WKU Basketball: Matt’s Stats: Examining Rick Stansbury’s Substitution Patterns
First of all, let me preface this: This is not an article to bash Rick Stansbury, but this is absolutely an article that uncovers some…
First of all, let me preface this: This is not an article to bash Rick Stansbury, but this is absolutely an article that uncovers some bizarre tendencies that break serious trends, both currently within the conference, out of conference, and comparing to WKU history all the way back to the mid-1970s (minutes were not recorded beforehand.
WKU is hovering a little bit outside of the top 100 in the country in the NET rankings, is 14–7overall and 7–2 in Conference USA. That is not a hideous record in the slightest. With some better luck, some of WKU’s opponents could have helped them have better national numbers.
However, despite being 14–7, there is room for concern. Is WKU operating a Ponzi scheme with its bench use? In other words, is WKU’s bench so thin that they may not be able to maintain this pace? When we look at the numbers, it is absolutely astounding when contrasting any tangible comparison.
When comparing to C-USA opponents in just about every measurable metric, WKU plays less of its bench than anyone else. When comparing to Western’s non-conference opponents, the same things happened. When comparing the rest of Rick Stansbury’s years, it seems to be a Rick Stansbury thing the more you break it down. In three of his four years, he has ultimately used a six-man rotation. Before and after the Charles Bassey injury, there is still the same tendency to play six or seven players, regardless of who is available.
Rick Stansbury is doing a great job this season. Let’s make no bones about that. With four comeback wins, one against an SEC team and three others from down more than ten points and all of them without Charles Bassey, an All-American caliber big man, there is no question he deserves a boatload of credit.
But again, this success is despite rotational frustration. As we get into this, look at the numbers yourself and decide. Either minutes don’t matter, Rick Stansbury has discovered something no one else rarely has, or Rick Stansbury is a complete and utter outlier in his substitution patterns and his players are way more likely to wear down over time.
You decide. I only bring the facts, and I will comment on what those numbers definitively tell us.
Purpose: To find out if my frustration about the rotation with WKU Basketball is remotely valid. If so, how extreme is the lack of rotation historically?
Parameters: I am going to break down all 20 of WKU’s regular-season games in 2019–20. We’ll look at the overall minute spread, individual game tendencies, look at the rest of the conference’s minute count, and even take a look at some previous WKU team’s tendencies. This is a deep dive, and I assure you I will let the numbers speak for themselves. Players will count as being part of the rotation if they both play at least ten minutes per game and have played at least 75% of the team’s games.
Hypothesis: Rick Stansbury plays his bench less than anyone else in Conference USA and his lack of substitution will rank as the highest in WKU’s recorded history.
*Note these numbers were before Thursday’s loss against FAU in which the bench only played 21 combined minutes (zero in the second half)
Analysis
WKU Minutes Stats
Six average 25 mpg or more.
Three average 32 mpg or more
There is no player (besides Charles Bassey, who does not count in the rotation) between Jordan Rawls (24.8) and Jeremiah Gambrell at 6.5 mpg
Top 7 players are only players to play at all in 85% of games eligible
Top 7 score 8+ ppg; Gambrell and others below him score less than 2 ppg
Matt Horton (6'11", 14 games played, 4.7 mpg) and Isaiah Cozart (6'7", 10 games played, 4.6 mpg) are one of only three players (Carson Williams) to shoot 50 percent or higher on the active roster.
Cozart averages a turnover every 16 minutes (fourth-least on the team); Horton turns the ball over every 32.5 minutes (least on the entire team).
WKU vs. Opponents
# of games WKU played all five starters 29* or more minutes: 7
# of games WKU opponents played all five starters 29* or more minutes: 1
# of games WKU played four of five starters 29* or more minutes: 16
# of games WKU opponents played four of five starters 29* or more minutes: 4
*29 minutes was selected because it is a huge tipping point for both WKU and its opponents. 30 was the original number I would have selected, but 29 seemed to be a number that grabbed more players into the statistic.
