Started looking through some numbers this morning to check on a half-baked theory about how teams are attacking the Kansas defense. The thought behind this was that teams who scout and game plan for the Jayhawks would alter their approach a little bit when Jeff Withey is on the bench compared to in the game. This would mean more feeding the post and more attacking the rim. It's not a huge secret that Kevin Young, Perry Ellis, and Jamari Traylor are all pretty raw on the defensive end. Compared to attacking Withey, teams can have success against those guys.
In order to check the theory, I went through the play-by-play logs of every Big 12 conference game and tracked the 2pt attempts (makes and misses) for when Withey was on the bench and when he was on the court. Looking at the numbers, I can't see much of a trend. This may be because Withey plays a lot and teams just don't get the opportunity to alter what they're doing for the short time he's on the bench or teams simply don't worry whether he's on the court or not. It's also very possible that just looking at 2 pt attempts isn't a good way about checking on the theory.
First up, I tried to figure out the number of 2 pt shots a team would be expected to take based on the amount of time Withey is on the court and the number of 2's a team took in the game. For example, Withey plays 90% of the minutes and a team took 10 two pt shots in a game, the team would be expected to take nine 2s with Withey on the floor.
With the expected shots figured, I compared it to the actual number of shots taken with Withey on the court. This is where I was hoping to see a bit of a difference but came away with nothing really.
So my theory's not holding up so far. The second thing I wanted to look at was a quick glance to see if teams are more successful in their attempts with Withey on the court compared to him on the bench.
Average FG% with Withey on the court is 39% compared to 41% with him off the court. The average may be misleading due to OU going 0-4 from 2 pt land with Withey out of the game, we know zeroes will skew averages. Ignoring that game changes the averages to 38% for Withey on the court and 43% off the court. Might be a little more what we expect.
The number of shots teams take with Withey on the bench probably aren't enough to draw meaning one way or the other. The other issue is that all 2 pt attempts are included, so easy layups following turnovers are counted and Withey has very little impact in that area of the game. Free throw attempts and the number of rebounds also weren't counted because the play by play doesn't give the information on where a foul occurred.