In this series, I’ll highlight six guys that my PGA models identify as high-upside players who are projected for low ownership. Note: these will not always be my core plays. For my core, find me in FTA+ chat each week!
Finally, here’s a breakdown of my model. You’ll find I refer to the top 10/3/1 percentages a lot. In my mind, Cut Odds is a floor projection, Earnings is a median/average projection, and the probabilities of a top 10/3/1 finish are akin to a ceiling projection.
Let’s get to it.
Course Breakdown – Pebble Beach Golf Links
I hate course history. The sample size for each player is too small to get a reliable sense of how a player fits at the course, or whether that fit even matters. However, what if we could see how numerous players perform at a given course in order to dramatically increase the sample size? Thanks to DataGolf’s Historic Event Data, we can.
To begin, I take the top 25 and bottom 25 golfers in CH Index (DG defines CH Index as “the average (adj. for field strength) strokes-gained at the course”). Next, I regress the CH Index of each player on their strokes gained profiles, bogey avoidance, and birdie or better percentage to find out which stats best translate into success at the course of interest. This strategy makes the model predictive, whereas most course breakdowns are inherently more descriptive. The difference is simple; predictive tells us what will happen while descriptive tells us what did happen. DFS is all about the former. Put another way, what use does knowing iron play plays a key role if you can’t predict whose irons will be on that week? The CF Model combines what will be important with what we can best predict.
Results & Key Stats
This week, I instead used DataGolf’s Course History Tool to get a bigger sample. The strategy for building the model was exactly the same.
The r^2 this week is around .3, which doesn’t jump off the page, but is quite reliable nonetheless. It’s important to remember that the US Open set-up will play different than the AT&T Pro-Am anyway, but more on that in a bit.
The key stats and their weights are:
- SGOTT – 28%
- SGAPP – 23%
- SGATG – 28%
- SGP – 21%
CF Model Favorites
The CF Model favorites (relative to price) are:
- Rory McIlroy (1st)
- Patrick Cantlay (2nd)
- Tommy Fleetwood (5th)
- Hideki Matsuyama (7th)
- Adam Scott (8th)
- Webb Simpson (4th)
- Lucas Glover (17th)
- Byeong Hun An (24th)
- Matt Jones (27th)
Datagolf also showed that bombers gain less than usual off the tee here and more than usual approaching the greens. As they point out here, this is likely a result of everyone laying back to specific numbers (like on holes 4 and 8). Therefore, to get really specific, we want to target players who gain off the tee primarily because of accuracy, all else equal. Fortunately, the players who stand out in the model already fit this mold.
Finally, let’s talk about the US Open set-up. The fairways have been narrowed and the rough is extra penalizing. Accuracy and precision, both off the tee and approaching the green, will be slightly more important than normal, but also remember that players with higher clubhead speeds (not to mention shorter approach shots) are less affected by rough. As usual, I think the edge is not overthinking any of this, especially since the type of players it all points to is already popping.
Tommy has to be the most notoriously underpriced player by DraftKings. His true-talent ranks are 8th in top ten odds, 5th in earnings projection, and 3rd in odds to make the cut.
Furthermore, his 5th place rank in the CF Model is actually enhanced by the way in which he gains off the tee. In 2018, Tommy was 13th in SGOTT while just 23rd in driving distance. This year, he’s fourth in SGOTT despite only ranking 46th in distance. In other words, he’s as accurate as it gets off the tee. (Note – using % of fairways hit as a measure of driving accuracy penalizes those who hit it far because they have a smaller margin for error in order to hit fairways – I prefer the just-discussed analysis).
The cherry on top for Fleetwood is that his ownership projection is a little under 15% right now.
Webb is priced almost identically to his Augusta price-tag, but this time he’s going to be mega-chalk instead of the sub-5% ownership we got at The Masters. Nevertheless, he’s a core play for me. He’s fourth in the CF Model and 8th in recent form. While his true-talent metrics don’t stand out, they’ve improved consistently throughout the last couple of years and his true strokes-gained 50-round rolling average sits just below two per round, which is elite.
