NFL Bets Analysis

See the previous posts on my NFL betting strategy here and here.

This year, I tested the waters of NFL season over/under bets. I made this statement before the season began.

Again, the theme was bet on unders, and most likely there will be more teams that finish with less than 5 wins than Vegas predicts.

My three bets went 2-0-1 with Tampa Bay and Detroit both hitting the under and Miami resulting in a push. Also my prediction that there will be more teams with less than 5 wins than Vegas predicts was spot on. Vegas predicted there would be one, there were actually 5.

Check out this sweet chart I made. Then check out the badass graph I made.



So I did the full analysis to see if my system of betting would have proved profitable this year. With my system I would have bet on 18 unders and 14 overs. I would have ended up going 18-11-3. So in theory if I would have place $100 on each bet I would be up $700 at the end of the season. But the way sports betting works is that payouts aren't always 1:1. Usually you have to bet $110 to make $100. So if I bet $100 on each team and went 16-16 I wouldn't necessarily break even. I would actually be down like $160. Some bets pay out better than 2:1 though. This is because bookies will change the payouts to get even betting on both sides. This season even though I would have won 7 more than I lost I would have only made $535.

But one of the issues with betting on every team is that my systems sometimes finds the Vegas odds to be nearly identical to my odds. Then putting a bet down is like a coin flip. But it can be worse if the payouts are less than 1:1. So if I paired down my system to only the biggest differences between my predictions and Vegas odds and bet those I would have gone 10-6 and won $335.

While I've only looked at one season I am confidant that my systems works pretty well. But the real question is if my system is a better predictor of wins then the sportsbooks are. As a side note, my system relies on the Vegas odds to make my predictions. So I guess my systems relies on Vegas to be at least somewhat close to the actual odds, and from what I can tell they are.

So, I looked at the total difference between the actual results and both Vegas and my system. Vegas was off by 61 wins, while I was off by 58.33 wins. My variance was also 1.61 wins compared to 1.65 for Vegas. But I'm not sure that this comparison is valid because Vegas odds are in increments of .5 and my predictions are decimals.

I like the idea of betting the season over/under because the bet can minimize the randomness of football outcomes by aggregating a teams performance over 16 games. Sure they might get upset a couple of times, but they might also pull a few of their own. At the end of the day usually they are who we thought they were. So will I take this action next season? You betcha.