The question as to why you should be betting on totals could probably be answered in one sentence: because you will often find lower staking limits and liquidity on totals. That may sound counterintuitive, but limits on totals are lower because bookmakers and line setters are less sure of their numbers, and as bettors, we like uncertainty.
Uncertainty always gives us a chance to beat the price before the wider market figures out what it should be.
So why are bookmakers and market makers less confident in their totals?
Part of it is simply that less money is bet on totals, so the number isn’t as solid. Another factor is that they are more difficult to model, and of course, models drive the modern market.
A lot of NFL models work by coming up with a power rating for teams, and working out how many points better or worse they are than an average team. That’s harder to do for totals, and there are a myriad of extra factors which you have to take into account, including weather, pace of play, paying surface and referee.
The impact of the weather
The weather is one of the most underrated factors in the NFL betting markets, and the wind, in particular, can make a huge difference to the way a game plays out. You often find the market adjusts a few points, but it is loathe to go too far, and this can be a mistake. Research has shown that when the wind speed gets above a certain level it can have a dramatic impact on the success of unders bets on the total.
But it’s not just about the wind when it comes to weather, and there are other ways the market doesn’t adjust enough to a change in conditions.
One weather misconception is that scoring plummets in the cold. Well, it does but not to the level the market thinks, as a rock hard ball makes turnovers and short fields more likely, while tackling is also just a little bit more painful.
Per SportsInsights, overs are 104-73-4 (58.8%) in low total games when it’s cold outside.
In fact over the last 240-odd NFL games under 40 degrees, the over has won at a 59.8% rate. If you only take games below freezing, the winning percentage jumps to 70.5% and the ROI skyrockets to 36.9%.
The totals in divisional games
Another mechanic you can look at for value on the under is divisional games. Per Bet Labs Sports, when taking the under in any divisional game, you’re going to be winning 52.7% of your bets since 2003. That’s slightly better than breakeven as is, but if you narrow it down to the second half of the season when the teams are playing each other for the second time and familiarity reigns, the under is an even more profitable bet.
Take out lower totalled games as well, and you have a productive system. Again as per Bet Labs Sports there is what they call the contrarian system.
The last factor of the total needing to be over 43 does start to look like data-mining in order to find a winning system, but I actually think it speaks to larger point that NFL games seem to find themselves drawn to an average of around 45.
In other words, if the total is set at 40 because we have two exceptional defensive teams, I will generally look to make the case for the over, and conversely, if the total is set at 54, I’ll be looking to make the case for the under.
Again it’s worth noting that all these rules are not hard and fast. They are broad trends which can point you in a direction, but you need to start with a true number based on the team metrics. Yards per play, success rate, Football Outsiders DVOA, and PFF’s defensive rankings can all be useful to try and establish the true talent of offenses and defenses before you start layering in those other factors.
Snap to it!
One area where there is probably an edge to be had is snap rate and no huddle percentage.
It’s worth having a Google to see where you can find the stats. But you can get them off the Snaps column published on Pro Football Focus each week, which you need some form of subscription for. It’s around £30 for the year and well worth it if it helps you win even one bet.
The column throws up some really interesting tendencies and last year it identified that Eli Manning was going no-huddle a high percentage of the time in an effort to simplify opposing defensive schemes and help his wet tissue paper of an offensive line.
It also identified that the Lions were milking the play clock on almost every road snap to help keep their wet tissue paper defense off the field.
I think they had seven straight unders at one point despite having a defense that was 30th in efficiency simply because the offense was taking so much time off the clock. But the point really is that looking more deeply at the snap stats can throw up some illuminating numbers, and you should always be looking for these small edges when trying to find a winning bet.
Looking at the schedules
Another area that can be exploited is strength of schedule and this is one of Warren Sharp’s favourite angles. In essence, if someone has played three straight “under” teams resulting in low-scoring games, the market often over-adjusts, and there’s money to be made coming back to the over in the next game.
Finally, an angle that can be exploited early in the season is very simple: just bet unders.
This has historically been very profitable as betting unders in all Week 1 games over the 2000-2010 NFL seasons, yielding a 13.6% ROI. The theory is pretty simple, in that it takes longer for the offense to get their timing and rhythm down, so defenses dominate early in the season.
Indeed this trend has got even stronger in recent years under the new CBA rules which give players less practice time before the season and offenses are even more disjointed.