Welcome to the fourth chapter of Finding Frankenstein, a PGA experimental series on FantasyLabs.
Finding Frankenstein is where I leave all my opinions and narratives at the door and rely only on what the tools at FantasyLabs tell me. First, I experiment with an unreal amount of data by using our Trends tool to run trends for literally every possible metric specific to rounds played at the week’s course. Next, I use those results to construct a tournament-specific model in our Player Models. Finally, once the model is built, I take my experimentation even further, trying to create a monster player capable of taking down a tournament.
The Wells Fargo Championship Recap
There was a calamitous accident in the Lab last week.
Anytime an amateur scientist is working with acids and neurotoxins, or statistics and advanced metrics, disasters are bound to happen. Who would have known that combining Long-Term Greens in Regulation with Long-Term Eagle Score and a large dose of variance could trigger such violent explosions? Shattered beakers were littered all about the Lab, with malfunctioning statistics spilled all over the place. I still have all ten fingers but the explosion did singe off my left eyebrow.
Last week’s Frankenstein, J.B. Holmes, played the weekend but finished T53 at three over par with only 53.5 DraftKings points. Even worse, he was rostered in 27.03 percent of lineups and was a top-four chalk across all contests. He clearly was not the Frankenstein we had hoped for. A catastrophic result.
What Went Wrong Besides Everything?
In hindsight, the Wells Fargo Model was obviously not up to par (pun fully intended). Although 68 percent of the twenty-five Frankenstein-eligible golfers made the cut, not one of them ever truly contended for victory. Lucas Glover’s solo eighth-place finish at six under par was the best finish among the contenders. Only five of the 25 golfers in consideration finished inside of the top 20.
Holmes was eventually selected because of his elite eagle score, and then he went out and didn’t card a single eagle during the entire tournament, despite multiple attempts at reachable Par-5s. I recall one hole: He was nine feet away — and he three-putted for par.
Enough of the horrifying flashbacks. It’s time to build another monster. We’re mercifully moving onto The Players Championship at TPC Sawgrass in Ponte Vedra Beach, FL.
Creating the Frankenstein Model for The Players Championship
Still recovering from last week’s misfortune, I was hoping for a statistical slam dunk of sorts. However, TPC Sawgrass presented me no such luck. Still, some very interesting trends popped in terms of both profibility and Consistency. And, of course, some trends that I had assumed would return gainfully were either not at all advantageous or frighteningly unreliable.
Here are a couple of observations I made in the Lab:
- Recent Scrambling Score was consistently profitable over multiple sample sizes and became extremely consistent at the top-end of the sample.
- Recent Adjusted Round Score produced a Plus/Minus of +6.26 with 56.3 percent Consistency on the upper end of the sample and was one of the only productive trends over a larger sample.
If you plan to build your own Frankenstein model this week, all the other sliders included in order of weighting are as follows: Consistency, Course GIR, Recent Field, Long-Term Missed Cut, Long-Term AdjRd, Long-Term Birdie, Long-Term Scrambling, Long-Term Field, Recent Birdie, Long-Term GIR and just a touch of Recent Bogey.
Unquestionably, there are some distinctive metrics weighted in the top half of the model, all but assuring us that this week’s Frankenstein will be an incomparable monster for tournament play.
Creation of the Monster
Following Frankenstein guidelines, all of the true studs must be immediately eliminated from consideration. It’s a star-studded field at The Players Championship, so I’m wiping out anybody priced above $9,000 this week. When you’re trying to create a hideous wretch capable of taking down a tournament, you can’t select from the cream of the crop. As a result, all the high-priced golfers from Rory McIlroy down to Phil Mickelson were eradicated.
It’s worth noting that five of the top-ten highest-rated players in the Frankenstein Model this week were eliminated due to price, including my personal favorite, Jason Day.
In creating my Frankenstein I used the following filters in order to eliminate the majority of candidates:
- Recent Field Score, which was one of the top metrics weighted in the model, showing steady profitability over multiple sample sizes for rounds played at TPC Sawgrass
- Long-Term Birdie Score, which was weighted in the middle of the model but produced a substantial Plus/Minus of +4.09 with 59 percent Consistency in a large sample
- Long-Term Par-4 Score, which is unable to be weighted in the model but week after week shows up as a consistently profitable trend.
- Long-Term Par-5 Score, which is also unable to be considered in the model but generates a Plus/Minus of +5.36 with 55.7 percent Consistency.
Working in the Lab is tiring, but it’s rewarding.
The Players Championship Final Four
After utilizing these previous filters, I was down to The Players Championship Final Four:
To narrow the field even more, I used a few more additional filters:
- Course GIR, which produces a Plus/Minus of +5.94 with 58.3 percent Consistency: Patton Kizzire and Patrick Reed were eliminated. Of the two, Kizzire seems the more promising, having matched a boatload of other trends.
- Recent Bogey Avoidance and Long-Term Scrambling Ability, both of which produce massive Plus/Minus numbers on the elite end of samples as individual trends: Kuchar was eliminated by this double trend.
Frankenstein Revealed
Only Branden Grace was able to survive the double-whammy trend. Grace is my Frankenstein for The Players Championship. He is the hideous wretch whom the data has deemed most apt to take down a tournament this weekend.
Grace is two for two in surviving the cut at TPC Sawgrass but hasn’t finished higher than 42nd. His 69.2 Course AdjRd Score is tied for 12th-best in the field.
So load up a Frankenstein lineup, and see what kind of monstrous results you can achieve. Feel free to berate me on Twitter throughout the tournament if Frankenstein doesn’t play up to your standards.
Good luck this weekend.
