— When Robbie Hummel went down with an ACL tear in October, the general consensus — traditional media and statistical models agreed — was that the injury killed any hope Purdue had of being an elite team. One of the only voices piping up to say that it wasn’t the end of the world for the Boilermakers was Ken Pomeroy.
As usual, we should have listened to Ken.
Purdue entered Thursday's game at Minnesota ranked No. 4 by Pomeroy, No. 5 by Sagarin, and No. 8 in both polls. Despite a 70-67 loss, the Boilermakers are 15-2 overall and 4-1 in the Big Ten and remain a formidable foe thanks to a variety of reasons. There is reason for concern, but I'll get to that in a minute.
How have they done so well despite the loss of such an important player? Losing a star isn’t always the end of the world (see Kyrie Irving, Duke), but it sure seemed it would be for Purdue because, well, wasn’t it last year?
This chart shows opponent-adjusted offense and defensive efficiency (points per 100 possessions), along with Four Factors numbers, for the stretches before and after last season’s Hummel injury, plus this season to date:
The offense tanked last year, but I’ll get to that. First, check out the defense. Despite Hummel’s defensive strengths (3rd on the team in Block% and Steal%, 1st in defensive rebounding percentage), the Purdue 'D' actually improved after he went down. The improvement has continued this year, and the Boilermakers currently rank as the nation’s 2nd best defense, according to Pomeroy. Maybe we should have seen this coming: giving Matt Painter a mulligan on his first year at Purdue, the rest of his teams have ranked 13th, 16th, 5th, 3rd, and now 2nd in defensive efficiency. The man can coach defense.
Big Ten Geeks noted recently that Purdue's defense this year has been Duke-like, in that they’ve extended their perimeter defense in order to take away the three (only 27.2 percent of opponent shots are 3’s, 30th lowest in the country), forcing teams to drive or pass into the middle, where they have to deal with JaJuan Johnson, who blocks 7.6 percent of opponent shots while he’s on the floor (80th nationally) while only fouling 1.6 times per game. Well, that strategy seemed to materialize last season, right around the time of Hummel’s injury:
It makes sense — when faced with with the loss of a key defensive player or two (Hummel and Chris Kramer), shift more defensive responsibility to the one stalwart you have left (Johnson). Whether it was truly a direct response to Hummel’s absence or not, the strategy does seem to be a factor in their defensive success.
Now, the offense. Purdue’s offensive rating plummeted when Hummel went down, but until then, the Boilermakers were very good. Counting only conference games before his injury, Purdue would have ranked 25th in offense — their best year yet under Painter. This season their offense sits at 22nd, meaning they’re playing even better. Why was Hummel’s loss so devastating last year, but a seeming non-factor now?
Well, Purdue’s offensive woes last season showed up mainly in the stat lines of two players: E’Twaun Moore and Keaton Grant. When Robbie Hummel went down, the bulk of his vacated shots were inherited by Moore, Grant, Johnson, DJ Byrd, Chris Kramer, and Lewis Jackson, who all saw increases in minutes and/or Shot% (the fraction of a team’s shots that the player takes while he’s on the floor — an average player takes 20 percent). Kramer and Jackson went from invisible to nearly-invisible, Byrd went from a scrub to a part-time spot-up shooter, and Johnson expanded on his already large role in the offense, somehow maintaining his efficiency despite taking considerably more shots. The increased role seems to have been a wake-up call for Johnson, as he’s continued — and even improved on — that performance this year:
Unfortunately, Grant and Moore couldn’t quite carry their heavier loads:
Moore’s 33.6 Shot% after the loss of Hummel would have been 3rd highest among major-conference players over the whole year, behind only Luke Harangody and Devan Downey. Being thrust suddenly into such an absurdly huge role was a little unfair, and it’s no surprise that his numbers suffered. If he had kept that up this season, Purdue would be in trouble.
However, with his Shot% back down to its natural level, his shooting percentages have returned to nearly their pre-Hummel levels. Combined with his decreased turnovers and increased rebounding, his overall stat line looks like a huge improvement over last year. But if you remove those crummy end-of-year numbers, it becomes evident that we shouldn’t be terribly surprised by his performance, given that he’s not being forced to carry things as he was at the end of 2010.
Now the questions becomes who has stepped up to A) allow Moore to back off the throttle, and B) replace the production of the departed Keaton Grant? Heading into the season, many hoped the answer would be freshman Terone Johnson. He’s been taking his share of shots, but in truth he hasn’t been helping much, with an offensive rating of only 93.9 (average is 100).
The real key seems to be the drastically improved shooting — and increased minutes — of DJ Byrd, John Hart, and especially Ryne Smith:
This is what was missing last spring. With Hummel out, and this trio throwing up bricks, Moore was left with no real long range threat to spot up when he drove the lane, so help defenders were able to collapse down on him. If Smith, Byrd, and Hart had been knocking down shots at their current rate, they might have drawn defensive attention away from Moore, or offered him a release valve to dish to, and we might have realized that there is life after Hummel for Purdue after all.
Boilermaker fans have permission to be antsy, though. Their status as an elite team is precariously balanced atop the sky-high 3-point FG percentage of a player who was a complete afterthought heading into the season. Need an example? He was just 1 of 4 from the field and finished with three points.
If Smith continues to be a non-factor like he was against Minnesota, help defenders will roam free, and Purdue’s offense could dry up very quickly.