# of games WKU played seven players more than 10 minutes: 12
# of games WKU opponents played seven players more than 10 minutes: 18
# of games WKU played eight players more than 10 minutes: 1
# of games WKU opponents played eight players more than 10 minutes: 14
C-USA Team Statistics
UNT: 8 players at least 13 mpg
Rice: 10 players at least 10 mpg
UAB: 9 players at least 11 mpg
MTSU: 9 players at least 13 mpg
ODU: 6 players playing at least 14.7 mpg
Charlotte: 8 players at least 10 mpg
Marshall: 9 players at least 10 mpg
FAU: 10 players at least 12 mpg
FIU: 8 players at least 11 mpg
La Tech: 10 players at least 10 mpg
USM: 7 players at least 18 mpg
UTEP: 9 players at least 10 mpg
UTSA: 8 players at least 13 mpg
WKU: 6 players at least 24 mpg
C-USA’s number of players averaging more than 24 mpg (WKU’s rotation’s minimum):
UNT: 4
Rice: 5
UAB: 4
MTSU: 4
ODU: 4
Charlotte: 4
Marshall: 2
FAU: 3
FIU: 4
La Tech: 3
USM: 4
UTEP: 3
WKU: 6
Raise That Number to 26 mpg:
UNT: 2
Rice: 3
UAB: 3
MTSU: 3
ODU: 3
Charlotte: 4
Marshall: 2
FAU: 0
FIU: 3
La Tech: 3
USM: 4
UTEP: 3
UTSA: 2
WKU: 5
How about 30?
UNT: 1
Rice: 0
UAB: 0
MTSU: 2
ODU: 1
Charlotte: 3
Marshall: 2
FAU: 0
FIU: 0
La Tech: 0
USM: 2
UTEP: 2
UTSA: 2
WKU: 3
WKU All-Time Minutes Statistics:
2019–20 (current): 6 players with at least 10 mpg
2018–19: 6 players with at least 10 mpg
2017–18: 6 players with at least 10 mpg
2016–17: 8 players with at least 10 mpg
Seasons under Rick Stansbury with 6 players or less with at least 10 mpg: 3/4 (75%)
Total seasons dating back to 1975–76 (when minutes were recorded on the stat sheet) with 6 players or less with at least 10 mpg (not including the Stansbury era: 4*/40 (10%)
Total seasons 7 players or less had at least 10 mpg (not including the Stansbury era): 13*/40 (32.5%)
*1976–77 only had three players play 75% of the games, so that team only had three players play at least 10 mpg for 75% of the games. This is the only season in WKU history (with minutes recorded) with less than a (technically) six-man rotation. Obviously, the team couldn’t have less than five in the rotation in every game.
Conclusion
Rick Stansbury plays his bench as much or less than anyone else in every reasonable category. Most opponents are not even close to the short rotation WKU puts out. Even in games when the metrics used show an even minute distribution when comparing opponents, or the opponent plays roughly the same amount of players, the teams “equal” to WKU were almost uniformly barely fitting within the qualifications. For example, they would have a player or two with eight or nine minutes but not quite ten. Not only is it conclusive to say that Rick Stansbury is an outlier when related to rotating players, but judging by everything in the statistics, Rick Stansbury is the stingiest coach to sub out his players of anyone WKU has played this season. As a WKU head coach historically, he is the least likely to sub out a player.
In WKU history dating back to 1975–76, only seven teams have ever operated with only six men playing ten or more minutes per game. Three of those are since Rick Stansbury took over at WKU. Rick Stansbury’s percentage of seasons in which he ultimately plays a six-man rotation is 75 percent. In WKU’s history (not including Stansbury, that number is 10 percent). The number for a seven-man rotation still only brings in another nine seasons in 40. Rick Stansbury still sits at 3-of-4 with this metric.
One other thing that was striking when looking through these numbers (both current C-USA and previous WKU squads) was the incredible amount of times that WKU’s current rotation all play more than the third or fourth starter on some teams. Plenty of teams (five in C-USA) don’t even have one single person that plays 30 minutes per game. In other words, 1-out-of-3 C-USA coaches refuse to play their players 30 minutes per game on average. This was true for Western’s statistics, as well. Plenty of really good WKU teams from yesteryear rarely played a player more than 30 minutes.
When I posed the question on Facebook, I had some people mention 1970–71 as a year that there was very little rotation. According to them, mainly the five starters played.
Although this can’t be proven, this seems to be absolutely true. Only seven players could have possibly qualified, simply because at least 23 games had to be played to fit within the 75 percent rule I set from the beginning. The top five players seemed to play and produce the most statistically. However, this team still had depth. Although it was completely top-heavy, other players produced something when they came in. Jim McDaniels and Jerry Dunn fouled out at least six times on the season each. Two other starters (Jim Rose and Clarence Glover) also fouled out multiple times. Clearly, the Toppers’ lone Final Four team found a way to make it work without as much depth.
