Before we get too far into this blog post you should know some things about me: I am an engineer by education, I know a little bit about a lot of things, I have two kids under five, I believe in a rich tapestry of pop culture references that would make Bill Simmons say uncle, and in college, a roommate once said about me “it’s almost like you can physically see the trains of thought collide in his brain”. Why do I tell you all this? Mostly so you are prepared for the poorly written stream of consciousness blog article you are about to inflict on yourself; also know that if I’m taking time to do this is matters cause….kids are a lot of work and I’m inherently lazy.
So over the last couple of weeks, there has been a lot of noise about the current head coach of Marquette University’s men’s basketball team (I had no idea where to put the apostrophe(s) so this is a great start) Steve Wojciechowski. Over the first five seasons of his tenure, the general fan attitude of Coach Wojo (that’ll save me some letters) was always seemingly “he’s fine but…..” or “he can really recruit but…..” and the but is always followed up with a list of complaints five or six deep. Things like “Wojo can’t adjust” or “Wojo just rolls the ball out for Markus and sees what happens” or “unless he has a powerpoint for it, Wojo can’t solve it”. You get the idea. It’s not flattering.
Well, this season the “yeah, buts” have gotten, we’ll say, more vocal and the disdain for Wojo has gone from the retched hives of scum and villainy that is Twitter and message board forums to out loud in person booing. Certainly concern has been warranted given some fairly spectacular screw ups in the current season. First there was the way Wojo handled the last 6 minutes of a winnable game against Providence, but then two weeks later he completely forgot the score in a tie game and ordered his team to foul with 15 seconds left. It was just dumb and given all the “yeah, buts” that people already had with Wojo, it clearly broke a dam. In all the teeth gnashing, booing, complaining, and otherwise general panic within the fanbase I spent some time trying to defend the Wojo era, in large part because I tend to be a pretty optimistic person and I really don’t want to go through another exhausting coaching change. So I would post things things about how particular times he made adjustments in games or cite outstanding articles like this one from Paint Touches which shows Wojo’s After Time Outs (ATO) plays are spectacular. However, it never seemed to convince anyone other than those who were already convinced, so I started to think perhaps I was wrong and my optimism was pointing me in the wrong direction. It’s true that Wojo seems to make adjustments and his ATO plays are good, but then there are the blow out losses to good teams and the perception that Wojo just can’t win games when it matters. This last thought stuck with me: does Wojo really not win in the clutch? Is it possible that Wojo can coach for the most part but when it matters, with the blood pumping in his ears and moments left on the clock, does he fall apart? That’s when the engineering part of my brain kicked in(see, the rambling preamble had a point) and said “ya know what that question is? That’s a hypothesis.” So that’s the point of this post, the journey I start on to determine if Wojo is a clutch coach with the hope that if I could answer that I might be able to settle the question of whether he was a good coach or not.
For those of you who aren’t familiar with statistical analysis (Why would you be? It’s very boring) you always have to start with a hypothesis. Basically a statement of something you believe to be true and then you use data to try and validate the statement is true…..if you can’t then it’s likely false. So let’s start there:
Hypothesis: Wojo is not a clutch coach
OK, cool so now I just need to test that against some data and I can kick back and bask in the glory of the six nerds who will actually want to read about it. Let me go find that data…………….yeah, so it doesn’t exist. Guess I’m going to have to tell #noredforOwen to go spend some more time building puzzles by himself, I’ve got a data problem. So to solve the data problem I had to figure out what data I needed to validate whether Wojo was clutch. I had to figure out what games would show whether Wojo was clutch and decided that “close” games and their result would suffice for my data set so let’s go figure out some close games. I settled on a highly sophisticated and calibrated algorithm of, well, looking at KenPom game summaries and applying some rules of thumb which would classify whether a game was close or not. Those rules of thumb are as follows:
-if a game goes to OT, it’s close (really stressing the ole’ brain on this stuff)
-if a game is within 5 points either way with 5 minutes to go, it’s close
-if a game is within 3 points either way with 1 minute to go, it’s close
-if the win probability flips from one to the other or is around 50/50 with 5 minutes or less to go, it’s close
-if the final score is within one possession, it’s close
With those rules in place, I determined that there have been 84 games in the Wojo era classified as close. For context, there have been 189 games in the Wojo era which means 44.4% of Wojo’s games have been close. Neat, what does that tell me? Shrug emoji. That could be great, could be bad, could be average. So here’s where we get deeper into the statistical nerdery….I started by gathering a data set on Wojo, but I need a control group. I need the same data set but on other coaches to give me some sort of indicator as to where Wojo falls in the continuum of clutch. So I did what any lazy person does, I crowd sourced via Twitter DOT com. I asked the internet what seasons or what coaches they thought were clutch and boy did you guys deliver. Here is the list of Teams – Coaches (Season Year) that I decided to make a part of my control group:
