Thursday, November 14, 2013

Debut of NUera Yields Familiar Result

Nearly all of the focus on Northwestern basketball heading into this 2013-2014 has been on rookie head coach Chris Collins and understandably so. The 2012-2013 roster was more than decimated by injuries (Sanjay Lumpkin, Drew Crawford, Chier Ajou, Aaron Liberman, Jared Swopshire, Reggie Hearn, and Alex Marcotullio all missed parts of the season) and an unfortunate suspension (JerShon Cobb). Last year’s team limped to the finish line, losing its final nine games and finishing with a 13-19 record (4-14 in the B1G). Athletic Director Dr. Jim Phillips finally had a clear opening to remove Bill Carmody and install someone more to his liking. Enter Chris Collins.

Collins comes to NU with a good deal of hype. He is regarded as a strong, enthusiastic recruiter, and so far his 2014 class bears that out; it is shaping up to be one of the best in NU history. He also brings with him a shift in playing style. Gone (supposedly) is the Princeton Offense with its intricate team play based on reads; entering is a more pro-style offense dependent on ball screens, isolation plays, dribble drives, and kick outs. Gone is the prevalence of zone defense, particularly the pesky 1-3-1; entering is an aggressive, overplaying, switching man-to-man. Collins has stated that he wants to push the tempo, give his players freedom on offense, and establish a tougher mentality. Many expect this to limit opponents’ scoring and improve the Wildcats’ rebounding.

A primary focus of this blog will be utilizing tempo-free statistical analysis to understand how the different style of Collins shapes the results on the court. There is good work out there analyzing how changes in style resulting from coaching changes generally take multiple seasons to coalesce. I expect the same at NU. With only one official game completed, there are a couple of things that I think we can say for sure about this year’s team: they will focus on man-to-man and they will try to play a different style of offense. Both of those things were on display last Saturday night against Eastern Illinois, although the team did spend about half the game running Princeton offense sets as well.

But how different were the results? For comparison’s sake, I want to look at this year’s EIU game against last year’s season opener against Texas Southern. Both games ended in decisive victories for NU. NU opened last season with a KenPom ranking of 89 and this season with a ranking of 86. Texas Southern was 253 last season, while Eastern Illinois was 255 this season. NU opened both games with an initial win probability right around 89.5%. I’d say those are fairly even comparisons. So let's jump into the numbers.

Four Factors

Let's look at the Four Factors for these two games. I've labeled them anonymously Game A and Game B.

eFG% TO% OR%
NU Opp NU Opp NU Opp
Game A 58.1% 32.1% 18.8% 10.9% 21.4% 27.1%
Game B 59.5% 33.6% 15.9% 20.3% 21.2% 23.9%

If I stopped here, it might be difficult to distinguish between these two games. Once I show the FTRate, though, I think you'll be able to guess which game corresponds to 2012 and which to 2013.

NU Opp
Game A 67.44 71.70
Game B 24.14 31.03

WOW! Look at those FTRates for Game A. That's insane! The NCAA D1 average FTRate for 2012-2013 was 35.9. I suppose if you've paid any attention to college basketball so far this season you don't need me to tell you that Game A is NU's 2013 opener against EIU. (I've got more to say about the foul calls, perhaps in a later post.)

Setting aside the huge jump in FTRate from last season's open to this season's, a glance the remaining three of the Four Factors suggests that NU was actually better in the 2012-2013 opener.

The eFG% numbers both for NU and its opponents are nearly identical, as is NU's OReb% for each game. It turns out, though, that NU did a better job of protecting its defensive glass in the 2012-2013 opener than it did on Saturday by just over three percentage points.

What really stands out to me from last season's opener compared to this season's is how NU took better care of the ball and how much more frequently NU forced turnovers. NU's TO% was three percentage points better last season, while its opponent's was TWICE as high last season.

The Four Factors for the two games suggest to me that, while on Saturday NU was as effective on offense as it was in last season's opener, NU was actually better defensively last year. This despite the much ballyhooed focus on tough defense and rebounding this fall.

Additional Components

Another point of emphasis with the new coaching staff is a desire to play a more up-tempo style of ball. NU's game against EIU featured 64 possessions. Last year's game against Texas Southern? 69 possessions. In other words, the team that played a supposedly deliberate and slow brand of basketball played a faster pace than this year's up-tempo model. I'll be interested to see how this plays out throughout the season.

