Recently

posted by DavidH on 2/13/2007 - -

It has seemed to me like Kansas has gotten their act together a little since the Texas Tech loss. They’ve stopped letting inferior teams hang around, they’ve pushed the tempo, and their offense seems to have benefited. I wanted to see what the stats said about this, so I played with Pomeroy’s.

His Game Plan pages list raw game-by-game efficiency stats on offense and defense. These tell you only half of the story, though. You can see a team’s output, but you need the context of that output. 1.1 points per possession against Texas A&M is excellent. Against Baylor, eh. To get that context, you need the opponent’s seasonal adjusted efficiency stats. Taking those two pieces of the puzzle, I worked backwards from the formulas and parameters Pomeroy lists on his site to obtain adjusted game-by-game efficiency stats.

So, was my perception correct? Has KU’s offense been on an uptick? Here are the game-by-game adjusted efficiency numbers since the Texas Tech loss. “Equiv Rnk” indicates what their Pomeroy Rating rank would be if they played like that every game… OR what rank team they could be expected to beat on a neutral court if they performed at that level.

Opponent A/H Off Eff Def Eff Pyth Equiv Rnk
Baylor A 113.4 69.0 .9967 1
Colorado H 113.5 94.5 .8916 53
Nebraska A 125.3 75.4 .9971 1
Texas A&M H 120.7 92.7 .9543 17
Kansas St. H 145.4 92.1 .9948 1
Missouri A 132.5 83.7 .9949 1
Average - 125.1 84.6 .9891 2
Full Season - 116.0 82.6 .9803 6

The offense has indeed taken a step forward. A full season Adj Off Eff of 125.1 would rank 2nd in the country (behind Georgetown). The defense has slipped a little - 84.6 would only rank 6th. One caveat, though - the one game where they needed to play like a top-5 team (vs. Texas A&M), they didn’t.

One other caveat - I cherry-picked this data to show only their recent warm streak. How do the Hawks stack up if I do the same for the other top teams? For each team, I used their most recent 5 to 10 games, whatever gave them the best results. I looked at the top 10 and a few other teams I suspected might be able to crack the top, so there’s a chance I missed some team on a ridiculous tear. But I think these are the 10 hottest teams. “.99+” is the number of .99+ games in the past 10.

Team Off Eff Def Eff Pyth .99+
North Carolina 123.5 77.2 .9956 7
Pittsburgh 120.8 80.8 .9903 4
Georgetown 143.2 96.7 .9892 4
Kansas 125.1 84.6 .9891 5
Florida 131.7 89.4 .9886 4
Ohio St. 128.6 88.1 .9873 3
Texas A&M 124.2 86.5 .9847 2
UCLA 120.6 85.3 .9817 2
Memphis 121.6 86.1 .9815 4
Wisconsin 122.4 87.9 .9784 1

Three things jump out at me:

  • Georgetown’s offense is unstoppable.
  • If not for the letdown against NC St, I’d be calling UNC’s defense unbreakable.
  • It’s UNC and everyone else.

I’m sure I’ll mess around with this sort of thing more as we get closer to the tournament. For now it’s nice to see that KU can get as hot as any almost team in the country.

5 Responses to “Recently”

  1. Hoopinion Says:
    February 13th, 2007 at 9:39 am

    Great post, David.

  2. Chalmersfan Says:
    February 13th, 2007 at 10:40 am

    eek- that’s some great research there, but I cringe at your use of adjusted stats. I can’t give you other teams numbers but I’ll give you the raw ones if you subtract Bill Self’s horrible decision to keep going to Kaun for 6 minutes in the A&M game. KU scored only 4 points in 9 possessions and blew an 8 point lead in that time.

    Opponent Result Site Pace Eff.
    Baylor W, 82-56 A 73 112.0
    Colorado W, 97-74 H 82 117.0
    Nebraska W, 76-56 A 66 115.0
    Texas A&M 62-55 H 55 112.7
    Kansas St. W, 97-70 H 73 132.5
    Missouri W, 92-74 A 78 117.7
    Average 84-66.7 72.7 116.1

    (hopefully this will format right)

    Defensive efficiency in that time is 91.7. You know what? These look like Roy Williams type numbers, except I can’t remember a Roy Williams team doing this well without senior leadership and without a dependable post option. I can’t remember a recent Roy Williams team having this type of balance so that it didn’t need one great player scoring 20+ every game.

    Looking at the four factors the only big problems left are forcing turnovers and opponents shooting but the latter may have been a result of getting such a big lead every game. Oh and Sasha Kaun. I think the team is ready to streak into the postseason like they did in Roy’s better years but they have found a different way to get things done.

  3. DavidH Says:
    February 13th, 2007 at 11:09 am

    Why for you cringe? Because you don’t like adjusted stats, or because you think I’m not calculating them correctly? Because the latter is certainly a possibility.

    And nice point on Kaun. If I use your Kaun-less numbers, the A&M game comes out to:

    133.8 … 87.1 … .9929 … 1

    and their line in second table becomes

    127.3 … 83.6 … .9921 … 6

    And yeah, I know I just took your numbers and converted them into cringe-inducing adjusted form. Sorry :)

  4. Chalmersfan Says:
    February 13th, 2007 at 12:35 pm

    I’m sure you’re doing them right… I don’t like them because I don’t understand why the margin of adjustment is what it is in many cases.

    Example: KU and UCLA have the same Raw Offensive Efficiency but UCLA gets a better adjustment. Why? Because according to Pomeroy they’ve played the 10th hardest group of defensive teams of any team. But how is that possible, they play in the offensive-minded Pac 10, and half of their 6 nonconference games were against patsies. According to this graph (http://tinyurl.com/38rtx5) at Big Ten Wonk most Pac 10 defenses are playing pretty bad. Ah, so that’s how it works- Pac 10 teams can’t stop each other so the offenses have their way, and if the offenses are having their way then the defenses deserve a big adjustment for playing such good offenses. And thus because Pac 10 defenses are adjusted to look a lot better UCLA’s offense will get a big adjustment to look better.

    Adjusted stats work well in the Ratings for comparing 2 teams, but I just have a problem with a system that will reward mediocrity with a big adjustment, i.e. Duke’s offense and Arizona’s defense. I prefer to use situational numbers (home, road, or conference) and the new game plan.

    JMO

  5. dominatindj11 Says:
    February 13th, 2007 at 12:39 pm

    that is what i was just talking about in response to another post. it just seems like this team flows better when Sasha doesnt play. We get into a rhytym and stay there when Sasha isnt in the game and when he comes back it disrupts our flow

Leave a Reply

Comment Policy:Be polite.

You must be logged in to post a comment.