Last week, the Giants beat the Saints 16-13. This was a disappointing loss for the Saints and also a disappointing game for anyone rostering fantasy players from this game. The closing total for this game was 54 points and everyone was expecting a shootout much like the game these same two teams had in 2015 in which 101 total points were scored. The question that arises is how often does this happen? We’ll use our archive to look at implied team totals and see just how much variance there is from game to game.
We’ll use the complete 2015-16 season as our dataset. We will look at the implied team total using Pinnacle’s closing lines against what the teams actually scored. Note that we are not looking at Pinnacle’s team totals they have listed to bet. We are looking at the implied total using the spread and total.
When we look at the average difference of the implied total and actual total for all games in 2015-16, we get .2 points. This looks pretty good, right? On average the actual total is .2 points higher than the implied total. Not so fast. That’s because when you have a team go over it’s implied total by 7 points and a team go under it’s implied total by 7 points, you get an average of zero even though we know that’s not correct. We need to take the absolute value and get that average.
Once we use absolute value, we get an average of 7.3 points per game. This means on average, there is more than a touchdown difference between the expected points scored for each team and what actually happens during the game. For those that are curious, when you look at the home/away breakdown, home teams’ average was 7.4 points while the visiting teams’ average was 7.2 points.
Here is the graph from the data with the x-axis displaying how well a team has done compared to their implied points and the y-axis being the number of occurrences:
And here is a table that breaks it down by percentage of games.
|-14 or worse||5.4%|
|-10 or worse||13.6%|
|-7 or worse||23.6%|
|-3 or worse||37.9%|
|Between -3 and +3||26%|
|+3 or better||36.1%|
|+7 or better||23.0%|
|+10 or better||15.5%|
|+14 or better||8.4%|
So roughly one fourth of the time, the team will be within +/- 3 points of their implied point total. The Giants and Saints in the example I started with both ended 12.5 points below their implied point total. 13.6% of teams end with a point total of -10 or worse so it’s not as uncommon as you might have thought.
53.4% of the time, a team will finish within +/- 7 points of their implied total. The other 46.6% of the time they will fall outside of that range. Which means it’s pretty close to a 50/50 proposition that a team will be within a 14-point gap for a given game. I don’t know about you, but a 14-point gap seems pretty large to me. So the next time you are using a team’s implied total to figure out who to start on your fantasy team, you may want to think twice. Because the reason we love watching football in the first place is because it is highly unpredictable.