March 14, 2011 - by David Hess
Duke’s Kyrie Irving hinted yesterday that there was a possibility he’d return to the team for part of the NCAA tournament, though head coach Mike Krzyzewski downplayed the possibility. Just to be safe, we thought we’d examine how Irving’s recovery might impact the Blue Devils, by looking at how they fared with him, as opposed to without.
To do that, we took the Predictive power ratings of Duke and their opponents, and looked at the expected outcome of each game on the schedule. If they outperformed expectations early in the year, with Irving playing, it might be reasonable to credit some of that higher level of play to him.
Here is Duke’s game by game performance, compared to expectations. The value plotted is simply [actual margin of victory] – [margin of victory predicted by power ratings]. So, if the games all went exactly according to projections, you’d see a horizontal line at 0. The thin blue line shows the game-by-game difference, while the thicker line is a 5-game rolling average. The red vertical axis splits the season into pre- and post-Irving.
As you can see, Duke was playing better at the beginning of the year, with Irving. They maintained that for a few games after he went down, but have since fallen back to the pack. Their recent performance has been right in line with their Predictive power rating. So, how much better was that early season performance?
Duke’s Performance Vs. Power Ratings Predictions
|Avg Expected Margin
|Avg Actual Margin
With Irving, Duke played about 7 points better than they did without him. We sincerely doubt that all of that difference was due to Irving’s presence, but it’s likely that some of it was. Exactly how much is beyond the scope of this short blog post. However, it’s totally within the scope of our Ask the Nerds Forum. So if you have something to add, head over there and join in the discussion.
Printed from TeamRankings.com - © 2005-2024 Team Rankings, LLC. All Rights Reserved.