Methods of Analysis - Football
By Bob Stoll - Updated July, 2008
I have been a successful professional handicapper for 21 years and my methods have evolved from my early years as mostly a technical handicapper. While the research I have done over the years indicates that technical analysis does work, I felt as if there were some games in which I was giving up line value on the team that my situational analysis dictated that I bet. In an effort to gauge line value, I developed a model that is mathematically sound and that math model has produced very good results over the 7 years since its inception. With 7 years of good results, my mathematical predictions have become a major part of my handicapping process. My research of fundamental indicators has also proven to be profitable and I will continue to explore the statistical profiles of opposing teams to find value in a particular match-up. Below is an explanation of the methods that I currently employ in my handicapping and a summary of how I combine those tools to accurately measure a team's chance of covering the pointspread at any given line based on the situation, fundamental indicators, my math model prediction and the line.
Situational Analysis is the study of performance patterns, either on a league-wide basis or on a team specific basis.
I tend to shy away from most team oriented trends unless the head coach or core of star players has been intact over the term of the trend. I certainly wouldn't pay much attention to a Carolina Panthers trend that included games prior to the arrival of head coach John Cox – who changed the personality of that team. On the other hand, longer term trends of the Pittsburgh Steelers do have some validity due to the long tenure of head coach Bill Cowher – even though the personnel have changed over the years.
Most of the situational analysis that I employ are league wide trends rather than team specific patterns. For instance, NFL home underdogs have been pretty good bets over the years after an upset win (173-128-9 ATS since 1980) since such teams tend to play with more confidence in that situation. That is a very simple trend with very few parameters and I expect that situation to continue to be profitable in the future. Many handicappers tend to back-fit past data by adding more and more factors (parameters) to a situation until they have a very high percentage angle (but also a much smaller sample size). However, my research has shown that a situation's predictability is sacrificed with each parameter added to derive that situation. For instance, a situation with a record of 40-20 (67%) that is derived using 10 factors isn't as predictive as the 57.4% home underdog situation that I presented above, which has just 4 parameters (this game home, this game dog, won last game, dog last game) and a much larger sample size. It's easy to find a very high-percentage situation if you use an unlimited number of parameters to get to that situation, but all that will result in is a situation that explains what has happened rather than something that helps predict what will happen. My research, and the theories of statistics, shows that the more predictive angles have fewer factors and a larger sample size, rather than a smaller sample situation with a high winning percentage that was derived by using too many parameters. Further research I did in the Summer of 2004 enabled me to accurately assess a situation's future performance based on the win percentage, sample size, number of parameters and more recent performance (i.e. record of the angle over the past 3 seasons). That research led to a more realistic use of situational analysis than I've employed in the past. For instance, I can now tell you that a situation with a record of 140-60-5 ATS that uses 6 parameters has a 56.8% chance of winning the next time it applies if the line is fair. Having a realistic expectation of a situation's value has helped my overall analysis immensely the last 2 seasons and I will continue to devote time each summer to update the research on the predictability of my situational analysis.
Fundamental indicators are based on the statistical profiles of the teams involved in an upcoming game, using historical spread results of match-ups that involved teams with similar profiles. For instance, I have learned that good running teams tend to perform better at home while good passing teams have a tendency to perform relatively better on the road. You might find it surprising to find out that teams that have turned the ball over a lot are actually pretty good bets – which is due to the fact that their previous turnover problems have likely led to disappointing past performances that will be already reflected in the current pointspread, while the likelihood of turning the ball over at the same high rate is not likely. That scenario creates line value in favor of the turnover prone team. I also use fundamental indicators that are based a team's statistical profile (i.e. good rushing team, average passing team, good run defense, etc) when matched against the profile of their opponent. For instance, I can query how teams with a good run offense have performed, against the pointspread, when facing teams with a poor run defense. I have found those sort of match-up indicators to be very insightful and my recent research shows that indicators based on season-to-date statistics are very predictive and weighing my fundamental indictors more heavily last season has helped produce better overall results.
Most handicappers that try to come up with a formula to predict future games tend to make the same mistake. That mistake is using regression analysis to find the correlation between different statistics and point differential. While that exercise is very useful for explaining which statistics impact a game's result, regression is not necessarily useful in using past statistical averages to predict future results since some important statistics simply don't correlate very highly to the future. For example, turnovers are the number one factor in point differential in football, but turnovers are also the least predictable statistic. A model that is based on regression analysis will weigh turnovers very highly, but since past turnovers do not correlate highly with future turnovers such models will over-weigh the affect of past turnovers – creating a model that is good at explaining what has happened but not very good at predicting what will happen.
