It will be interesting to watch these numbers evolve over the year. So far we have the types of numbers we’d expect. We have three tougher games coming up, so it will probably be worth another look then.
You need to be careful with Ken Pom until about early January. The data now is not entirely based on the games of the early 2014 season.
Before the season, you may have noticed preseason rankings. From what I have gathered, Pom comes up with these predictions through a formula, then uses a moving average some it’s some predicted data and some real data. This should clear up as I said in early to mid January and then it is all performance and not prediction.
For example, you will see Illinois has had a much more efficient offense this season than Michigan, but ranks below them in offensive efficiency; if he did not do this, his website would be useless for the initial part of the season. FWIW, the predictions end up being pretty close to accurate but are really hard for some teams where the addition or subtraction of one player or many players changes the dynamic of a team.
If you want to use Pom early in the season, then you need to be a member for sure. You can check a lot of valuable statistics outside efficiency that can give you a good indication of a team. So far, with such little data, the only stats so far that I think hold merit are the tempo stats. Despite what it may look like with Levert and Walton pushing the pace and the team playing more transition basketball, they are still very slow in pace, 280th in the nation.
First post on the board, but I will try and stick to metrics and questions about recruiting.
Agree with using the KenPom rankings this early… They are still weighted for his preseason projections.
Michigan’s transition game is lethally efficient, but it is different than teams that like to run all of the time. Still, it’s devastating enough that it only takes a few mistakes to really turn a ball game around (i.e turnover for a dunk then two missed shots to early clock threes).
KenPom wrote something about this being a slow season (http://kenpom.com/blog/index.php/weblog/entry/the_slowest_season) and the more I think about it, the more I don’t think pace can be equated to transition.
Michigan runs so many 5 seconds possessions and so many 30 second possessions. It would almost be most useful to be able to see a full distribution, because there are probably other schools that mostly run 20 second possessions.
Lack of TOs will always work in our favor, I’m not convinced we will get clean enough looks against better teams to shoot on par with the last 2 years. Time will tell
Sorry to keep going post to post, but I don’t foresee it being much of a serious problem. With two good creators in Levert and Walton, they should be able to collapse plenty of defenses. Last night, guys who got plenty of clean looks last year as well, just missed them. Also, I think Chatman is going to be able to create a new dimension for that offense once he gets comfortable in game action and will certainly be able to make plays to create shots; I think he will be able to open up the floor far more than Glenn Robinson did. I simply do not see a situation where as the year progress he is much less effective for the team at the offensive end than G-Rob; points, yes, but he will make it into the same deviation of help with passing and creating. Nonetheless, you’re right, each game will bring a different challenge and lead the team through a different evolution- only time will tell.
On that topic of Chatman, I think he may be the best option to go in the “middle” against Syracuse’s 2-3 zone. Donnal would be a good option as well, because of his perceived ability to hit mid range shots, but Chatman’s length and passing ability actually make him a better choice
I totally agree with the transition thing. We are lethally efficient when running, problem is we are simply not a transition heavy team, hence the transition effectiveness is mitigated.
Matt,
I agree that JB would prefer to score out of the half court set usually, but he I also think he wouldn’t mind speeding things up in tempo right now to get easy buckets. The problem is the way teams are defending Michigan. While the hottest topic with Michigan is the lack of an effective big, they are still one of the 3 most efficient defensive rebounding teams in the nation right now. Why? Teams are not crashing the offensive boards, because they are afraid of Michigan’s transition game. If this continues, I am more than fine with it. The teams they play will get bigger, so it likely won’t, but putting Michigan in the half court will eventually do very little to benefit the opposition, because of their effective scheme, excellent shooting, and lack of turnovers. The latter two quantitative measures have been very good thus far also.
Chatman isn’t the best option in the middle of that zone, quite frankly he’s the only option
Dylan,
Do you mean going to a % of possessions under 10 seconds, % 10- 20 seconds, % possessions 20-30 seconds, etc?
You need to be careful with Ken Pom until about early January. The data now is not entirely based on the games of the early 2014 season.Before the season, you may have noticed preseason rankings. From what I have gathered, Pom comes up with these predictions through a formula, then uses a moving average some it’s some predicted data and some real data. This should clear up as I said in early to mid January and then it is all performance and not prediction.
