A regression-based dose of hope
Posted by kj on Thursday, February 14th, 2008
Before we get to more number crunching, you should check out the Bleacher Guy’s take on MSU’s seven-year conference title drought. It’s a critical, but balanced, commentary on MSU’s struggles in conference play in recent years, compared to the improvement over the course of the full season the early Izzo Final Four teams exhibited. As far as this season’s team goes, here’s the key paragraph:
But a lack of toughness doesn’t make you turn the ball over nearly 20 times a game. Toughness won’t help you overcome an offense described as “very predictable” by analyst Tim McCormick (another huge fan of Izzo, by the way), or offset a 21-0 deficit from three-point range in a game against Purdue. Sometimes it seems that the constant search for toughness, or for another leader in the mold of Mateen Cleaves, is a bit of a red herring. I want Michigan State to be a tough basketball team, but I’m not sure I buy “get tougher” as answer to every question that’s raised about the Spartans anymore.
Definitely click through to read the whole post, though. Note that I’ve linked to it twice to emphasize how worthwhile it is.
OK, now to the numbers. After my stab at using linear regression to examine MSU’s turnover woes, there were a couple of good suggestions for revisions to the model from the comment section faithful:
-
GBBound noted there may be multicollinearity between the opponent’s defensive TO% and defensive efficiency stats. (“Multicollinearity,” by the way, is a great word to throw around to make yourself sounds smarter than you really are. Which is not to say that GBBound isn’t as smart as he sounds.) In other words, defensive efficiency is partially based on defensive TO%, so including both as variables is redundant.
-
Hubert suggested a dummy variable for Big Ten opponents, on the theory that our conference foes know us better and can, therefore, come in prepared to disrupt our offense.
A third possible variable occurred to me: A time-based variable to see if MSU is improving or getting worse at holding on to the ball once you control for other variables. Taking these three concerns into account, here’s the list of independent variables for the new model:
-
AWAY: Dummy variable for away games. Playing in hostile environments could cause a team to turn the ball over more. (Note: Texas coded as home game; Missouri/BYU coded as away; 0.5 for UCLA game.)
-
DefTO%: Opponent’s defensive turnover % (for the season). The more turnovers an opponent tends to create, the more turnovers MSU is likely to make.
-
DefShoot%: A stat I cobbled together based on the opponent’s effective FG% and free throw rate (for the season). It’s the team equivalent of PPWS but on the eFG% scale. Replaces defensive efficiency as a measure of whether a team forces tough shots–as opposed to being good at creating turnovers. It could be MSU turns it over forcing the action against solid defenses.
-
PACE: My working theory has been that the slower the pace of a game, the more MSU bogs down in the half-court offense and turns the ball over.
-
BIGTEN: Dummy variable for whether the opponent is a Big Ten team.
-
TIME: Variable running from 1 to 24 for the sequence of the games MSU has played to date.
The dependent variable is MSU’s offensive turnover percentage (MSUto%) for a particular game.
Data was pulled today, so it accounts for two additional games. Here’s the equation resulting from the linear regression I ran in Excel:
MSUto% = 38.70 + (2.05 x AWAY) + (0.42 x DefTO%) + (.06 x DefShoot%) + (-0.38 x PACE) + (10.83 x BIGTEN) + (-0.82 x TIME)
This model has two advantages over the previous model:
-
The R-squared is higher: 0.54 vs. 0.44. So we’re explaining a little more of the variation in MSU’s turnover % from game to game.
-
The directions of all the coefficients are intuitive. MSU turns the ball over more (1) on the road, (2) against opponents that create more turnovers, (3) against opponents that force tougher shots, and (4) against Big Ten teams. The turn the ball over less (1) when the game is played at a higher pace and (2) as the season progresses.
The big difference versus the previous model is in which variables are statistically significant. Previously, they were the opponent’s defensive TO% and pace. Neither is significant in this model. Rather, the Big Ten dummy variable and the Time variable are now significant–both at the 0.025 level.