Marquette – Buzz Williams (2013)
UConn – Jim Calhoun (2011)
Maryland – Mark Turgeon (2016)
Wisconsin – Bo Ryan (2015)
Butler – Brad Stevens (2013)
Michigan State – Tom Izzo (2016)
Villanova – Jay Wright (2016)
Virginia – Tony Bennett (2016)
Some brief (yeah, right) commentary on the control group. This is somewhat of an arbitrary control group that is focused on teams in the last 10 years (peak KenPom data set) and it is by no means exhaustive. It does have a lot of coaches in it that the general public would say, yes they are a great coach so it feels like a solid control group. As people suggest more “clutch” teams they can be added to the data set, but this is what we have to roll with now. The control group has 108 close games out of 300 games played giving an control average of 36%, meaning we have our first comparison….Wojo has played more close games then our control group on average. However, lets dive deeper as I gathered a lot more information than just close or not close (if that was all I had this would be a short and boring article instead of a just boring one). For each close game I collected details about the game like Home/Away/Neutral, KenPom rank of opponent, and then used those attributes to assign a quad classification as defined in the Net Rankings (yes it’s a mixing of concepts but Net rankings are produced by dark elfin magic for all I know but the Quad format makes a lot of sense). I’m using the Quad format as a metric of game quality, ie a Quad 1 win is a better quality win than a Quad 4 win. So with all that, we come to our first data visual
A couple of things jump out just by looking at this table. Holy salmon on an iced cracker, that UConn 2011 team must have killed a lot of its fans! Generally speaking the new old Big East (or is it the old new Big East) was RIDICULOUS and something like it will never be seen again. Second, this season has had a lot of close games already and we’ve got four to go…buckle up!
So let’s break down the quads and see if anything jumps out for Wojo
First, red cells indicate the result is below the control average in each quad while green means above that average. The first thing that jumps out at me is that the coaches in the control, by and large win the games you are supposed to(Q2 and below). Buzz’s numbers might be a little misleading because again the old new Big East was insane so while he lost more than half of his Q1 games, he had a lot of them. What also stands out to me is the almost linear progression of Wojo’s performance in the Q1/Q2 games. This is a clear indication to me that the complaints about Wojo’s ability to win close games was warranted but he has steadily improved on that culminating in an outstanding performance to date this season. We can’t throw a ticker tape parade for him on this data alone as we don’t know if this improvement is because of better coaching or better players, though both would be directly impacted by Wojo’s tenure. Either way, we’ve got something useful out of this (YAY!), so let’s press on.
We’ve got some evidence that Wojo has been getting better at winning tough games that are tight but how many tight games are happening/should happen? I’ve got a graphic for that.
The first column is a calculation of how many games the team played that have been tight, theoretically you want this percentage as small as possible (assuming you have a good record) because then you are winning even tough games handily. Looking at the control group, other than essentially the super good old new Big East teams, good teams have a relatively low percentage of tight games they play. Conversely, Wojo’s teams have seemed to have a good amount of tight games, especially this season. That’s a strike against Wojo to be sure, but there is some context. Look at the Schedule Strength column, MU this season has faced very high level of competition in the tight games so playing good teams tight isn’t necessarily a bad thing especially when you combine it with the fact that Wojo is by and large winning those games. Very notable is the middle column, Wojo this season to date has been outstanding at winning close games. Yes the Butler and Providence games stick out in a very negative way, but other than those two games, Wojo is getting the job done. One of the reasonable conclusions is that Wojo is getting on the job training and it seems he is applying those lessons this season. One note on the last column, this is a made up metric (by my crazy brain) but it basically is a weighted score of tight games (translation a tight Q1 game is weighted more than a Q2 one, etc). What the metric does a good job of showing is A) I’m super duper glad I wasn’t a UConn fan in 2011 (though they won it all so I guess it worked out) and B) I hope all MU fans have seen a cardiologist at some point this season cause the ol’ ticker is getting a work out C) that generally Wojo’s teams have been playing tighter games at higher levels than most high quality teams historically which tells me that Marquette isn’t an elite team right now (when you’re done picking yourself up off the floor from shock please keep reading).
