Monday, February 14, 2011

How Do You Win By 50? One Basket at a Time

Matt already gave his thoughts on Saturday's big win over Alma. I don't mean to rehash those but I thought such a wild game deserved a little more commentary, so here goes.

I was struck by how consistent Calvin was over the course of the game. They didn't seem to be prone to the feast or famine offensive trends we've seen so much this year. The Knights got to 115 points, the most they have scored since putting up 115 in a sectional final game against McMurry in 2000, by steadily pouring in baskets from beginning to end.

Most blowouts happen when one team makes a run or two while the rest of the game is played relatively evenly. That is how a team puts together a 20 point blowout victory. 50 point victories take blowouts to a new level. To test my impression of Calvin's consistency, I plotted their lead over the course of the game. See the graph below (click for larger view):



There are certainly stretches of the game that look more impressive than others. On the positive, Calvin did a quick job of building a double-digit lead. On the downside (it just doesn't feel right to use that word when talking about a 50 point win!), Alma was able to keep their deficit around the 25 point mark for the last few minutes of the first half and the first part of the second half.

The solid black line is a linear trendline set with an intercept at zero (since a game starts with a zero point differential, duh). The trendline produces an r-squared value of 0.9454. A data set which is perfectly described by its trendline produces an r-squared value of 1. In other words, nearly 95% of our data set (Calvin's lead over time) is explained by the fitted linear trend line. Calvin very clearly set a pace and stuck to it for 40 minutes.