Scoring Efficiency
Pts/Pos Pt/FGA eFG% TS%
2013 NU 1.12 1.67 0.58 0.63
EIU 0.86 1.04 0.32 0.39
2012 NU 1.14 1.36 0.59 0.61
Tx So 0.71 0.84 0.34 0.37

The scoring efficiency numbers show NU this year in similar opening game form to last year. NU's points per possession, eFG%, and TS% numbers from both games are nearly identical. Texas Southern and EIU also put up nearly identical eFG% and TS% numbers. You'll notice, however, that EIU actually put up more points per possession (and per FGA) than Texas Southern did last year. The higher Pts/FGA numbers this year are attributable almost exclusively to the very high number of fouls in the NU-EIU game. Still, EIU's more efficient scoring again suggests that last year's team played better opening game defense.

Miscellaneous Components
3P% 2P% FT% Block% Steal%
2013 NU 0.48 0.46 0.76 0.11 0.02
EIU 0.11 0.40 0.55 0.18 0.11
2012 NU 0.44 0.55 0.71 0.22 0.09
Tx So 0.19 0.38 0.56 0.09 0.07

A few quick thoughts on these miscellaneous components. Both EIU and Texas Southern are terrible shooting teams. Good defenses can affect 2P%, but very few defenses have a discernible affect on 3P%, so those 19% and 11% 3P%s for Texas Southern and EIU tell you more about those teams' bad shooting than NU's defense. In the 2012 opener, NU was better at blocking shots and generating steals on defense and at avoiding both of them on offense. In the 2013 opener NU shot lights out from 3P (slightly better than 2012), pretty well from 2P (though not nearly as well as in 2012), and great from the foul line.

Style Components
2013 NU 0.49 0.80
EIU 0.34 0.25
2012 NU 0.43 0.83
Tx So 0.45 0.59

Two hallmarks of Carmody's NU teams were lots of 3PAs and a very high assist-to-made basket ratio. One game into the Collins era not much has changed there.

Point Distribution
3Pts 2Pts FTs
2013 NU 0.42 0.28 0.31
EIU 0.11 0.51 0.38
2012 NU 0.42 0.46 0.13
Tx So 0.31 0.49 0.20

I think the point distributions illustrate two things. First, there were a lot more fouls called this year and consequently more points scored via FT. Second, EIU is worse at 3PT shooting than Texas Southern was.


So what's the point of all this? I'm very interested in separating reality from perception, truth from hype. The perception and hype around Collins is that he represents a sea change in the direction of Northwestern basketball. I think there's some truth to that, off the court. Reading some recaps from his first game as head coach, you might think he's already made a dramatic change on the court, too. This simply isn't true. It may be true sometime in the future, but right now this team looks an awful lot like it did after one game last season. Both games featured good defense and stellar shooting from NU, marked by a lot of assists and 3PT shots, resulting in blow out wins over vastly inferior competition. The truth is that the on-court changes NU fans expect will take time. So enjoy the win, but don't make it out to be something it's not. We've all seen this before.

Looking Ahead

The team has a showdown with Stanford in Palo Alto tonight @ 10 p.m. central. Stanford is coming off a 112-103 home loss to BYU. That game featured 89 possessions, 52 personal fouls, and 77 free throw attempts. Stanford scored 1.16 PPP and gave up 1.26 PPP. This one could be a barn burner, and it should be a decent early-season measuring stick for how good this team could be. For what it's worth, KenPom projects a 75-68 Stanford win in a 67-possession game.

I may or may not have a game recap, but I do anticipate a new post soon focusing on individual players on the NU roster.


  1. Great stuff!

    Excited to see some advanced statistical analysis on 'Cat hoops here in the blogosphere.

    The FTA/FGA is crazy so far this year, and I'm wondering if it's going to hold up. (i.e. will the Refs ease up and/or will defenses play more to the written rule/) I'd expect us to be better with that in the Collins era vs. Carmody era since the former depends more on slashing to the basket and iso, vs. the latter's system which seemed to favor more open shot takes and less attacking the rim. Or perhaps we'll get more Olah usage in this new set, which could be a source for free throws.

    I'm also interested in tracking the OREB% and DREB%. Under Collins the latter is priority one, and I'm hoping that the focus on man-to-man will increase those numbers.

  2. Thanks, MacArthur! I love the new foul calls & enforcement. I hope the refs keep calling it tight. I actually think the new rules would have helped in the old regime just as much, since our cutters were always getting bumped & chucked. I think Sobo will really benefit. He's very savvy at drawing contact even without attempting a shot. Dude had 8 FTAs on 1 FGA vs. EIU. That's an insane FTRate. Hit 7 of those FTAs, too. That alone makes him so much more valuable than Demps. But that's a post for another day.