Since I had previously performed an analysis to determine whether there was a significant difference in the frequency with which home teams and away teams beat the spread, and had also performed an analysis to determine whether there was a significant difference in the frequency with which favorites and underdogs beat the spread, it only seems natural that I would cross-pollinate these two variables in designing a new analysis. In simpler terms, I decided that I wanted to run an analysis to see if there was a difference between the following groups in WATS (Wins Against The Spread) %: Home Favorites, Away Favorites, Home Underdogs, and Away Underdogs. I used WATS%, rather than average number of WATS for each group because there were large differences in the amount of times each team found themself in the above scenarios (Detroit only managed to be an Away Favorite 18 times between 2009 and 2013, whereas Miami accomplished this feat 161 times in the same time span).
In case my readership is too lazy to follow the links to the above results, I will summarize them here for you: I found that from 2009-2013, away teams beat the spread significantly more often than home teams, with respective means of 106.03 WATS and 101.27 WATS. I also found that there was no difference in the WATS% between favorites and underdogs between 2009 and 2013, with percentages of 49.35% and 50.42%. Based on these findings, and paired with another analysis in which I discovered that straight up wins were positively correlated with WATS, going into this analysis I predicted that I might find Away Favorites to have a higher WATS% than the other groups, and Home Underdogs to have a lower WATS% than the other groups.
Results: After looking at the data, I uncovered the following winning percentages against the spread: Home Favorites – 49.28%, Home Underdogs – 46.48%, Away Favorites – 51.82%, Away Underdogs – 50.72%.
Analysis: By segmenting teams into the four conditions above, I had hoped to find that one group would have a winning percentage against the spread that was greater than 52.4%, and would thus provide us with a profitable long-term betting strategy. Although my predictions about the results ended up turning out to be correct, unfortunately none of the groups were able to surpass the profitability barrier. However, it is possible that in the future I will be able to take the Away Favorites group and distill it down even further, in order to find a condition that does beat the spread > 52.4% of the time.