Course Profiles / TPC Toronto at Osprey Valley (North Course)
Course profile · no validated fit signal

TPC Toronto at Osprey Valley (North Course):
what wins here.

RBC Canadian Open · 2 editions, 2025–2026 · 885 rounds analyzed

What separated the leaderboard
OTT 23% · tour 19%
APP 27% · tour 34%
ARG 11% · tour 14%
PUTT 40% · tour 33%
Each bar is that category’s share of the spread in strokes gained among the players who finished, across two editions (2025–2026, 142 finishes); the mark is its share in a typical tour week (293 editions). Two editions of category data describe those weeks, not the course, so no difference from a typical week is called out.
Course fit is not scored here. It first appears in our archive in 2025, after the 2019–2023 seasons the weights are fitted on. When this course hosts, the Lab ranks the field on player quality alone.

How it plays

Not what the scorecard says — what this course does to the same players, measured against everywhere else they play. Built from 885 rounds across 2 editions, and ranked against every course in the archive.

MeasureVersus elsewhereWhich meansAcross the archive
Driving distance +6.6 yds longer off the tee here 13th longest of 90
Fairways hit +3.4 pp easier to find 27th easiest to find of 90
Greens in regulation +3.5 pp easier to hit 18th easiest to hit of 89
Scrambling +1.1 pp easier to save par middle of the pack

Scoring. These players average 1.00 shots fewer per round here than they do elsewhere.

Plays most like. Vidanta Vallarta, The Old White TPC, TPC Deere Run — the closest courses in the archive on those four measures together.

A record of what happened, not a forecast. Estimates are shrunk toward the tour mean by how much evidence each course has, and every baseline leaves the course itself out, so a venue cannot define the average it is measured against. Split-half reliability 0.888. Weather is not controlled for.

What it has taken

The numbers the week has actually demanded, edition by edition, counted from the scores rather than from anyone’s summary of them. The cut line is the worst halfway score that still saw a Saturday.

-17
typical winning score
Range -18 to -17 across 2 editions.
-13
to reach the top ten
The tenth-place score, median of 2 editions.
-2
to see the weekend
Where the cut fell, median of 2 editions.
YearFieldCutTop 10Won with
2026 146 -2 -12 -17
2025 153 -3 -14 -18

The week, round by round

Scoring average for each round of the tournament, drawn against this course’s own four-round average. Round 1 has played 1.47 of a shot easier than round 3, the widest gap in the week.

hardereasier
ROUND 1
-2.84 avg
ROUND 2
-2.15 avg
ROUND 3
-1.37 avg
ROUND 4
-1.53 avg
Longer to the right is easier scoring than a typical round here; longer to the left is harder. Built from 568 rounds — every player who completed an edition at this course, counted in all 4 of his rounds. Only finishers are in it, because Thursday is played by the whole field and Sunday by the half of it that survived the cut: compare those as they fall and every course on tour looks like it eases off for the weekend. Weather is not controlled for, so a single storm-hit afternoon is in these numbers along with everything else.

What the data says

None of this is a forecast. Course-fit weights are kept only where they predicted results on seasons the model had never seen; TPC Toronto at Osprey Valley (North Course) has not been through that test, for the reason given above, so the Lab does not score course fit here.

How the winners won

Strokes gained per round for every champion here since 2019. The highlighted figure is that winner's best category for the week. A dash means the category split is not published for that event — only the total is.

YearWinnerTo ParOTTAPPARGPUTTTotal
2026Bud Cauley-17+0.63+0.46+0.74+1.67+3.50
Method. What separated the leaderboard splits the spread in finishers’ strokes gained exactly into its four categories, edition by edition, and sets it beside every tour edition in the archive; a difference is named only when it survives resampling the editions and the players in them, and only with three editions or more. It describes weeks already played and ranks nobody. Course-fit weights come from regressing top-20 finishes on players' strokes gained in each category, using only rounds played before each event, and comparing each course against the tour-wide baseline — so a weight measures what this course rewards more than golf generally does, not simply which skills help. Weights are kept only where they predicted results out of sample; where they did not, we say so. We tested our previous method, found it did not predict, and published that too: how the model works.

The field, ranked before they play.

Every tournament week we score the field on course fit and player quality. Lab Notes lands every Tuesday with what the numbers say, free.

Free every week. No spam. Unsubscribe anytime.

More: all course profiles · live tournament stats · the research series