Examining Michigan’s transition offense

Dylan Burkhardt
on

Michigan-at-Bradley-24[1]
Dustin Johnston

The numbers say that Michigan has played progressively slower basketball in every year since John Beilein arrived in Ann Arbor. The Wolverines were the nation’s 189th fastest team in Beilein’s first season, 2008, and rank 328th in adjusted tempo thus far this season. Through 11 games, Michigan is averaging roughly the same number of possessions per game as the most chided slow paced team in the country, Bo Ryan’s Wisconsin Badgers.

Despite what the tempo statistics say, something about Michigan’s slow tempo doesn’t add up. Listen to any pundit, announcer, opposing coach or analyst and you’ll hear them rave about Michigan’s transition offense. And they aren’t wrong; Michigan’s transition offense has been extremely effective. Although the tempo statistics don’t show a significant increase from year’s past, even casual spectators can see that this team has been more effective pushing the ball and scoring in transition. What gives? We take a closer look.

Michigan is running more often than we’ve seen in recent seasons. According to statistics by Synergy Sports, 17.2% of Michigan’s offensive possessions come in transition. That’s a 25% increase from a season ago when just 13.8% of Michigan’s offensive possessions came in transition. Despite ranking in the bottom quarter of the league in possessions per game, Michigan ranks squarely the upper third when comparing percentage of offensive possessions in transition.

Team Poss. PG % of Poss. in Transition
Indiana 69.4 20.4%
Michigan State 65.6 20.2%
Ohio State 68.1 18.2%
Michigan 62.6 17.2%
Iowa 69.4 16.6%
Minnesota 67.2 15.3%
Penn State 64.6 13.1%
Illinois 67.3 13.0%
Purdue 67.5 11.3%
Northwestern 63.8 8.8%
Nebraska 62.2 8.1%
Wisconsin 62.4 7.0%

Source: Ken Pomeroy, Synergy Sports

While traditional slow Big Ten teams like Northwestern, Nebraska and Wisconsin have possession per game tempo figures closer to the Wolverines, they all push the ball on less than 10 percent of their offensive possessions. Meanwhile Michigan’s 62.6 possessions per game is a far cry from the 68 possessions per game that the other five Big Ten teams with at least a 15% transition percentage average per game. Michigan’s combination of transition offense and slow pace is abnormal and scatter plotting the data makes it fairly obvious:

image

The other 11 Big Ten schools appear to show a general correlation between tempo and transition percentage while Michigan is squarely in the upper left quadrant – by itself. There are a couple of explanations for the discrepancies. First of all, if transition opportunities aren’t there, John Beilein is very content to run his offense and utilize the entire 35 second clock. Michigan has a point guard he trusts and more than enough scoring options to make plays late in the shot clock. Second of all, it means that Michigan has done a decent job preventing transition opportunities for its opponents while also forcing opponents to utilize a majority of the shot clock before attempting a shot. Extending the possessions of opponents can still shorten the number of possessions in a game relatively quickly.

The story here isn’t just that Michigan runs more than its tempo statistics would indicate. The Wolverines are also a significantly better team when they push the ball.

Through 11 games, the correlation between Michigan’s tempo and its offensive efficiency is actually stronger than the correlation between shooting – the most obvious of the four factors – and offensive output. Pomeroy’s correlation score (correlation multiplied by 100) is +83 for tempo to offensive efficiency, compare that to just +7 a season ago and it’s clear that Michigan is a more effective team this season when it pushes the pace.

Correlation Offense Defense
Pace eFG% TO% OR% FTR eFG% TO% OR% FTR
To Offensive Efficiency 83 77 -9 9 2 -6 -33 21 -25
To Defensive Efficiency 6 -19 6 17 74 87 -34 48 -4

Source: Ken Pomeroy

According to statistics from Synergy Sports, Michigan is scoring 1.31 points per transition possession – a dramatic improvement from the 1.09 points per possession that the Wolverines netted on transition possessions last season. That number ranks in the 96th percentile nationally and compares favorably to the rest of the conference.