With only 60 and ties making it to the weekend, I will not pass up an opportunity to get an elite golfer at such a massive discount. Diversify elsewhere!
The GPP magic. With little separation in terms of ownership at the top, no one stands out as the clear path to high-upside leverage. Consequently, I will likely settle on a tight core in the upper and mid tiers, while hammering a nice mixture of the following players at extreme discounts and low ownership:
Grillo is quietly putting together a strong season, particularly in the ball-striking department. He’s gaining half a stroke per round off the tee and over .8 per round with his irons. His putter is failing him, as he’s losing half a stroke per round on the greens, but that should bounce back. He actually gained over half a stroke per round putting last season and is slightly positive on the greens in his last five events, furthering the argument for positive regression.
His recent form rank is 18th thanks to his ball striking and he’s 16th in made cut odds. Like Fleetwood, his course fit rank (40th, right in line with his price) is enhanced by the fact that he gains so much off the tee despite being 67th and 106th over the last two years in driving distance.
Grillo is projected for 5% ownership so he will be a favorite target of mine.
Glover continues the trend of targeting players who project as significantly more likely to make the cut than their price would suggest. Glover is 18th in odds to make the cut, thanks to a long term bogey rate of just 14.2% and strong driving performance. He also continues the trend of targeting players who gain off the tee without overpowering courses (.514 SGOTT & 85th in distance in 2018, .259 SGOTT & 87th in distance in 2019). Glover doesn’t come in with the same form as Grillo, but he pops more in the CF Model with a ranking of 17th.
At just 9% ownership, Glover is a great way to chase what should be a very rare 6/6 this week.
Erik Van Rooyen
EVR is making a name for himself this season with strong play in Europe and the states (and Canada!). It’s a small sample, but you have to love what he’s done off the tee and approaching greens. In just 12 shotlink rounds, he’s gaining over a stroke per round off the tee and half a stroke per round approaching the greens.
He’s now played five PGA sanctioned events, with the following finishes, courtesy of Fantasy National:
Again, there’s some sample size concerns here, but the uncertainty that comes with that is suppressing his ownership a bit.
Byeong Hun An
Back to our FTA darling, Benny An! An ranks 24th in the CF Model and there’s reason to conclude he;s in fine form despite a disappointing rank of 56th in recent form. Thanks to Fantasy National, we can dissect his last ten events from a ball-striking perspective:
As you can see, his ball striking has consistently been there, outside of one horrendous showing off the tee. The Recent Form model can’t recognize that event as an outlier and adjust accordingly. We can.
An is 27th in top ten odds, 19th in earnings projection, and 33rd in odds to make the cut. It looks like we’re going to get him at or below 5% ownership, too.
Matt Jones continues to pop for us this season. He is 27th in course fit and 29th in recent form. His true talent metrics don’t measure up, but it’s understandable when looking at his last three years:
The true talent model is thrown off by a career-worst slump in 2018, but Jones is playing his best golf since 2016. I’m more inclined to buy his recent improvement since he’s played at or above this level many times in his career. If you’re searching for a 1% play, this is it!
Fade – Henrik Stenson
Two players are projected for over 20% ownership: Webb (core play!) and Stenson. Though Stenson fits the narrative of shorter, accurate ball strikers, his numbers don’t actually support it this year. For example, in his last ten events, he has gained strokes off the tee just twice, and gained a total of just 1.2 strokes in those two events. What he has done well is stick irons close, as he always does. However, even with his stellar iron play, he is 67th in course fit and 41st in recent form. One of the main conclusions in my model research was that the industry over-values iron play. I believe this is exhibit A.
Another reason for Stenson’s massive ownership projection is the fact that he’s coming off an eighth place finish in Canada. It was his best finish of 2019, sparking many “he’s back!” takes, but it should be noted that he gained 4.6 strokes putting, too. That was his best putting week since March 17th of last year at Bay Hill, so good luck if you’re hoping for a repeat there.
In the past several weeks, choosing a fade was really difficult. This week it was rather easy. Let’s hope it pans out.
Good luck this week and find me on Twitter at @alexblickle1!Fantasy Golf