Europe defeated the United States 16.5-11.5 to win the 2023 Ryder Cup—the 5th consecutive victory for the home team in the biennial matches.

In the past we watched and digested Ryder Cups without strokes-gained data, which made it difficult to draw any substantive team or player-level conclusions from the result due to the inherent randomness of team match play. So this year we decided to spend a good chunk of our week collecting and cleaning (to account for concessions) strokes-gained data from the matches, hoping it would add context to the final outcome. While the strokes-gained data isn’t perfect, it does a good job of separating a player's performance from that of their partner and direct opponents, helping us better identify who played well or poorly throughout the week.

Let's start with the team totals in each strokes-gained category:

Europe beat the US by 39 strokes over the 36 rounds played by each team (counting 2 rounds per Fourball). As you would expect, the majority of those strokes were gained during the two Foursomes sessions where the Europeans outscored the Americans 7-1. Europe only won 22 more holes overall than the US which, given their 39-stroke advantage, means they won more holes by multiple strokes or lost fewer strokes from their non-counting partners in fourballs than the US. We don’t have any historical data to compare it to, but it seems like Team USA was lucky to only lose by 5 points given the staggering SG deficit.

Now that we have a decent idea of what happened in the aggregate, let's drill down to the player level. Here's a summary of points earned and cumulative strokes-gained by player:

Team Europe had the four best strokes-gained performers in Hovland (+9.7), Rahm (+8.7), Fleetwood (+7.8) and McIlroy (+7.6). The Big 3 (sorry Fleetwood) were never paired together in team sessions and combined for 10.5 out of a possible 14 points.

On the American side, Homa and Cantlay both played well but lackluster performances from key players in Scheffler and Schauffele made it difficult for the Americans to put up much of a fight. Thomas and Spieth, who ranked 21st and 23rd respectively in strokes-gained, managed to ham-and-egg their way to half a point on Friday afternoon but ultimately couldn’t plug the holes in the US lineup.

Bob MacIntyre got the most out of his week—thanks in large part to Justin Rose and some soft opponents—earning 2.5 points in 3 matches despite being the 3rd worst strokes-gained performer.

Ryder Cup Musings

Before, during, and after the playing of the 2023 Ryder Cup, we had several discussions with each other about the public discourse surrounding the matches. While we typically bite our tongue on more general golf topics, we figured it was worth putting down a somewhat disjointed collection of our Ryder Cup thoughts...

To explain why Team Europe won, it’s really hard to look past the basic facts presented in the SG data above. Course setup, optimal pairings, team camaraderie—it almost feels silly to discuss these things when the Europeans were 20 shots better than the Americans off the tee alone over the course of the event. It’s a hopeless task to try to explain this massive advantage by pointing to specific things Europe did. The unsatisfying explanation for most of it is simply that the Europeans mashed a lot of drivers down the middle while the Americans did not, and we don’t really know why that was the case.

It’s important to remember that to the extent that anything influences the outcome of a Ryder Cup, it has to eventually operate through these strokes-gained numbers. You can have better SG and lose a Ryder Cup—because it’s not a stroke play event—but the only thing that the players can control is the quality of each golf shot they hit, and strokes-gained measures that. So if Europe had a better team room, and you think this helped them win last week, it’s probably useful to come up with a ballpark number for how much that could realistically have improved their SG numbers.

The SG data also highlights individual performances, which makes it easier to see through team-level narratives. For example, Hovland, Rahm, Fleetwood, McIlroy, and Hatton were responsible for almost all of the 39 strokes Europe gained over the US. Why was it mainly these 5 players that performed well for Europe? If their team was unified, and the leadership was strong, wouldn’t we expect all the players to respond positively to that? The obvious answer is that it might have had a small positive effect for all the players, but because random variation in golf performance is huge, we still have big over and under-performers within the European squad.

One of the big themes in recent Ryder Cups has been the importance of using analytics to form pairings and set up the golf course. While it’s been great to see the use of statistics embraced at the highest level of golf, its ability to impact the outcome of this event in particular has become wildly overstated by the media and even some analytics people. Just as with the narrative-based explanations above, it’s useful to think through how much of Europe’s 39-stroke advantage could reasonably be explained by things like optimizing course setup and pairings.