Team % of Poss. in Transition Points Per Transition Poss.
Minnesota 15.3% 1.35
Michigan 17.2% 1.31
Ohio State 18.2% 1.30
Indiana 20.4% 1.19
Wisconsin 7.0% 1.16
Purdue 11.3% 1.15
Illinois 13.0% 1.01
Michigan State 20.2% 0.98
Iowa 16.6% 0.96
Penn State 13.1% 0.94
Nebraska 8.1% 0.94
Northwestern 8.8% 0.82

Source: Synergy Sports

Good things happen when Michigan pushes the ball in transition. Obviously there are more easy opportunities in transition (the Wolverines average .96 PPP in half court sets – a very good figure in its own right) but the Wolverines have maximized their transition effectiveness. A counterargument to Michigan’s transition efficiency could be that the Wolverines are so selective as to when they push the ball that they are bound to be more efficient than teams that push the ball in almost every scenario (Michigan State, Indiana). But a majority of that credit has to be given to Trey Burke, who is able to control the game so effortlessly and make the right decisions whether to push or slow the game.

Burke, Stauskas and Morgan are all very good in transition but Tim Hardaway Jr. has been Michigan’s shining star in the full court offense. According to Synergy Stats, the Wolverine swingman is scoring 1.7 points per transition possession compared to just .82 points per possession half court situations. Hardaway is 18-of-20 from the floor when pushing the ball and scores at least 1 point on 79% of the transition possessions that he utilizes. To borrow from MGoBlog, cackle with knowing glee when you see Hardaway streaking down the floor with the ball in his hands after a defensive rebound.

Transition opportunities are generally created by live ball turnovers or clean defensive rebounds – either by a guard or from a great outlet pass. The Wolverines have struggled to create turnovers but have been very strong on the defensive glass, ranking fifth nationally in defensive rebounding percentage. Finding easy opportunities in transition will become increasingly difficult as play becomes more physical in Big Ten play but the combination of turning defensive rebounds into easy offense could be the most telling barometer for Michigan’s success in league play.

  • Adam

    One of the big reasons for our increase in transition offense is the improvement of THJ’s handle. It makes such a difference in the offense.

    • http://www.umhoops.com/ Dylan Burkhardt

      Yep. Hardaway has been absolutely lethal.

    • gpsimms

      Timmy’s handle is still bad. The improvement is because he is killing the glass, I think. In transition, Timmy’s handle doesn’t matter so much because he just outruns people. He’s so fast. He did that last year, the thing is he only rebounded 11.4% of opponent’s misses vs. 17.6% this year.

      When I see him get ripped 30 feet away from the basket or when he drives into traffic, I really don’t see any improvement from his dribbling last year. He is passing well, and making better decisions. The transition thing, though, is all due to the rebounding in my opinion.

      Honestly, if we had a guard that could dribble (or if Stauskas wasn’t the slowest person in the history of slow white men), I think he’d still be at the three.

      • John

        Saying the Hardaway’s handle hasn’t improved is just ridiculous. When West Virginia was pressing and playing a complete ball denial defense on Burke, Hardaway would have the ball in has hands for the 15 seconds of the shot clock basically playing the point. Do you really think he could of done that last year? And for your second point about not having an off guard the can beat someone of the dribble thats the beauty of a Beilein offense, You can have players like Novak and Douglass who aren’t going to blow by people on the dribble but if you run his sets the right way the offense itself will create a shot.

        • gpsimms

          Timmy’s usage stats are all at career lows this year. He shoots less of the teams total shots, and assists on less of the teams total shots than he ever has. This means the ball is actually in his hands less often than in the past.

          Even with that, his turnover rate is at an all time high this year.

          I believe the stats agree with me here.

          http://kenpom.com/player.php?p=10308

          Note: I am NOT saying tim is having a bad year. In fact, this is his best year. I just think he’s still not a great dribbler. He’s out of position (even though he insists that it is his ‘natural’ position), and is helping the team by playing that spot, and doing well considering.

          note note: he’ll be a 2 in the nba, but that’s a lot less handling responsibility than in beilein’s two guard sets.

          • http://www.umhoops.com/ Dylan Burkhardt

            I understand the stats, but don’t agree with the conclusion.

