3-Year NFL Success Rates with ProMathletics
Positive expected value and long-run profitable returns are a couple of phrases that are thrown around lightly in both the financial advisory and gaming industries. There are distinct differences in the two industries that both hurt and help market participants achieve their desired results. Financial markets are extremely efficient, which makes it harder to outperform the crowd. However, the market in general grows and trends in a positive direction. The gaming markets are far less efficient, so it’s easier to achieve an edge, but it is also a zero-sum game where a 10% implied fee is charged for every ticket.
There are both quantitative and qualitative factors that impact and differentiate all markets. While the financial markets are often hard for laymen to understand, sports are something near and dear to the heart and soul of most individuals. The economics of long-run returns for sports speculation are negative for the masses because of the charges associated with taking a position. Most people don’t sit down and sketch out the math of each decision they make, but they feel it in a very real way. The chart below displays what happens to most people over the course of a 256-game regular season. This assumes they take a position in half of the games played.
The average novice market player is equally successful and incorrect in their actual choices. With no advanced statistical data modeling techniques or non-public knowledge, the average person does exactly what the mathematical probability would suggest…they get one out of two correct for a 50% success rate. That doesn’t sound bad, but unfortunately, because of the 10% fee charged by the bookmaker, it is a greatly unprofitable proposition. What’s even worse is that due to various psychological factors, most people will alter their units of risk as they are gaining frustration during a cold streak and end up facing ruin faster than if they would have kept equal or reducing units of risk.
Ok, so we have established that participating in the gaming markets is a fool’s errand and a waste of your time and financial resources, right? That conclusion is not true if an individual possesses an ability to perform better than the mathematical base probability of 50%. Perhaps you are a former athlete who has a very deep understanding of a sport and can use your eyes and gut to read trends or assess matchups with a better degree of accuracy than the Average Joe. Or, perhaps you have a firm understanding of statistics and are able to use historical results to model future expectations. Or lastly, you are like ProMathletics and combine those knowledge sets to create a repeatable approach that stands the test of time. If that’s the case, a person may observe a 60% success rate. Let’s take a look at the economics of an NFL season with that assumption…
Well now, that changed the results of season-long exercise quite a bit! The success rate assumption dictates long-run results to such a degree that 50% vs. 60% is almost a $14,000 difference in the average participant’s NFL season. It is also important to note that this assumes a flat unit of risk sizing methodology. The Kelly Criterion is the optimal methodology over a long run, but due to the timing of NFL games, it must be applied with a great deal of finesse. A single risk result outcome that is contained unto itself is the perfect setting to utilize the Kelly Criterion. For example, while playing a hand of blackjack you have a unit of risk, odds of success, and a payout given success. The player will know before the next hand whether the previous hand resulted in success or loss, and thus, they will know their updated bankroll. Football games are not that simple.
During an average NFL Sunday there will be approximately 11 to 12 games. The majority of these will occur simultaneously at 1 o’clock EST. If a participant has 4 or 5 games of interest that all are being played at the same time, it is impossible to apply the Kelly Criterion, per its definition. If the probability of success and payout given success equate to a 22% optimal unit of risk size, how is a participant supposed to simultaneously lay that on 5 games being played at the same time? Additionally, what happens if an 0-4 or 0-5 results at the same time? If you want to know how easy it is for an average person to go through a streak of 4 or 5 consecutive losses, take out a quarter and flip it 100 times. Keep track of how many “hot” or “cold” streaks there are with heads as a “winner” and tails as a “loser”… you will be surprised at how often this sort of short term variance will happen with a small sample size.
Possessing a deep understanding of the Kelly Criterion is fine, but in order to optimize returns without overexposing oneself to risk, a participant must have an extremely deep knowledge regarding the probability of success and failure. Our proprietary databases incorporate over a decade of historically observed outcomes that ensure stability with our year to year performance. Our NFL commentary not only includes the probability of success, but also arms our subscribers with the unit of risk sizes that allow for our long-run positive expected value edge to take hold. You will never see a “game of the week” or “play of the season” with us because the reality is those are gimmicks and marketing tools. Depending on the intrinsic value of our calculated outcome versus the overall market’s expectation, we know with a very high degree of certainty the probability of success for our suggestion. That is why ProMathletics, unlike most marketing sites who make outlandish claims, has been able to sustain positive season-long returns over the duration of our services.
For questions on our services or historical results, please contact us at firstname.lastname@example.org