Every course asks a different question. Some weeks it is length, some weeks it is a wedge to fifteen feet, and some weeks it is whether you can read grainy Bermuda at five in the afternoon. Set what this week actually demands and watch the field reorder itself. Runs entirely in your browser — move a slider, the board moves with it.
Course fit is the idea that a course rewards particular skills more than others. A long par 71 with small greens pays for approach play; a short course with heavy rough pays for accuracy and scrambling. Weight each skill by how much the week demands it, score every player against those weights, and the ranking that falls out is the fit.
This board is live in your browser every week the feed updates — there is nothing to subscribe to for that. What lands in your inbox is The Card: unfiltered player ratings and the stories PR reps tried to kill, every Friday.
Each player carries four ratings from 0 to 100, where 50 is a tour-average performer: distance, approach, short game, and putting on the surface being played. The fit score is a weighted average of those four against the weights you set. Set approach to 100 and everything else to 0 and you get a pure iron-play ranking. That is the whole model, and it is deliberately simple enough to argue with.
A fit score is not a probability. It says a course asks for skills a player happens to have. It knows nothing about form, injury, travel, the draw, the weather on Thursday afternoon, or whether someone is putting well this month. Golf is won by whoever plays best over four days, not by whoever fits best on Tuesday. Treat the board as an argument, not an answer.
The shipped field is archetypes, not real players. Every card on the board is labelled.
They exist so the model is usable and testable out of the box. A real field is loaded by editing
/data/tournament-field.json with real strokes-gained numbers from a source you can
point at. Estimating a named player's numbers by feel and publishing them would be inventing
statistics about a real person, so this tool ships without doing it.
More than almost any other single factor. Bermuda grain, afternoon Poa bumpiness and the true roll of bentgrass reward different reads and different strokes. Players with a real surface preference show it in their numbers, which is why putting is split three ways here.
No. The shipped list is archetypes, each one labelled on the board. Load a real field by editing the tournament feed with sourced numbers.
No. It ranks skills against course demands. It does not price markets, take stakes or predict winners, and a fit score is not a probability.
Into your own browser, alongside everything else in the locker. Nothing is uploaded and there is no account. Clearing browser data deletes them.