My math model incorporates the predictability of past statistics to future games and uses each team's compensated statistics rather than their raw stats, which adds to the accuracy of my prediction. Compensated statistics are derived by comparing a team's statistics to the statistics of the opponents that they have faced. For instance, a team averaging only 3.6 yards per rush on offense is actually a better than average running team if they have faced a schedule of opponents that combine to allow an average of just 3.4 ypr on defense. Using compensated statistics in combination with the predictive nature of each statistic used in my model produces an accurate measure of the true differences between two teams future performances – not the difference between their past performances. I also adjust my projected numbers based on current personnel for each team and those extra hours of statistical work have paid off handsomely over the years (and I get better each year at making those adjustments). I also take out meaningless plays such as kneel downs at the end of a half or game and quarterback spikes, so the game statistics that I use are more representative of a team's performance than the statistics used by other handicappers.
I've been using my current math NFL model for 8 years and the record is a very good 199-140-8 ATS (59%) when my math model prediction is 5 points or more away from the actual pointspread and my College math model had produced 57% winners from 2001 to 2004 using differences of 7 points or higher. I made improvements to my College Math Model prior to the 2005 season and the results have been even better than I anticipated. My College math model kicks in week 5 and it has been 55% picking every single College game from week 5 on since 2005 and an incredible 60% ATS in games where the difference between my prediction and the line is 6 points or more (5.8 or more, actually) as long as both teams had played 3 or more games.
The key to the 2004 research on my methods was finding a way to combine my situational analysis, fundamental indicators and my math model to give me an overall chance of a team covering at any given number. My performance on my Best Bets the last 4 seasons is an indication that I succeeded in that endeavor and I will continue to refine the accuracy of my methods each year. An example of combined analysis is a game in which Team A applies to a 140-60-5 ATS situation that uses 6 parameters. Team B applies to a statistical profile indicator with a record of 86-28-4 ATS and my NFL math model favors Team A by 12.4 points when Team A is a 7 point favorite in reality. As discussed above, a situation with a record of 140-60-5 and 6 parameters has a 56.8% chance of winning if the line is fair. The fundamental indicator favoring Team B has a 58.2% chance of winning given a fair line and my math model would give Team A a 56.9% chance of covering at a line of -7 points. The trick is assigning a point value to the situation and the fundamental indicator based on their chance of covering at a fair line. I simply put everything in terms of points based on the relationship between point differentials and the chance of covering of my math model. Each point difference in my math model is worth about 1.3% in chance of covering, so each percentage point is worth about 0.8 points (1/1.3). In this case, the situation favoring Team A is worth 5.3 points while the fundamental indicator favoring Team B is worth 6.4 points. My math model favors Team A by 10.4 points, so adding the value of the situation and the indicator would result in an overall prediction of Team A by 9.3 points (+5.3 – 6.4 + 12.4 = 11.3), which would give Team A a 55.5% chance of covering at the line of -7 points. Obviously, things can become a lot more complicated when there are multiple situations and indicators applying to a particular game - which is most often the case, but my years of studying probability theory at Berkeley have given me the tools to sort through it all and come up with an accurate measure of the overall affect of the situations and indicators.
A lot of handicappers use situational analysis and math models in their handicapping but few, if any, of them have studied the predictability of their methods, as I have, or found a realistic way of combining their methods for an overall measure of predicted success on every game.
Sports Betting as an Investment
My analysis has also shown me that there is no game that has higher than a 70% chance of covering - so ignore those handicappers that advertise "Locks" and "Guaranteed Winners". You'll also notice plenty of handicappers that claim to win well over 60% of their games. I have had plenty of seasons in which I've won over 60% of my Best Bets, but there is no handicapper that is better than 60% over the long run. I feel that I'm on my way to that sort of long term success in football given the research I've done the past 3 summers on my handicapping methods. While 60% may not seem that impressive to those of you that are new to sports betting, you should consider that a bet with a 60% chance of covering is a 16% investment at -110 odds (.60 - .40 - .040 = .016), and that investment is compounded since you can use your winnings immediately for future sports investments. I consider any season in which I win 56% or more to be a successful season, as 56% is a solid 7.6% investment per bet while winning 56% of 150 Best Bets would lead to a record of 84-66, which is a profit of 11.4 units. If you have a bankroll of $10,000 and expect to win 56% on 150 bets then you can safely wager an average of $400 per game (or $160 per Star using my Star ratings) with less than 1% chance of losing your bankroll. Winning 56% of your 150 bets at $400 a game would result in a profit of $4560, which is actually a 45.6% investment on your $10,000. So, you can see why I consider 56% to be a successful season and a very good investment. For more information on money management you should consult my Sports Betting as an Investment section on this site.