Implication: It’s playing Big Ten teams that are familiar with MSU that causes them the most problems in holding on to the ball. A higher percentage of MSU’s Big Ten opponents (1) create more turnovers defensively and (2) play at a slower pace, which accounts for why those two variables showed up as significant in the previous model.
As for the Time variable, this makes some sense in light of this graph:
The last time we looked at this graph–right before conference play started–we were feeling good about the downward trend in turnovers over the nonconference season. Then MSU turned it over 25%+ of their possessions against their first three conference opponents and we didn’t feel so good anymore. Since then, you’ll note there has been a subtle downward trend in turnover % again, albeit not as distinct as the first downward trend. The Big Ten dummy variable accounts for the spike going from nonconference play to conference play.
The regression results above indicate that (1) playing a Big Ten opponent makes MSU’s turnover % go up nearly 11 percentage points per game (!) and (2) after you factor everything else out, our turnover % is going down about 4/5 of a percentage point each game.
Those are some dramatic results. I’m not sure if their magnitude makes me more or less likely to believe this version of statistical reality. But it’s a pretty good reality: It implies MSU is, in fact, improving over the course of the season and can be expected to perform much better once we get out of conference play and into the NCAA Tournament.
I’m a tad skeptical, though, for three reasons:
- What my eyes see when I watch MSU play make it really hard for to believe this team is improving.
- If they are indeed improving, there is a natural limit to how much further improvement can occur. You can’t go down 4/5 of a percentage point of TO% per game indefinitely.
- At some point, won’t it be patently obvious to nonconference opponents how they should play us? Slow down the pace, force us into a half-court game, and do everything possible to disrupt us from running our set plays.
Time permitting, I’ll try to run last season’s data through this model to see if it holds up.
Anyway, here’s hoping the numbers trump my intuition in this case. An NCAA Tournament run would be great. Of course, adjusting our offense to become less predictable against Big Ten opponents would also be great.
Update:
Responses to comments:
1) Sample size is clearly an issue. Tough to do a study of a single team in a single season.
2) The interaction of TIME and BIGTEN is a key issue. Could be a fluke of the small sample size. The fact that both are statistically significant is curious, though.
3) There may be an issue in measuring quality of opponent outside the defensive variables.
4) I get a predicted TO% of 17.8% for the IU game.
In a real-world conversation, BoilerBrewer (Official Purdue Fan, Home Brewer, and Economist of the Spartans Weblog) notes that OLS really probably isn’t the best methodology here–since TO% can never go negative. To which I replied: It’s the only methodology I know how to use!
Here are the results running the model on last season’s data:
MSUto% = -2.96 + (1.06 x AWAY) + (1.93 x DefTO%) + (-0.41 x DefShoot%) + (0.19 x PACE) + (2.70 x BIGTEN) + (-0.32 x TIME)
R-squared is lower: 0.34. Coefficients, including the constant, vary pretty extensively from this season’s results.
The significant variables are DefTO% and TIME. BIGTEN has a pretty large coefficient, but doesn’t meet the statistically significant threshold.
I think, at this point, the thing to do is . . . give up. But hopefully this exercise has at least been useful to think through some of the potential issues causing MSU’s turnover woes. And I’ll stick with my intuitive judgement, which these models have generally confirmed:
MSU’s offense has to become less predictable, through some combination of pushing the ball more on offense and allowing more offensive freedom for our guards (Lucas, in particular–but Neitzel and even Walton, as well) when our set plays aren’t working.
Filed in links, msu basketball, stats analysisComments Off
No Responses to “A regression-based dose of hope”
gbboundon 15 Feb 2008 at 10:21 am 1kj-
Good stuff.
1. That constant of 38.7% is just sad. I wonder how it compares to other schools.
2. With as many variables as you have, it certainly increases df. Right now it is at 17, and you typically want 20. Certainly more game games played will get it to 20.
3. I wonder if your Big Ten dummy variable and time have a bit of an interaction effect, in that we are approaching the point where half of TIME is the Big Ten.
