| # | School | Team | Rec. | |
|---|---|---|---|---|
| 1 |  |
Brown University | Bears | 6–1 |
| 2 |  |
Bowdoin College | Polar Bears | 6–1 |
| 3 |  |
Dartmouth College | Big Green | 6–1 |
| 4 |  |
Tufts University | Jumbos | 4–3 |
| 5 |  |
Yale University | Bulldogs | 3–4 |
| 6 |  |
University of Rhode Island | Rams | 2–5 |
| 7 |  |
Massachusetts Institute of Technology | Engineers | 1–6 |
| 8 |  |
Harvard University | Crimson | 0–7 |
Each side is the sum of its three boats (the skipper's fleet rating times 0.72 plus a team race offset learned from their past matches, plus the crew's offset and a small experience term) plus the school's team race strength and a home-water edge for the host. The win probability is a logistic function of the difference. The forecast was made before the event: the model was refitted on every match sailed before the first day, with fleet ratings as of the week before.
Fitted on the 12,273 team race matches sailed before 2019-03-03, using the line-ups that actually raced. It called 25 of 28 matches (89%) with a log-loss of 0.353 (a coin flip scores 0.693), so the event went well for the forecast. Orderliness +0.70 ± 0.39: an orderly event, more orderly than 98% of college team race events (temperature of the forecast log-odds that best fits the results, relative to the league average; nothing about the field, venue or wind predicts it, see the Weather tab). vs forecast is a team's wins minus the wins the model expected from the matches it sailed: positive means it beat its rating.
For each side, the performance strength is the value that would have made its record least surprising: the strength at which the model's expected wins over exactly these opponents equal the actual wins, with every opponent fixed at its pre-event strength. For a skipper, the same shift is applied only to their boat in the matches they sailed, with team-mates and opponents fixed; team-mates who sailed every match together therefore get the same number. Surprise is in bits: -log2 of the probability of the record at the pre-event strength minus the same at the performance strength. P(this good or better) treats matches as independent coin flips at the model's pre-event probabilities, so it ignores form within the day; read it as a rough guide.
Pre-event strength against the strength that would have made each record least surprising, everything else fixed. Δ is in the model's logit units, the same scale as the boat values above (a side is three boats plus the school term); green is a clear over-performance, red a clear under-performance. Skippers need three or more matches.
| Place | School | Strength | Perf. | Record | Exp. wins | Δ | Surprise (bits) | P(this good or better) |
|---|---|---|---|---|---|---|---|---|
| 3 | Dartmouth College | 9.93 | 11.78 | 6–1 | 3.9 | +1.86 | 2.6 | 8% |
| 2 | Bowdoin College | 10.50 | 11.54 | 6–1 | 4.9 | +1.04 | 0.8 | 31% |
| 1 | Brown University | 10.94 | 11.42 | 6–1 | 5.6 | +0.48 | 0.1 | 55% |
| 4 | Tufts University | 10.02 | 9.98 | 4–3 | 4.1 | −0.04 | 0.0 | 68% |
| 6 | University of Rhode Island | 8.87 | 8.50 | 2–5 | 2.4 | −0.37 | 0.1 | 80% |
| 5 | Yale University | 9.79 | 9.38 | 3–4 | 3.5 | −0.40 | 0.2 | 81% |
| 7 | Massachusetts Institute of Technology | 9.51 | 7.18 | 1–6 | 3.2 | −2.33 | 3.2 | >99% |
| 8 | Harvard University | 6.97 | 3.97 | 0–7 | 0.4 | ≤−3.00 | 0.6 | >99% |
| Skipper | School | Value | Perf. | Record | Exp. wins | Δ | Surprise (bits) | P(this good or better) |
|---|---|---|---|---|---|---|---|---|
| Max Clapp | Dartmouth College | 2.96 | 4.81 | 6–1 | 3.9 | +1.86 | 2.6 | 8% |
| Timothy Greenhouse | Dartmouth College | 2.99 | 4.39 | 4–1 | 2.8 | +1.40 | 1.1 | 25% |
| Christopher Keller | Tufts University | 3.21 | 3.18 | 4–3 | 4.1 | −0.04 | 0.0 | 68% |
| Alden Grimes | Bowdoin College | 3.30 | 3.04 | 3–1 | 3.2 | −0.26 | 0.0 | 81% |
| Dylan Dimarchi | Yale University | 2.91 | 2.51 | 3–4 | 3.5 | −0.40 | 0.2 | 81% |
| Trevor Long | Massachusetts Institute of Technology | 3.65 | 1.78 | 1–3 | 2.1 | −1.87 | 1.3 | 99% |
| Ivan Shestopalov | Massachusetts Institute of Technology | 3.17 | 0.84 | 1–6 | 3.2 | −2.33 | 3.2 | >99% |
| Francis Doyle | Harvard University | 2.07 | -0.93 | 0–7 | 0.4 | ≤−3.00 | 0.6 | >99% |
| Eric Hansen | Harvard University | 1.85 | -1.15 | 0–7 | 0.4 | ≤−3.00 | 0.6 | >99% |
Probability the model gave the side that went on to win. Red bars are upsets (the winner was given under 50%).
