Masters: Trying to predict the winner? Don’t bet on it.

The Masters is just around the corner, with gamblers, fantasy players and casual fans looking to predict the winner. However, week-to-week performance variations make actually calling a major event’s outcome close to a fool’s errand.

For the next few weeks, fantasy sports sites and real-world gaming centers alike will do a brisk business catering to people who believe they can predict the winner of the Masters.

It’s a trending activity given the profusion of theories regarding what constitutes the winning “formula” at Augusta National. Beyond that, both in legal gambling centers and also in country club polls across the nation where the stakes extend beyond “entertainment purposes only,” the lure of imagined riches draws folks to wager on the winner.

This year in particular, a recent U.S. Supreme Court ruling making it easier to bet on sports generally will only fuel interest.

For dedicated golf gamblers, there’s only one problem: actually projecting the winner of the Masters – or any golf tournament on a week-to-week basis – comes pretty close to being a fool’s errand. Despite what your cousin with the inside info tells you, there is no reliable method of predicting, from week to week, which player will out-perform all the others.

The numbers are as clear on this as it is possible to be. Begin with the raw problem of the concentration of talent on the PGA Tour. To date in 2019, 211 players have competed in one or more PGA Tour events. From best to worst, the spread of their stroke averages totals only a fraction more than four per round. That means an average of more than 50 players are differentiated by just one stroke.

Most Masters bettors, of course, aren’t interested in 200 players; their eyes and brains are trained on the movements of the top 15 or 20. So let’s confine ourselves to the top 15 in the current FedEx Cup standings.

In predictive statistics –political polling being the example most people would be familiar with – experts generally seek what is called a 95 percent confidence level.  In other words, they want to  know the margin across which one can say with a 95 percent degree of confidence that a result will occur.

Applied to golf, we can calculate this using a two standard deviation rule, since two standard deviations covers 95 percent of any mathematical set of data.

The question then becomes, for any of our top 15 players, what is the two standard deviation spread of their performance? In the table below, I’ve laid out that two standard deviation spread based on each player’s actual Strokes Gained Scores in the four major component categories – driving, approach shots, chips and putting – for the 2019 season to date.

The table shows the actual standard deviation spread for each player as well as his seasonal average score and his resulting two standard deviation range that we can, with a 95 percent confidence level, expect the player to shoot in any given round:

          Player                            2 SD Spread              Average                  Normal Range

• Dustin Johnson                 6.142                     69.580                   63.438 – 75.722
• Rory McIlroy                      2.430                     69.594                   67.164 – 72.224
• Xander Schauffele           5.438                     69.890                   64.452 – 75.328
• Matt Kuchar                       5.442                     69.879                   64.437 – 75.321
• Paul Casey                          7.666                     70.219                   62.553 – 76.885
• Charles Howell III              4.802                     70.144                   65.342 – 74.946
• Gary Woodland                 6.016                     69.857                   63.041 – 75.873
• Justin Thomas                   6.268                     69.432                   63.164 – 75.700
• Marc Leishman                 5.432                     70.271                   64.839 – 75.703
• Rickie Fowler                     6.092                     69.827                   63.835 – 75.919
• Brooks Koepka                  6.394                     70.472                   64.078 – 76.866
• Phil Mickelson                   6.080                     70.345                   64.265 – 76.425
• Justin Rose                         5.876                     69.858                   63.982 – 75.734
• Bryson DeChambeau      5.386                     70.093                   64.707 – 75.479
• Keith Mitchell                    6.882                     70.768                   63.886 – 77.650

Using Dustin Johnson as an example, let me simplify this for you prospective gamblers. Based on his performance to date, all we can say with 95 percent confidence is that when he gets to Augusta National, Dustin Johnson will shoot rounds between 63 and 76. That translates to an expected four-round score somewhere between 252 (36-under par) and 304 (16–over par). Is that helpful, gamblers?

I didn’t think so.

Essentially the same is true of all of the game’s other top players. Among them, the most predictable player – based on his 2019 results – is Rory McIlroy, with a two standard deviation spread of 2.43 strokes against a 69.594 average. But even that means the most we can say about McIlroy with 95 percent confidence is that he will shoot between 67 and 72, which over four Masters rounds puts him somewhere between 20-under and even.

Aside from McIlroy, the only other leading player with a two standard deviation range under five strokes is Charles Howell III. The 95 percent predictability of his performance is 65 to 75, which over four rounds brings him home somewhere between -28 and plus 12.

The average two standard deviation spread of our 15 most appealing players is 5.762 strokes. Across their 70.015 scoring average, that’s a predicted range of between 64.253 and 75.777. Across four Masters rounds, all we’re truly comfortable saying is that they’ll shoot somewhere between 257 and 303. Duh!

These of course are tour averages, and they are not Augusta National-specific. Yes, the course matters, and yes, we can adjust for it.

Of our top 15 players, 11 have completed at least 10 tournament rounds at Augusta, enough to provide a performance sample.  Here is the same table for those 11 using only their Masters performances. In this case, the players are listed in order of predictability:

           Player                              2 SD Spread           Average                  Normal Range

• Brooks Koepka                  3.652                     72.333                   68.681 – 75.985
• Matt Kuchar                       5.292                     72.143                   66.851 – 77.435
• Dustin Johnson                 5.620                     71.846                   66.226 – 77.466
• Phil Mickelson                   5.878                     72.667                   66.789 – 78.545
• Justin Thomas                   6.170                     72.670                   66.510 – 78.830
• Justin Rose                         6.176                     71.731                   65.555 – 77.907
• Charles Howell III              6.536                     73.962                   67.426 – 80.498
• Gary Woodland                 7.140                     73.857                   67.717 – 80.997
• Marc Leishman                 7.406                     72.778                   65.372 – 80.184
• Rory McIlroy                      7.452                      71.618                   64.166 – 79.070
• Paul Casey                          7.466                     72.333                   64.867 – 79.799

Even for Koepka, the most predictable “top-tier” player at the Masters, the normal two-standard deviation range of his performance at Augusta exceeds 3.5 strokes, meaning all we can project is that he’s likely to post a four-round score between 12-under and 16-over. For sexier picks like McIlroy, the projected ranges are even broader.

Here’s the bottom line. The stultifyingly close level of competitive balance on tour, in combination with the wide week-to-week normal performance variations among the game’s best players, makes predicting the outcome of any major – or any tournament, for that matter – far more a matter of luck than skill. And you can take that to the bank.