For example, you will see Illinois has had a much more efficient offense this season than Michigan, but ranks below them in offensive efficiency; if he did not do this, his website would be useless for the initial part of the season. FWIW, the predictions end up being pretty close to accurate but are really hard for some teams where the addition or subtraction of one player or many players changes the dynamic of a team.
If you want to use Pom early in the season, then you need to be a member for sure. You can check a lot of valuable statistics outside efficiency that can give you a good indication of a team. So far, with such little data, the only stats so far that I think hold merit are the tempo stats. Despite what it may look like with Levert and Walton pushing the pace and the team playing more transition basketball, they are still very slow in pace, 280th in the nation.
First post on the board, but I will try and stick to metrics and questions about recruiting.
Taking care of the ball skews tempo stats, though. Turnovers are directly proportional to # of possessions. If Michigan turned the ball more, say the median 13 times per game, it would boost them up the tempo rankings.
Dylan,Do you mean going to a % of possessions under 10 seconds, % 10- 20 seconds, % possessions 20-30 seconds, etc?
Yeah I’d love to see a distribution down to five second intervals combined with efficiency. Could probably put some of that data together at some point, could be interesting.
Dylan,Do you mean going to a % of possessions under 10 seconds, % 10- 20 seconds, % possessions 20-30 seconds, etc?
Yeah I’d love to see a distribution down to five second intervals combined with efficiency. Could probably put some of that data together at some point, could be interesting.
I think it’d paint a better picture of transition if you take TOs out of the equation.
Chatman isn't the best option in the middle of that zone, quite frankly he's the only option
What about Max if he has actually found his shot? Especially since his passing is looking pretty decent.
I think it’d paint a better picture of transition if you take TOs out of the equation.
I think that’s hard to say. You certainly can dictate tempo and pace with your defense to “speed people up.” However, I agree with what you’re saying. If you are like 2013-2014 Indiana, your pace can be deceptive because of how badly you turn the ball over.
Here’s Indiana’s profile from last year:
http://kenpom.com/team.php?team=Indiana&y=2014
They were 63rd in the nation in tempo, but 330th in turnover percentage offense and 328th in steal percentage offense. However, I don’t think you can say that the relationship of the pace at which they intended to play did not have a factor in those statistics. Frantic, hurried offense can often lead to more turnovers, more steals, and actually in their case, more blocked shots. I also think examining a team like Indiana will show you just how bad their man in charge is.
Chatman isn't the best option in the middle of that zone, quite frankly he's the only optionWhat about Max if he has actually found his shot? Especially since his passing is looking pretty decent.
Not really efficient because Kam would then become a corner shooter, and that will essentially allow the D to double the entry pass or another shooter because nobody respects Kam’s shot
I think Kam is a liability on offense right now because the game is just to quick for him. I think it’s been stated on here a couple times his level of compitition at the high school level wasn’t the greatest. He has the natural skills and once the game slows for a bit at the college level I think we will start to see those skills come to the surface note. I felt the same way watching Walton and Zak last year and McGary to an extent the year before. Obviously all three are different position and type players but the speed is the same mindset wise. The same might apply to DJ as well, I feel like by beginning to middle of Big Ten play we will see those two start to make serious contributions.
Chatman isn't the best option in the middle of that zone, quite frankly he's the only optionWhat about Max if he has actually found his shot? Especially since his passing is looking pretty decent.
Not really efficient because Kam would then become a corner shooter, and that will essentially allow the D to double the entry pass or another shooter because nobody respects Kam’s shot
Unless Kam is on the bench, and the two point guard lineup is in there, then you have 4 guys that can hit the three on the perimeter.
Donnal is definitely an option. Irvin can run the baseline as well
Don’t think Donnal is a guy you want at the high post making decisions at this point. He lacks the requisite assertiveness and confidence to make quick, efficient decisions
Been thinking, Dylan.
If you fit an equation to represent the spectrum of expected value from 0-35, it would probably look like the cumulative distribution function, where the seconds are worth a lot at the beginning and decrease in value. You could do that and standardize the results of every possession based on time, but If there actually are team specific tendencies like you present with Michigan, the pure estimate would be worse than the observed results that don’t factor in time.