Analysis of Best 37 NFL Handicappers Reveals the Extreme Difficulty of Picking Against a Spread

Best-Known Experts Fail More Often Than Not as Dr. Murray Launches Website to Pick Games
By John F Murray, Ph.D.

Who is Dr. John F Murray?

A leader in the field and profession of sports psychology, people know Dr. John F Murray as the clinical and sports psychologist in South Florida who helps prepare athletes and business folks for competition with mental training, and general counseling when needed, so that they will win more and have more fulfilling lives.  “I love helping my clients and working with them to achieve great things. It works much more often than not because mental training and sports psychology is extremely important to success but still vastly under-utilized in our society,” explains Murray.

Mental Performance Matters

“After opening my practice in 1999, I realized that being a good professional in my rare field was going to demand that I find ways to (1) tell more people first that the field and profession exists, and (2) show them that it really works! It’s the showing that it works part, along with my passion for NFL football, that led me on an exciting journey that culminated in my second book “The Mental Performance Index: Ranking the Best Teams in Super Bowl History (2011, 2013).”  In this book, Murray explained how vastly important mental performance is to success, but that it had not yet been taken seriously or even measured along with some 40 other more traditional factors used to quantify a football team’s performance. To correct this oversight, he invented a statistic to fill this void and called it the “Mental Performance Index” or MPI.  The extensive research showed that not only was this factor important to winning the Super Bowl, it was by far the most important factor in winning the big game!  He had tested whether this new statistic (which looked at a team’s overall performance including how well they managed mental aspects such as pressure, efficient execution, and reduction of errors) mattered in winning games, and not only did it matter, the MPI hit the ball out of the park. Correlations between this statistic and winning were above .80 whereas the next best statistic, turnover differential, ranged between .50 and .60. Clearly, the mental game in football had not been properly measured and he had discovered something amazing.

Support from NFL Experts

The book on this MPI was very well received by top people in the NFL including the forward written by 4-time Super Bowl champion Tom Flores, the book’s epilogue written by America’s Hall of Fame sports broadcaster Lesley Visser, and cover endorsements and quotes by Don Shula and Steve Sabol of NFL Films among many. In addition, “using MPI data I’ve gone public prior to 11 NFL Super Bowls and my record of picking the correct team against the spread is 9-2 (82%) even though I realize that is a terribly small sample size and might be due to chance. I don’t think it is due to chance, but I have to be humble with such a small number of games. The first season overall record is 54-41-3 (56%), and that is an amazing success rate for anyone who understands the challenge of picking against a spread.”

A Shift Toward Predicting Future Games

While the initial purpose in creating the new statistic was to help football teams understand more precisely how they were performing mentally, so that the coach would get great feedback leading up to the next game, he also began to realize that since he was capturing an aspect of football that had been largely ignored, there might be some predictive value in this statistic. In other words, he had already shown that mental performance was absolutely vital in winning the Super Bowl, but could he re-configure this measure to make predictions of future NFL games? This was not about gambling, and Murray has never been a gambler, but it hit him “like a ton of bricks” that if he could show that the MPI could predict future NFL games (since there is rich data that others are not getting) he would be able to more convincingly show the football and sports world how vital the mental game truly was to success, and more people would then understand the immense value of mental training and sports psychology!

Extensive Research

Murray added, “These thoughts led me on a 5-year mission beginning in 2013 to analyze almost 40 years of NFL games with a database of almost 10,000 past games. I adapted the MPI slightly for this purpose to make it easier to rate games based on available published data, rather than having to watch the games, and I rolled up my sleeves and went to work in my free time with what I considered a very fun and challenging pursuit. I involved statisticians at times, and computer experts who helped me create countless programs to properly analyze this massive database with the scientific method driven by hypothesis testing always at the forefront. While it is beyond the scope of this article to go into great detail, the end product was that I came up with an MPI derived system to predict NFL games using the scientific method, and I believe I can now predict NFL games as good or better than anyone on the planet.”

