Using Averages in Sports Handicapping

Using averages in sports handicapping is the most common method of determining values for a particular team. They're relatively simple to find, as many publications and websites already provide this information, but they may not be the best numbers to use, especially when handicapping football teams, and to an extent college basketball teams, as one or two extremely high-scoring or low-scoring games can distort the statistics when you consider the small number of games played.

Instead of using averages, you may want to consider using medians, instead. Medians are simply numbers that separate a higher group of numbers in a sample from the lower half of the numbers in the same sample. The number in the middle, essentially.

In the following group of seven numbers (3, 7, 11, 15, 18, 21, 23), the median is 15. The average of the numbers is 14.

Using the 2007 New York Giants an example, through nine games the team has allowed 190 points or an average of 21.1 points per game. But a closer inspection of New York's past results show that 45 of those points were allowed in one game against the Dallas Cowboys. The Giants have allowed 17 or fewer points in five of their nine games, so it's logical to conclude the Giants are a slightly better defensive team than their scoring average indicates.

Medians in Action

Imagine your son's soccer team is hosting a two-day tournament with seven games played each day. As a parent, you are required to work on one of the two days and elect to work on the second day so that you can spend time with your neighbor Joe, who also has a son on the team. As it is going to be a long day, you and Joe decide to liven things up a bit, by placing friendly wagers on each of the soccer games.

Since you don't want to be betting against little kids, the two of you decide to make wagers on the number of goals scored in each game. Joe tells you to pick a number and he will bet either over the total or under the total. You agree, and pick up the results from the first day of the tournament, which read:

Galaxy 8, Strikers 4
Thunder 5, Cosmos 1
Fire 3, Challengers 1
Tigers 7, Waves 6
Wizards 6, Crew 1
Rapids 6, Dynamo 2
Force 4, Shooters 2

You add all the scores up in your head and arrive at a total of 56 goals or an average of eight goals in each game. You tell Joe the number is eight and he says he will bet the unders.

As you look at the scores once again, you noticed that if all of today's games have the exact same scores, you will win two bets, lose four bets, and tie on one. Had you used medians, you would see that three games had eight or more goals and three games six or fewer goals and one game had seven goals. The median number of seven would have been a number more reflective of the number of expected goals you would see in the tournament.

A more practical use, for sports betting purposes, would be to take the 2007 Minnesota Vikings football team. Through nine games of the season the Vikings have scored 166 points for an average of 18.4 points per game, and allowed 188 points, or an average of 21.1 points per game.

Looking at Minnesota's results gives the following:

Atlanta 24-3
Detroit 17-20
Kansas City 10-13
Green Bay 16-23
Chicago 34-31
Dallas 14-24
Philadelphia 16-23
San Diego 35-17
Green Bay 0-34

Using medians, however, gives the Vikings 17 points on offense and shows the Minnesota defense allowing 23 points, so it's logical to conclude that Minnesota may be rated a bit higher than they should be.

As sports bettors, what we are looking for are games where one team is overrated on both offense and defense playing against a team that is underrated on both offense and defense. When that situation occurs, don't be afraid to back the underrated team, knowing that you will be getting good value with the point spread.