UDM Improvement ’13: Faceoffs

Faceoffs were often a struggle for UDM in 2011 (and 2012)

This post might feel familiar: I made a similar one last summer. To sum up that article, I figured the Titans would perform a bit better on faceoffs in 2012 than they did in 2011, though that was more based on “feel” than the stats contained in the post.

While the Titans did improve – they won .428 of faceoffs compared to .409 the previous season – it was still not enough to even out the possession game. While faceoffs weren’t the biggest problem for the Titans (that would be the offense, which was regularly a trainwreck in the short-lived tenure of coordinator Bill Tully), they were far from a strength.

The Personnel

The Titans went with two primary faceoff specialists in 2012, after mostly riding Brandon Davenport in 2011. Davenport (.446) was slightly better than he had been the previous year, and Tyler Corcoran (.465) was even better than that.

Part of what made the Titans struggle was the use of non-specialists on faceoffs. In 2011, 21 faceoffs were taken by non-specialists (Davenport, Joe MacLean, and Corcoran), and those players won just three of them, good for .143.

This season, non-specialists (MacLean didn’t take any faceoffs despite regular playing time as a defensive midfielder, so only Davenport and Corcoran apply here, though underclassman Tyler Harper, brought in as a faceoff specialist, does) took 36 faceoffs, winning just seven, good for .194.

Obviously, there are strategic reasons to voluntarily concede a faceoff to prevent fastbreak situations, so even though those hurt the overall number, they don’t affect the win rate when UDM was actually trying to win a draw.

Going Forward

It’s clear Davenport and Corcoran are your top two guys. Both started the year poorly – that’s bound to happen against excellent faceoff specialists from Delaware, UNC, and others – and rebounded to draw about even on the year. Here are their cumulative win percentages over the course of the year:

Both improved over the course of the year.

Both finished on the uptick, but it appears Davenport leveled out at around .450 and Corcoran’s numbers were still improving. That could be a matter of sample size – more of Davenport’s draws came early in the year, whereas Corcoran had more attempts in later games.

My “gut feel” at the end of the year was that Corcoran was impressing to close out the season, and the numbers definitely appear to bear that out. Another feeling I had? The more attempts either player had, the better he seemed to perform. That’s one to be careful with, since the causal relationship can go in either direction (he’s winning more because he’s getting in a rhythm, or he’s just getting more reps because he started out playing well), but it’s still notable.

Win% by attempts.

As you can see, both performed better with more attempts. The difference – aside from a much more “boom or bust” status with Corcoran – is in the slope of the line. Corcoran’s is about 0.83% improvement per faceoff, with a maximum number of attempts at 15. Davenport’s is around 0.45% per faceoff, with a maximum attempts of 26.

As noted above, there are a lot of confounding factors at play, but it might be worth giving Corcoran the chance to win the job next year.

Conclusion

As noted in last year’s post, there are several factors that go into improvement on faceoffs. Here, I’m looking mostly at how the top two options are used. Both had moments of brilliance and struggle last year, with a much wider spread for Tyler Corcoran.

Better overall team health next year (a variety of SSDMs missed significant time last season, and though there’s no way of knowing which impacted the performance on faceoffs, it probably played a factor) should be a benefit to the performance, and yet again, I think the Titans will edge toward .500.