recomputable from public sources: stats.nba.com, Basketball-Reference, nflverse, StatsBomb open data.
- Era- and context-honest. Stats are normalized inside their own season or tournament before anything is compared.
- Free. No account, no ads, no tracking.
- It is called dumbmodel for a reason. It is wrong all the time. That's the fun part.
Every number on every game is recomputable from public source data — an accuracy harness gates every deploy.
Back to the roster →