Let’s look at this another way, let’s compare the amount of winning in tight games to the difficulty of the opponent:
(Note: Red lines represent the control group average) The take away? MU as a team is continuing to improve in the Wojo era, but they also we pretty bad originally. One note, if not for the collapse at the end of the season last year (all late losses against almost all Q1 opponents) MU would have been approaching elite status last season (no more letter writing campaigns, m’kay?).
For me a picture is starting to form, that Wojo theory is coming into range (we’ll get to the title at the end), but I was troubled by one other thing. Going back to the Paint Touches well, there was concern about MU not being able to win away from home (namely they are 1-25 against KP Top 25 teams in the Wojo era), which is a problem. Winning on the road/neutral sites is the cornerstone of good coaching, without it the tournament is going be a hot mess. So that’s what this next chart is, how Wojo (and the control group) did in winning tight games at home versus in general
The long and the short of it is you want to be as high as possible on the chart and as far to the left as possible…anything in the upper left is elite level coaching (god I hate Bo Ryan showing up here). Both the raw data and the graphic show that the Wojo era has shown growth here. He definitely still seems to have too much dependency on the home court in close games (ie close games on the road are more likely to fold than not) BUT it is getting better.
Let’s close this out and get back to the whole point of this post. In physics (again the opening paragraph matters ), there are four interactions that define how the universe works – electromagnetic, gravitational, weak, and strong, and these interactions have their own theorems and laws. However, physicists have been trying for decades to unite these interactions into a single theory in which they all are interrelated, or a Grand Unified Field Theory (Einstein famously tried and failed to unify his general relativity model with electromagnetism). Well, taking that as inspiration, and using the research of others as well as what you’ve seen in this blog post, I’ve created a Grand Unified Theory of Wojo, one that unites all of the concepts, complaints, and concerns that exist within the #mubb galaxy brain…..are you ready?
Wojo is a decent coach, good molder of men, and great recruiter who is learning on the job and showing improvement. That’s it, that’s the theory. Every indicator is that Wojo was average to below average early in his career and has gotten better. Take this stat collected from poster Frenns Liquor Depot on MUScoop.com (yeah I know, but remember I’m lazy so here we are), Wojo’s record against KenPom Top 25 teams:
2015: 0-12
2016: 0-6
2017: 1-6
2018: 0-10
2019: 4-1
2020: 2-5 (with 1 top 25 team to go)
Certainly not a ringing endorsement but it’s the trend line that matters. Also keep in mind that Top 25 teams are Top 25 teams because they are hard to beat. We can argue all we want about Wojo’s mistakes, his failures, and his lack of public persona but what I don’t think we can argue is that Wojo is getting better every year as a coach. I have no doubt that some of that is due to improving talent on the roster, but if Wojo isn’t responsible for that I don’t know who is. Further evidence of that is this very recent article from Anonymous Eagle. It is clear that the team is not yet elite and certainly has warts, but we may be getting fewer warts each successive season. It’s also clear that Wojo is learning to win the tight games that he wasn’t winning early in his career. Keep in mind that Wojo has A) never been a head coach previously and B) where he was the lead assistant his teams almost always had the talent edge over opponents. Simply put, Wojo hasn’t had to “coach hard” until he got to Marquette. I think the narrative of Wojo being a bad coach (generally people view him as a good to great recruiter) got set early with some teams that had considerably less talent than fans had become accustomed and that narrative has been hard for Wojo to shake.
What I am ultimately saying here is that the chaos we all perceive is the symptom of a coach learning what it takes to win every day in one of the best conferences in the country. The Chris Beards or even Buzz Williames of the college basketball world are by far the exception. Tom Izzo wasn’t Tom Izzo until he was, same with Jay Wright or Tony Bennett. Sometimes these things take time and sometimes we have to give people the space to learn and grow. Wojo has demonstrated he is growing and I see no reason to believe he won’t continue to grow. I get the frustration of the two steps forward, one step back cadence to the Wojo era, but one of these days I think we’ll see steps forward with very little steps back. All of these lessons learned may pay off as soon as this season in the tournament where winning tight games is what it is all about. If Wojo can make a run in March, riding the nuke all the way in will have been worth it.
No comments:
Post a Comment