FootballShrink.com Was Born

Murray began publishing select NFL game predictions each week in the 2018 season on a website (FootballShrink.com) due to the fact that the Sun Sentinel referred to him as the “football shrink” in an early article about the MPI and Super Bowl. The Washington Post had already dubbed him “The Freud of Football” but “football shrink” was chosen for this new site. “While I am obviously far more than a football shrink in my day job as a clinical and sports psychologist, I kind of like the nickname that was given me, so I am going to use it. It’s catchy and bold.”

The Challenge of Prediction

Murray is directly quoted in the remaining text. It is very hard to beat the official spread of an NFL game by picking a side. When I first began this quest, I had visions of hitting 70% or even 65% success against the spread, but those early notions proved foolish, and I am now far wiser. 50% represents chance, or the worst performance possible. A newborn baby or a dog with his paw making NFL picks against a spread will get closer and closer to 50% success over time based on pure chance. It’s a coin flip. The spread more or less evens the teams to get equal money for the house on each side. To win money at the betting window when a person lays $110 to make a $100 bet, the person over time would need to averagae 52.38% over many picks just to break even! That 2.38% over chance is what it costs to make the bet, and that fee is called the vigorish or vig for short.

How Do the Best in the World Do?

It occurred to me that I needed to know how the best handicappers in the world were doing, so I tuned into one of the best websites in the world for this which has a panel of 37 NFL experts whose picks are published and archived, some going all the way back to 2003. I have looked at the numbers on this site over several years, and never once saw an error or an inflation in success rates. The handicappers on that site are well known nationally, publish magazines and articles in pre-season guides, and have often been doing it a long time. I recently completed a study of all 37 handicappers on this site, and their picks since 2003, and I would like to share some interesting results when it comes to making picks against a known official line. The take home message is that the overall average is below what it would take to make money!

Results:

All 37 NFL handicappers combined:
15,228 wins, 13,976 losses, 873 pushes = 52.14% against the spread (-323 on average)

Top 10 NFL handicappers based on their total wins:
7,562 wins, 6,914 losses, 412 pushes = 52.24% against the spread (-316 on average)

27 NFL handicappers who have data on at least 300 games:
14,341 wins, 13,143 losses, 804 pushes = 52.18% against the spread

The best handicappers in the world are averaging 52% success against the spread in NFL football. It does not matter if you look at all 37 handicappers, the top 10 handicappers, or the 27 with over 300 games, they all get very close to the same results of 52% and it is below what it would take to even break even at 52.38% if the bets were all -110. If the bets, however, are -105 (in other words, you pay $110 or $105 dollars to win $100) the percentage needed to break even is a little more than 51%, but since the format for many is -110, I am using that for the sake of example.

FootballShrink.com Ready to Launch in 2018

It is now the perfect time to evaluate how my picks compare with the top 37 NFL handicappers in the world. While I am quietly confident that I have built something that will be as good or even much better than what is available with the best handicappers in the world, time must now speak. It is one thing to show strong correlations with winning past games, but quite another to predict the future! Picks will be posted each week of the 17-week NFL season at FootballShrink.com later in each week and before each game, using an established line that is published. Since there are precise requirements for what is considered a good pick, there might be weeks where 5 or 6 picks are made, and other weeks where only 1 pick is made.

Meaning of the Predictions

Creating a tool and then using it to predict games against the spread and comparing results to the best handicappers in the world should prove challenging if nothing else. If over time the success rate is on par with the best handicapper experts in the world, that would be good. If success over time is above 52% it would provide definitive proof that the mental game not only matters, it so vital to performance that knowing something about it in a serious way can actually increase prediction! If 30 to 60 picks are made in the 2018 season, this is still an extremely small sample size. Success in year 1 cannot be used as evidence that the MPI is a viable prediction tool. At the same time, lack of success will not mean that it does not work. Hundreds of games predicted in advance and archived on FootballShrink.com, however, will tell the story over time. Statistics requires a large sample size and there is no way to rush this.