4. Plugging in the numbers with Indiana, who have a DEFto% of 20.025, an average pace of 67 possessions, and game #25, one would get that the offensive turnover% = 14.0305 +.06*DefShoot%, correct?
Spartalyticalon 15 Feb 2008 at 10:49 am 2Might the TIME be interacting with who we happen to be playing as well? I.e., the order in which we play better/worse teams (defined other than by DefTO% and DefShoot%) may be playing an unwanted role? I’m in stats and numbers, but it’s been a while since I’ve been heavy into regression analyses.
Huberton 15 Feb 2008 at 11:13 am 3Great fun, these regressions! With only 24 observations, you need to check for multi-collinearity, and you can probably eliminate correlated variables.
BUt I want to address the comments fromt he bleacher guy. I think there may be a degree of truth in this critique, which has become a common one in the blogosphere. We know there has been a kind of script in recent years, and it is becomming tiresome, particularly when the absence of big ten titles is not redeemed by a long dance in march.
But I do think it is only a partial explanation for the team’s problems this year and last. I also want to point out two other real differences between the glory years and the last couple years. The most obvious difference is the talent level. There is no obvious NBA player on the current roster, and I believe Richardson and Randolph were the last NBA drafts from MSU. the 2000 team had four NBA players (Bell, Cleaves, Peterson and Richardson) plus about 5 players good enough to play in Europe or Australia. I guess some will argue Morgan is a future pro, but I think he is too small to play the four in the NBA and lacks the handles or the shooting range to play the three. Maybe Lucas? Yes, indeed, his impact on games this year just highlights the lack of athleticism of players like Neitzel. In any event, I would argue that the talent level is just much lower than a decade ago, and that before the current freshman class, there were a couple really mediocre classes, and that this gets reflected in the results on the court. The limited athleticism of this year’s team makes them more reliant on Izzo’s sets, and much more vulnerable to a break down in the play. I don’t remember specific sets being so dominant ten years ago. I remember a lot of times, the clock running down and Cleaves manufacturing something on his own, or MSU getting points off of offensive rebounds.
Linked to this is that Izzo’s recruiting has resulted in some odd redudancies and weaknesses. The glut of pretty similar wing players, for example, and the absence of talented big men the last couple years is striking. Not totally Izzo’s fault, since Gray was so highly thought of as a HS players. The absence of outside shooters, particularly from the forwards is another significant lacuna. Granger’s outside shooting was a key dimension of the 2000 team.
A second key difference is the size of the guards in the program. Cleaves and Bell were fantastic rebounders from the guard positions. Big strong guys. In rugged big ten play, the current team suffers hugely from the small size of Neitzel and Walton. Izzo’s system works much better whith big guards — the 2005 team in the end and despite their inconsistency had big strong players in the back court. Think Torbert, Hill, Ager and Anderson, all 6′4 or taller.
kjon 15 Feb 2008 at 11:30 am 4Good comments, once again. See the update to the post above.
With regard to Hubert’s comments, I’ll think some more about this but I’m not sure about the talent thing. Of the group that won the first three Big Ten titles, only Cleaves was highly touted coming in as a freshman. Bell and Peterson simply improved consistently over their careers to the point of being NBA-level players by the time they graduated.
Then we hit the more talented group: Richardson, Randolph, Taylor, Torbert, Anderson, Davis. This created problems with (1) players leaving early and (2) finding cohesiveness.
I think Izzo’s been trying to find a balance since then–getting guys with enough talent to compete, but not obvious NBA early-entry candidates. On paper, this team seems to strike a pretty good balance in that respect. Hard to say if Izzo would be better off going the UNC/Duke route (take the talent, even if they leave early) or the Wisconin route (focus on getting less talented players who will learn the system) or sticking with the current approach.
Agree that (1) Cleaves’ ability to create shots was a key for the early Final Four teams and (2) lack of outside shooters hurts the balance on offense. Hopefully, Lucas and Allen can fill those voids going forward.
