Third Quarter Warriors: Fact or Fiction?
If you have been following the playoffs, you have heard quite a lot about the 'Third Quarter Warriors,' a menacing version of an already great team that appears in the third quarter. I wanted to see if this notion could be supported by data. To do this, I scrapped data on every NBA game this season (except Game 4 of the finals which I am watching right now). Then for each game, I calculated the point differentials of both teams before and after all four quarters.
The first thing I noticed from this data is that most teams play a lot better at home than away, which is to be expected. Secondly, the Warriors do play noticeably better in the third quarter, especially when they are at home. Their point differential after the second quarter is 2.8 on average while their point differential after the third quarter is 8.16 on average, which means that the Warriors outscore their opponents by more than 5 points in the third quarter, on average. While this might sound like a tamer figure than what you would expect from hearing overzealous commentators, this is the largest increase across a quarter for all teams.
After I finished my calculations, the challenge was visualizing the results in a meaningful way. This is where slope graphs come into play. Slope graphs are useful for visualizing how a quantity differs at two different points in time. I first learned about them from this excellent post by Cole Knaflic. This is exactly what I need since I am trying to compare points differentials across quarters.
In the following interactive slope graph, you can pick your favorite NBA team and see how their point differential compares before and after each quarter, both home and away: