I learned a simple way to calculate a baseline for NFL totals during my first bookmaking job almost a decade ago. I still use it today to spot numbers on the board at open which may be too low or too high. I thought I would explain today in a thread below to offer some value. 👇
The most important thing I learned to think about was how teams are affected by their opponent. Rather than thinking “what can these teams do together” to bet a total, I think, “does this opponent increase or decrease the likelihood of success in the matchup”. In order to do...
this, I like to look back at how a team has made their opponents better or worse this season. I set up my paper with the defences in question and list out opponents and their yards per play/rush attempt for the entire season against all opponents. The second temp is taking the... 
median of those values per pass and rush to create a basic strength of pass and strength of rush opponents faced. I then look up the yards per pass and rush allowed for the season for each defence in question. The next step is to compare the allowance vs the median of opponents..
For example, the median of OAK opponents yards per pass is 7.05. OAK has allowed 8.60 yards per pass this season. 8.60 / 7.05 = 1.22 or 122%. The OAK pass defence has allowed opponents to gain 122% of their expectancy this season. These “factors” can be used later on. The next...
step is to evaluate how the two teams playing make each other better or worse by multiplying the pass and rush factor by the yards per pass/rush gained. For example, against OAK, HOU is expected to increase 122% from 7.80 ypp to 9.51. OAK expectancy is just shy of season average.
Now that I know what both teams are expected to gain against each other both per pass and per rush, I can begin to create a game scenario. For the sake of this thread I will do a neutral game state (more on this later). I need to know plays per game and pass play percentage.
I then multiply the average play count per game per team by the yards per play expectancy for both teams to get a yardage total in both passing and rushing yards. The final step is to take the sum of all yardage expectancy to get a total yards expected for both teams.
In this game the yardage total is expected to be 847. I then use the yards per point median of the league (15.50) to generate an expected points value. In this case 847 / 15.50 = 54.6 points. This game opened at 48 and has since moved to 52.
Some folks reading this may take this and run thinking it is a way to “beat” the NFL. That is absolutely not the case, nor do I encourage that. Instead, what I hope you take away from this is that game state matters when handicapping the NFL. I believe that the best way to...
handicap NFL totals is to look at negative (behind by 8+), neutral (within one score) and positive (leading by 8+) play calling tendencies. The variance of run/pass percentage depending on the scoreboard and situation within teams is ENORMOUS. I think with every total a bettor...
should think about handicapping the game six different times with each team playing the majority in a specific game state. Why this total is interesting is because HOU despite likely having a lead changes their play calling balance less than any team in the NFL (3% diff)...
In comparison, if trailing OAK jumps from 53% to 63% pass. Within the context of this game, not only does the neutral game state suggest the total is short, but the second most likely game state (OAK trailing) projects even a higher total than 54. I think that often as bettors...
we fall into a trap where we think that any advantage suggests an unlimited range to the over or under. If we think both teams have success passing against weak defences, all we think about is “more points!” and not, “how many more points?” That gets problematic because it...
devalues the importance of different odds, we just take it and hope the offences blow the doors off the defence. I hope this helps some folks put a number on their bet and think about how leads or deficits often provide an advantage or disadvantage when betting NFL totals.
In addition to this thread, I have a write-up about the nuts and bolts of the game in my weekly column for @TeamMatchbook if you would like to read it: insights.matchbook.com/adam-chernoff-…