Anyone who watched the NBC telecast last week was treated to Paul Azinger repeatedly stating that Team Europe had “tricked out” Marco Simone to exploit their advantage on long-iron play and neutralize Team USA’s superior wedge play. As it turns out, Azinger was right: coming into the event Europe was the better team from 200+ (and weaker from inside 150), and Marco Simone featured more shots than the typical course from that yardage. However, the margins here are small: we estimate that this gave Europe at most a 0.06 stroke advantage per round. (For anyone interested, there is a detailed walk-through that describes how we arrived at this number at the end of the newsletter.)

Six hundredths of a stroke per round seems small, but over the course of a Ryder Cup (36 rounds per team) it adds up to 2.2 strokes. This still pales in comparison to Europe’s overall SG margin last week, but it would provide a meaningful edge: according to our Ryder Cup model, 0.06 strokes per round translates into a 2-3% increase in pre-event win probability. Changes to course setup are probably the cleanest edges a team can manufacture through analytics. Edges from optimizing Foursomes pairings are likely smaller, and also need to be balanced with non-analytics considerations like whether the players play similar golf balls or how well they get along.

As above, the point here is not that analytics doesn’t matter, it’s that we aren’t going to be able to attribute that much of any specific Ryder Cup outcome to it. One other odd thing with the analytics discussion last week was that despite every pairing being heavily scrutinized, which might make a difference of a tenth of a stroke, there wasn’t much said about Bob MacIntyre playing a session where Jon Rahm sat, which right away cost the European side 2 strokes in expectation! The justification for this was that Rahm was resting for his singles match, but it seems doubtful that a rested Rahm is anywhere near 2 strokes better than a tired Rahm on Sunday. (Europe's win probability was also 95% at this point, which likely made it easier to sit Rahm.)

The one explanation that does seem to hold water is the home-country advantage. With 5 straight wins and 10 of the last 12 going to the home team, it’s hard to argue against it. Home-field advantages also exist in other sports, so it makes sense it would have an effect in golf as well. However, keeping with the common theme here, it’s hard for us to attribute that many strokes to it. Can playing in front of an away crowd really make Scheffler a half shot worse per round? Maybe, but it seems pretty unlikely. It’s almost not worth saying, but the role of randomness in Ryder Cups is easily large enough to create the illusion of a home-field advantage that’s much bigger than it truly is.

Course Setup Walk-through

According to our approach data, over the last 12 months players on Team Europe were on average 0.025 strokes worse than Team USA from 50-100 yards from the fairway, 0.014 strokes worse from 100-150, 0.01 strokes worse from 150-200, and 0.022 strokes better from 200+. So Azinger had good intel: the US’s largest approach advantage was on wedge shots, while Team Europe’s advantage lay with long irons. (If these numbers seem tiny to you, consider that the best off-the-tee player might gain 1 stroke per round; given 14 tee shots per round, this works out to 0.07 strokes per shot.)

For these approach differentials to provide an advantage to Team Europe, more shots need to be hit from their favored distances. We know the typical approach distribution on the PGA Tour, and we also have the approach distribution from last week at Marco Simone. The net effect was that players were hitting about 2.5 fewer shots per round from 100-150 yards and 1.3 more shots from 200+ last week (there were fewer approach shots overall at Marco Simone because of the 3 driveable 4s). Combining this with our skill differentials above, we get an overall benefit of 0.06 strokes per round for Team Europe at Marco Simone compared to a typical PGA Tour venue.

It’s worth mentioning that this is likely an overestimate, because when predicting performance from noisy data you always expect gaps to narrow (e.g. Europe’s 0.02 stroke advantage from 200+ might only be predicted to be 0.01 strokes going forward). Also, to simplify things we looked at performance differentials on fairway shots, but of course some of the approach shots last week were from the rough where margins are smaller. But the point of this analysis is not for it to be perfect, it’s to illustrate where an edge can come from and roughly how large it can be.