Pro Golf Now
Fansided

The Desert Classic upset: Statistically, it’s a Long story

By Bill Felber

LA QUINTA, CALIFORNIA - JANUARY 20: Adam Long of the United States shakes hands with Phil Mickelson of the United States after winning the final round of the Desert Classic at the Stadium Course on January 20, 2019 in La Quinta, California. (Photo by Donald Miralle/Getty Images)
LA QUINTA, CALIFORNIA - JANUARY 20: Adam Long of the United States shakes hands with Phil Mickelson of the United States after winning the final round of the Desert Classic at the Stadium Course on January 20, 2019 in La Quinta, California. (Photo by Donald Miralle/Getty Images) /
facebooktwitterreddit

The Desert Classic was just that, as Adam Long came from seemingly nowhere to take down Phil Mickelson and Adam Hadwin for his first PGA TOUR title. Diving into the numbers, three specific factors made this possible.

To most golf fans, Adam Long beating Phil Mickelson (and Adam Hadwin) at the Desert Classic is something close to inexplicable. Save for the magnitude of the stage, it’s on a par with Francis Ouimet over Harry Vardon or Jack Fleck over Ben Hogan.

It’s a natural reaction. Mickelson has 43 PGA Tour victories, five of them majors, plus the 2007 Players Championship. Long is a tour rookie who came into the Desert Classic with just 10 rounds under his belt, scoring under 71 in only two of those rounds. He ranked 210th out of 217 in scoring average. Mickelson averaged 70.1 strokes last season, 25th best on tour.

Who would you have bet on?

Three factors combined to flip Sunday’s script, and each of those factors deserves more than a brief acknowledgement.

The first is the simple axiom that in a PGA Tour event, almost literally anything can happen. As I explained in greater detail in my recently released book, “The Hole Truth,” professional golf is easily the most competitive of American major professional sports. Last season, the spread between the best and worst scoring averages on tour was less than five strokes per round. That’s only a matter of seven percent, and the seven percent spread encompasses nearly 200 players.

Compare that with the standard performance spreads in professional baseball, football, basketball or hockey and the true meaning of the axiom “on any given week” becomes clear.

Adam Long Desert Classic 2019 PGA TOUR
LA QUINTA, CALIFORNIA – JANUARY 20: Adam Long of the United States lines up his winning putt on the 18th green during the final round of the Desert Classic at the Stadium Course on January 20, 2019 in La Quinta, California. (Photo by Donald Miralle/Getty Images) /

The second factor coming into play this past weekend is the data set shaping our assessment of Long. It’s not very strong, but it’s also not very deep.

Here’s the full statistical portrait. The Desert Classic was Long’s fifth start on the season. In his first start, the Safeway, he shot 66-72-74-75 for a total of 287 that brought him home in a tie for 63rd. In his three starts since then, he averaged 72.67 strokes and missed every cut.

Long’s scoring average was 2.1 percent worse than the averages of the fields against which he was competing. His Strokes Gained breakdowns were all also poor: 142nd off the tee, 144th approaching the green, 216th around the green and 218th putting. He ranked 216th of 218 in scoring average.

The normal way of examining a player’s scoring potential based on his existing data set is to delve about two standard deviations deep, that depth encompassing about 97 percent of his scoring potential. The standard deviation of Long’s scores entering the Desert Classic was 0.89 strokes, giving him a potential low score of 69.03 strokes on a standard PGA Tour course.

Since the Desert Classic is contested on courses that tend to play about a half stroke easier than usual, that potential low requires an adjustment that takes Long’s potential low down to about 68.53 strokes. Instead, he shot 63-71-63-65.

More from Pro Golf Now

That level of over-performance is only possible due to one reason, data insufficiency. Remember that our entire statistical portrait of Long is based on just 10 rounds, a total that is unlikely to be large enough to base reliable performance indicators on.

When Long has had a chance to compile a full season’s worth of data, we may look back on this past weekend as his coming-out moment, just like the 1991 PGA at Crooked Stick represented John Daly’s coming out moment. It’s also possible he retreats into the oblivion of the field he occupied so anonymously one week ago.

The third factor influencing our assessment is Phil Mickelson, who may be the flightiest player on tour.

That unpredictability is, of course, one of the reasons the world loves Phil. But it is very real. We’ve already noted that last season Phil ranked 25th on tour in scoring average. That’s very good. But the standard deviation of his performances was 3.43 strokes, making him an extraordinarily chancy bet.

The two standard deviation range of Adam Long’s performance entering the Desert Classic would have been calculated at about 3.5 strokes. That’s pretty normal on tour. By contrast, Mickelson’s two standard deviation range was 13.7 strokes. Given that predictable performance spread, why should anything Phil does surprise us?

Next. Adam Long stares down Phil Mickelson to claim Desert Classic victory. dark

Not that Phil threw in any major clunker rounds in the desert. Among the top 17 finishers this weekend, the average spread between best and worst rounds was just 4.4 strokes. Nobody equaled Phil’s 9-stroke range and the only one who came close was Long, whose pair of 63s and 71 created an 8-stroke performance range.

Home/PGA