Donaldoon 15 Feb 2008 at 2:08 pm 5Shannon Brown was a highly ranked player as was Paul Davis. I believe that Brown was drafted in the first round and I seem to recall Davis and/or Ager as second round picks. The current group of Spartans are weak as far as big guys. Philosophically, I think to keep a program at a consistent level Izzo’s formula is a good one – bring in talented guys that will stick around and develop over four years. I do think it is helpful to have one impact player such as an Oden, Beasley, or Gordon that can bring you to the promised land, even if only for a year.
The problem I see is that the system doesn’t fit the athletes we are bringing in. We don’t have a dominant big guy that we can feed in the post and get easy points. We don’t have any consistent three point threats. We have a few good ball handlers and some quickness. We rebound the ball well. The offense needs to be geared to our strenghts – transition baskets, more penetration and movement to create space for open looks and drives to the basket. We shoot free throws fairly well (up until recently) yet our best free throw shooter rarely gets to the line, and most games our free throws are not the result of aggressive moves to the hoop. J.J Redick was an outside threat, yet how often was he shooting free throws.
This is where our coaching staff has failed. They haven’t changed the system to fit the strengths of players we have recruited. I think that is a big piece of why we don’t see much growth in our team and players over time.
Huberton 15 Feb 2008 at 4:20 pm 6Yes, of course, I don’t know why I forgot about Ager/Davis/Brown, although I suspect none will end up staying in the league long. That class would have done really well, if only the team had had one other big man — for two years, it lost every big game in which Davis had foul trouble.
But what kind of system do you build for the current personnel? The one big weapon offensively is Neitzel outside shooting, and indeed, the team spends the first thrity seconds of every 35 second possession trying to open him up for a shot. I do agree, though, that in recent losses, the team did turn into a static jump shooting team, and failed to get Suton and Morgan touches.
TMadisonon 15 Feb 2008 at 5:09 pm 7I am surprised that talent level is truly in dispute. We have been consistently getting appropriate recruits for Izzo’s system. However, early entrants to the draft has hurt in some years. I believe that it is plain-and-simple Izzo’s system, combined with mediocre talent, that leads to sligthly above average finishes in the rough-and-tumble Big Ten. I obviously want a Big Ten title, but I would trade them in for the excitement that Izzo brings to the table come March. It’s the best playoff system in all sports, so being tied with UNC for the most visits to the Final Four in the last seven or eight years works for me.
I am not delusioned about the talent level of our conference. There is a clear-split and we are one of the few upper-level teams that loses to the likes of Iowa and Penn St. However, I think that we are team more geared for neutral court endeavors with faster-paced squads.
Not too much to back this thought process up. Call it blind faith in Izzo, I guess. I just want to see some progress before March. Maybe some year we can take a title.
Oh yeah… regarding that comment from Hubert about the lack of a big guy down-low that can fit into Izzo’s system, hopefully Delvon Roe will be that piece. Maybe Raymar’s stock is slipping and will stay another year. If so, that has to be the clear-cut best inside presence that we have had since the title. Right?
Go Green!
Donaldoon 15 Feb 2008 at 5:45 pm 8Hubert,
I am not trained in the Xs and Os of basketball, but I observe closely. I think we can make better use of our speed to disrupt the defense. We rebound well. If we push the ball up the court after the rebound we can catch teams off balance. I realize in the Big Ten there are fewer opportunities to fast break, but we can beat the team down the court with our speed guys before they have a chance to set up defensively. With more space on the floor and particularly Lucas’s ability to penetrate, there can be opportunities for easier shots or the chance to draw a foul.
In the half-court, I would utilize Suton’s passing ability to a greater degree. He has a knack for finding the open man when he gets rid of it before he has time to fumble it. I would also use Lucas more as a penetrator, but he does need to learn to kick it out. Once he drives, the defense is out of balance, and with a kickout and some quick passes we can get easy shots. One strength of this team is unselfishness and passing ability. We have had many games where the assist-to-basket ratio has been quite high. Let’s take advantage of our strengths.