| Team | Rec. | Bears | Bears | Green | Jumbos | Bulldogs | Rams | Engineers | Crimson | |
|---|---|---|---|---|---|---|---|---|---|---|
| Brown University Bears | 6–1 | ✕ | L | W | W | W | W | W | W |
| Bowdoin College Polar Bears | 6–1 | W | ✕ | L | W | W | W | W | W |
| Dartmouth College Big Green | 6–1 | L | W | ✕ | W | W | W | W | W |
| Tufts University Jumbos | 4–3 | L | L | L | ✕ | W | W | W | W |
| Yale University Bulldogs | 3–4 | L | L | L | L | ✕ | W | W | W |
| University of Rhode Island Rams | 2–5 | L | L | L | L | L | ✕ | W | W |
| Massachusetts Institute of Technology Engineers | 1–6 | L | L | L | L | L | L | ✕ | W |
| Harvard University Crimson | 0–7 | L | L | L | L | L | L | L | ✕ |
| # | Winner | Loser | Pos. | ||||
|---|---|---|---|---|---|---|---|
| 1 | |
Engineers | 1-2-4 | def. | |
Crimson | 3-5-6 |
| 2 | |
Big Green | 2-3-4 | def. | |
Rams | 1-5-6 |
| 3 | |
Bears | 1-4-5 | def. | |
Jumbos | 2-3-6 |
| 4 | |
Polar Bears | 2-3-5 | def. | |
Bulldogs | 1-4-6 |
| 5 | |
Big Green | 1-2-4 | def. | |
Crimson | 3-5-6 |
| 6 | |
Bears | 2-3-4 | def. | |
Engineers | 1-5-6 |
| 7 | |
Bulldogs | 1-2-5 | def. | |
Rams | 3-4-6 |
| 8 | |
Polar Bears | 1-2-5 | def. | |
Jumbos | 3-4-6 |
| 9 | |
Bears | 1-2-5 | def. | |
Big Green | 3-4-6 |
| 10 | |
Bulldogs | 1-2-4 | def. | |
Crimson | 3-5-6 |
| 11 | |
Jumbos | 1-2-5 | def. | |
Engineers | 3-4-6 |
| 12 | |
Polar Bears | 1-2-3 | def. | |
Rams | 4-5-6 |
| 13 | |
Bears | 1-2-3 | def. | |
Bulldogs | 4-5-6 |
| 14 | |
Big Green | 1-2-5 | def. | |
Jumbos | 3-4-6 |
| 15 | |
Rams | 2-3-4 | def. | |
Crimson | 1-5-6 |
| 16 | |
Polar Bears | 2-3-4 | def. | |
Engineers | 1-5-6 |
| 17 | |
Jumbos | 2-3-4 | def. | |
Bulldogs | 1-5-6 |
| 18 | |
Bears | 1-2-3 | def. | |
Rams | 4-5-6 |
| 19 | |
Big Green | 1-2-5 | def. | |
Engineers | 3-4-6 |
| 20 | |
Polar Bears | 1-2-4 | def. | |
Crimson | 3-5-6 |
| 21 | |
Jumbos | 1-2-5 | def. | |
Rams | 3-4-6 |
| 22 | |
Bulldogs | 1-2-4 | def. | |
Engineers | 3-5-6 |
| 23 | |
Bears | 1-2-6 | def. | |
Crimson | 3-4-5 |
| 24 | |
Big Green | 1-3-6 | def. | |
Polar Bears | 2-4-5 |
| 25 | |
Rams | 1-2-6 | def. | |
Engineers | 3-4-5 |
| 26 | |
Jumbos | 1-2-5 | def. | |
Crimson | 3-4-6 |
| 27 | |
Big Green | 1-2-6 | def. | |
Bulldogs | 3-4-5 |
| 28 | |
Polar Bears | 1-4-5 | def. | |
Bears | 2-3-6 |
Orderliness is how tightly a fleet race’s finishing order follows the pre-regatta ratings, relative to a typical event (0): positive is the favourites’ day, negative the underdogs’. A value of ±0.03 means a rating gap counts about 3% less than on a typical day. This event is not a fleet race the rating model scores, so there is no actual value for it: this is what this venue’s fleet racing history says about the breeze on the day, read off the chart at the star. How this is estimated · compare venues.
Short write-ups of what the ratings and results show, with the numbers and how they were computed.