            I’m just not sure you can say that Hardaway isn’t more comfortable handling the ball away from the basket this year. He’s driving more aggressively than he has in year’s past and is getting to the rim much more regularly. Some of this is a product of his rebounding (obviously dramatically improved) but Hardaway rarely rebounded the ball and took it the length of the floor last season – we see that constantly this year and he’s been successful with the ball in his hands full court.

          • gpsimms

            I don’t know man. I think the big thing is that his 2pt% is up. So of the shots he gets off, more go in. If that’s what you mean by “with confidence,” then ok. he is turning the ball over on the same rate when he goes to the basket as any other year, id guess. it’s just when he gets a shot off, more are going in. we remember made buckets.

          • mikey_mac

            I’m with you. I think the proof is in his efficiency stats … very efficient in transition, much less so in the traffic of the paint in half-court sets, even with a good 2pt%. He’s displaying more “confidence” in that he’s taking on more ball-handling (although likely just a function of his new position), but it’s really only paying off in transition. I find it hard to make the argument that overall he’s been a markedly better ball handler this season.

          • Champswest

            In other words, Tim’s handle passes the eye test. I would think that everyone (except gpsimms) would agree with that. Not everything can be explained or proven with numbers and charts.

          • UM Hoops Fan

            Whatever the merits of Timmy’s handle, I’m not sure a lower shot-rate and assist-rate = ball is in his hands less often. Especially in JB’s offense, a player can handle the ball a bunch without racking up a ton of assists or shots. Stu was a good example of this, I believe. In fact, the higher TO rate is evidence against your conclusion, isn’t it? In my completely anecdotal guesstimate, THJr has been invovled in initiating the offense more this year than last, which would make sense given his position change.

          • gpsimms

            yeah, i regretted writing that right after i did. you’re right, it def does not mean the ball is in his hands less often. however, the conclusion i am pushing is more that his handle isnt improved. a higher TO rate with lower Arate and less shots taken seems to support that.

            should not have said he touches the ball less, though.

        • Mattski

          I’d split the difference–Timmy looks quite improved to me, but remains vulnerable. It’s his initial burst that makes those plays go, though, more than his dribbling, wrestling down those rebounds in the first place.

          • gpsimms

            He does have lightning first step, and huge long strides. At first I thought Manny was crazy quick with subpar dribbling. Timmy is another level, completely makes up for lack of dribbling skill with insane feet.

      • Mith

        Stauskas is the slowest person in the history of slow white men? Have you forgotten Smot already? They’re not even comparable.
        Actually, I’m not sure I can take you seriously…I think Stauskas is reasonably quick.

        • gpsimms

          Haha, good point about Smot.

          OK, but seriously, I am not talking about quickness, I am talking about speed. Next time you see Stauskas in a full-out sprint to get back on transition defense, get back to me. I believe you will be convinced that he has a mind-blowingly low top speed. It is important to have top speed if you are a guard.

          Note also that Stauskas is an amazing shooter and has never gotten a successful backdoor cut, while Vogrich, who is an average shooter is good for a backdoor cut every 15 minutes of playing time. Vogrich is fast, Stauskas is not.

          • Mattski

            In a related vein–is Stauskas getting any rebounds? Where can I go to look at box scores of all the games so far?

          • rlcBlue

            33.
            Here.
            I read somewhere that one of the differences between the 3 and the 2 is that the the 3 man is usually responsible for boxing out one of the opposition’s primary rebounders while the 2 is usually free to go for the ball. That’s one of the reasons Hardaway is getting so many more rebounds this year.

  • Dollarz2982

    Pushing the ball up and down the court to fast causes more turnovers which we don’t need to do cause it helps the other teams have opportunities to score more points and possibly beat us in games. Just let the boys do as they have been cause it has kept us undefeated so far.

    • http://www.umhoops.com/ Dylan Burkhardt

      I don’t think there’s a 1:1 correlation between tempo and turnovers. Michigan does a great job of avoiding turnovers as a rule and perhaps pushing the pace could have the opposite effect and force opponents to turn the ball over more often.

