A self-built statistics project on 26 seasons of college sailing
Ranking college sailors with a model that actually gets validated
Existing sailing rankings are mostly eyeballed win totals. This project fits
a real statistical model — a temporal Plackett–Luce
ranking — to 600k+ scraped race results, and checks it against races it never saw.
Quinn Brighton · Applied Math–CS, Brown University · Class of 2027EmailLinkedInGitHub
100K+
Training races
24K+
Sailors modeled
13.7M
Parameters
70%
Held‑out accuracy
This week's predictions
Live output from the model above — win probability from 5,000 Monte Carlo race simulations per regatta, built on the predicted lineups. Check back after results post to see how it did.
Quinn's race this week
NEISA Women's Singlehanded Championship
Boston College · 09/18/2026 · 7 schools · 5,000 simulated races
1Harvard University32% win
2Yale University26% win
3Bowdoin College11% win
4Brown University10% win
5Dartmouth College10% win
6Connecticut College7% win
Rose Bowl
USC · 01/02/2027
Boston College — 82% win
KP Women's Invite
Kings Point · 09/12/2026
2
18%
Hobart and William Smith Colleges
1
61%
Cornell University
3
14%
University of Pennsylvania
KP Dinghy Open Invite
Kings Point · 09/12/2026
2
18%
Hobart and William Smith Colleges
1
29%
Princeton University
3
16%
Fordham University
ILCA North
Kings Point · 09/12/2026
2
22%
Hobart and William Smith Colleges
1
30%
U. S. Merchant Marine Academy
3
15%
Penn State University
NEISA Open Singlehanded Championship
Boston College · 09/18/2026
2
13%
U. S. Coast Guard Academy
1
22%
Connecticut College
3
13%
Boston College
Regis Trophy
Boston University · 09/19/2026
2
30%
Yale University
1
32%
Harvard University
3
10%
Dartmouth College
SAISA Women's Top 9
Charleston · 09/19/2026
2
19%
Rollins College
1
46%
College of Charleston
3
16%
University of South Florida
Susan Rogers
Cornell · 09/19/2026
2
8%
Fordham University
1
80%
Cornell University
3
6%
University of Pennsylvania
Hatch Brown Trophy 2026
MIT · 09/19/2026
2
22%
Boston College
1
24%
Brown University
3
15%
Roger Williams University
SAISA Open Top 9
Charleston · 09/19/2026
2
17%
University of South Florida
1
18%
Florida State University
3
13%
College of Charleston
Mystic Lake Invite
Tufts · 09/19/2026
2
25%
Boston College
1
30%
Roger Williams University
3
19%
Massachusetts Institute of Technology
Jack Boehringer '52 Open Invite
Cornell · 09/19/2026
2
30%
Fordham University
1
42%
University of Pennsylvania
3
19%
Hobart and William Smith Colleges
Donaghy Bowl
U. Connecticut · 09/19/2026
2
9%
Sacred Heart University
1
79%
Brown University
3
5%
University of Massachusetts at Amherst
GLIOR J105 Division
Chicago · 09/26/2026
2
17%
Harvard University
1
17%
Brown University
3
16%
U. S. Naval Academy
SMC Open
St. Mary's · 09/26/2026
2
29%
St. Mary's College of Maryland
1
44%
U. S. Naval Academy
3
14%
University of Pennsylvania
SAISA Open Top 6 Invite
Eckerd · 09/26/2026
2
19%
North Carolina State University
1
20%
Florida State University
3
17%
College of Charleston
JAX Open
Jacksonville · 09/26/2026
2
39%
Rollins College
1
44%
Jacksonville University
3
8%
Palm Beach Atlantic University
Cayuga Clash
Cornell · 09/26/2026
Cornell University — 100% win
GLIOR PHRF Division
Northwestern/Chicago · 09/26/2026
2
24%
University of Wisconsin
1
25%
Northwestern University
3
17%
University of Notre Dame
Seneca Showdown
Hobart & William · 09/26/2026
Cornell University — 84% win
CAZENOVIA FALL OPEN
Syracuse · 09/26/2026
2
24%
Rochester Institute of Technology
1
57%
Queen's University
3
17%
Penn State University
Boiler Cup
Purdue · 09/26/2026
Saint Mary's College — 100% win
Bowdoin Mid Week Invite
Bowdoin College · 10/01/2026
2
25%
Maine Maritime Academy
1
47%
Bowdoin College
3
23%
Bates College
WACC Round 1B
Bowdoin College · 10/03/2026
2
26%
Brown University
1
41%
Harvard University
3
17%
Roger Williams University
WACC Round 1A
Dartmouth College · 10/03/2026
2
23%
Tufts University
1
35%
Yale University
3
19%
Dartmouth College
Smith Trophy
Massachusetts Institute of Technology · 10/03/2026
Real output from the live model — weekly top‑3‑sailor average rating
for four ICSA programs, June 2024–April 2026. Ratings drift gradually
week to week instead of swinging on a single regatta, because that drift is
exactly what the model's smoothness prior constrains.
How it works
Every sailor gets a latent skill value θ[sailor, week]
for every ISO week they've been active — a full trajectory, not one static number.
It's fit by maximum likelihood under the Plackett–Luce distribution (the standard
way to model a full finishing order, not just win/loss), with an L2 anchor plus
log–cosh penalties on week‑to‑week velocity and acceleration so skill can
drift realistically instead of chasing noise — trained in PyTorch on GPU, batched
by fleet size, over roughly 13.7 million parameters.
A second tool, the Boat Speed Analyzer, is built on the same data:
it compares a school's fleet-race performance across two time windows and flags
whether the difference is statistically significant, so a real speed change isn't
confused with normal race-to-race noise.
Validated, not just fit
The model is trained on 100,000+ races across 24,000+
sailors, and checked against a random 1-in-20 sample of regattas held
out before fitting — races it never saw during training.
70%
of the time, the model correctly predicts which of two sailors
finished ahead — on races it never trained on.
Why I built this
College sailing rankings are usually eyeballed or based on raw win totals — I wanted something that could actually be validated against held-out results.
As Club Captain of Brown Sailing, I use this data to make real decisions about lineups and scheduling, not just as a portfolio piece.
It's the project that convinced me I want to work somewhere that rewards building real statistical models against real data.