      • rlcBlue

        The table in the article shows a negative correlation between tempo and turnovers – both on offense and on defense. Michigan turns the ball over on slightly fewer possessions when it plays a faster game, but forces fewer turnovers as well.

        • http://www.umhoops.com/ Dylan Burkhardt

          The table is showing correlation between turnovers and offensive/defensive efficiency in that scenario. Not tempo. Sorry if that’s unclear.

          As in, as turnover percentage increases, Michigan’s offensive efficiency decreases — which makes sense.

          • rlcBlue

            Oops. Thanks for clarifying.

    • Billiam

      The end goal of offense is points. Yes, we may turn the ball over in fast break opportunities, but we still score more points per possession, which takes turnovers into account. Even with more turnovers, our offense is more efficient.

      This isn’t arguing for MORE fast break, it’s arguing that the amount of fast break we have has been great for us.

  • gpsimms

    I see that we have more transition opportunities this year than in past years, but I don’t think this is due to any philosophical change. The number of possessions is always low because we limit other teams’ transition opportunities by not crashing the offensive glass too hard. I don’t think we’ve ever not looked to run on offense when we get the chance.

    I think this has been the case for a while. The big noticeable difference this year is in your last parahraph: we own our defensive glass this year. 76.6% DReb compared to in the high 60’s in previous Beilein teams.

    That rebounding effect is compounded by the fact that our defense is holding our opponents to the lowest Efg% in the Beilein era by far as well.

    I would wager that if you looked at some sort of %of running possessions per rebound/steal, that the ratio is about the same as always. Our defense is better, so we get more opportunities, so people are noticing more.

    This does not explain the improved efficiency however, and I definitely am in agreement with the fact that this team is more dangerous on the run than any before. I just disagree that we’re *trying* to run more often.

    Great post, good food for thought.

    • Mattski

      Fair criticism of what is nonetheless really fun analysis by Dylan. Would I be naive in saying that this is probably an ideal way to play the game? I mean, to be able to run. . . and be able to work through your offense, patiently, from one player to another–each a threat to score–until you obtain your good look.

      • gpsimms

        I absolutely think so. I love this offense. I love what coach has done. It’s definitely not exactly the same as it was when Beilein first got here, obviously, and I think Meyer, Bacari and Jordan deserve a lot of credit. As well as to coach for getting a great staff and being flexible.

        But I think the overall philosophy: we will get some run, our opponents never will, is definitely the way to go and the way Beilein has tried to go since getting here.

  • http://www.facebook.com/dan.lavoie.58 Dan Lavoie

    Hardaway is the Real Deal.

  • rlcBlue

    Great topic for a post.

    The upshot seems to be that Michigan now has players who can score in a variety of ways, and they are doing a great job of choosing which style to use at any given time. Beilein’s offense has always relied on good decision-making by the players, and having more players with more skills creates more possibilities to choose from.

    Clearly much of this is due to Burke, but the ball reaches the other guys as well, and the fact that they are making the optimal decision so many times a game is the reason Michigan has been in the top five all year.

    Of course they aren’t perfect – Burke sometimes relies on himself too much, Albrecht tends to throw backdoor passes to guys who aren’t quite open, Stauskas will dribble into traffic – but they are very, very good. And there is certainly hope that a team with only two upperclassmen in the rotation still has room for improvement as the season rolls on.

  • UM Hoops Fan

    Nice work Dylan — this article was on the afternoon links in the ESPN College Hoops Blog. Also noticed UMHoops.com on their blog roll.

  • mikey_mac

    Love this type of post! Great job.

    I agree with another poster, that our number of transition possessions probably highly correlates to our defensive rebounds … the only thing to see here is how much better we’ve been with these opportunities.

  • jakelam2116

    Great post, Dylan. This makes a lot of sense and all your numbers point to just how comfortable Michigan is in any kind of offensive situation (half court, fastbreak). Good back-and-forth on Hardaway, too. I sit in the middle. He has improved his handle, but there are also times — a couple on Saturday — when he looks shaky.

    On an unrelated note, I found it silly how concerned people were with Hardaway’s shooting “slump.” As you noted, his stroke is beautiful and most of his misses were just short or long. Saturday’s shooting performance didn’t surprise me at all.

  • David