Oh, the irony that WKU’s lone Final Four team was the outlier to possibly debunk my theory about the fact that a team just can’t handle a season-long spring for a marathon season! Obviously, it is possible to win with five or six players. However, at some point, something crazy happens. Somebody gets hurt. Somebody fouls out. The game goes to triple overtime. Who, then steps up? The teams that have a little bit of depth have an ability to withstand those types of crazy nights more often.
In a tournament run, something crazy generally happens. The teams that make runs and the teams that win it all still generally have some close calls. Those close calls are saved generally by some random schlub coming off of the bench. Anytime that starting lineup and those few trusted players have a bad night, who can come behind them and provide ten or 15 points to save the day?
Also, just from a wear and tear standpoint, what kind of mileage will these players’ bodies have on them if they continue to each play upwards of 25 minutes a night? Sure they’re young and talented, but they’re human, as well. They face a lot of pressure when each of them plays such a pivotal role. With each of them playing 24 or more minutes per game, will each of them be able to do their part? 24 minutes is 60 percent of the game.
What do those numbers mean? Each night, every player that plays consistently for WKU is responsible for at least ten percent of the team’s result. How many scores are recorded where the score lies within ten percent? A vast majority. What that means is every player has to play well, or the Tops will lose. There is absolutely zero margin for error.
That is where the concern goes for me and probably a lot of knowledgeable basketball people. Heck, you don’t have to be knowledgeable to wonder how they do it twice a week as hard as they go. Imagine doing it three or four days in a row. That is where it gets the most difficult: Tournament time. What is the number one goal for this program? Get back to the NCAA Tournament. How does that happen? Win the C-USA Tournament.
It will be a difficult road if the Tops cannot find at least a seven-man rotation, if not eight.
Therein lies the problem.
What Stansbury Could Do to Fix the Rotation
WKU generally plays six players real minutes, right? That is definitely true. It can’t be argued that Stansbury is not adamant about playing his “best” players. This trend could continue, and you know what? Maybe the Tops win the C-USA Championship in March. For me, almost every great team in WKU history was deep. Most teams that make real runs in March have some depth.
WKU’s starting five depth is rarely matched in all of the NCAA. Very few teams have starters averaging eight or more points per game and yet another borderline bench player (Cam Justice) that averages well over ten points per game. In addition to that impressive stat, most teams that have a type of starting production don’t have the efficiency that WKU has (no current starter shooting less than 42 percent).
That being said, how do we minimize the wear and tear on these players moving forward? There is way more risk for soft tissue damage and more major injuries if players are on their feet running around more. With WKU’s breakneck pace, these guys are running miles per game, not to mention however much practice they get in each week.
The solution for Stansbury needs to be trusting his bench more. He has made it clear that he doesn’t want to trust players like Isaiah Cozart, Matt Horton, and Jeremiah Gambrell. He certainly is not ever going to trust Jackson Harlan or Patrick Murphy to do anything during important minutes. That is just fine. That’s your philosophy. Fine. Whatever. Agree to disagree.
What about those moments when WKU gets ahead by double digits? There have been several games that WKU had a lead and controlled the game of late. For example, MTSU was a perfect example of an opportunity to play a few guys in minutes that were not crucial until the outcome of the game.
WKU was up no less than 12 points up 24–12 with 8:37 remaining in the first half. That is 28:23 of a possible 40:00 that WKU was up comfortably. Sure, put in Matt McCay and that lead would probably crumble. However, I’m not the seventh man on the Tops right now, thank the Lord.
Cam Justice was injured against MTSU, so Gambrell was the sixth man, playing 19 minutes. For that to be your first man off of the bench and only playing 19 minutes in a blowout. Why? Jordan Rawls was 1-for-6 from the field but did have three assists and no turnovers. In 36 minutes, he only had one rebound, though.
In only 19 minutes, Gambrell was 3-for-7, had seven points, six rebounds, and had no turnovers in all of that time. Could Gambrell not play a few more minutes and take a few of Rawls’ 36+ minutes? Could he not take a few more Taveion Hollingsworth’s 36 minutes? Jared Savage fouled out, playing only 28 minutes. All of those guys are viably able to be subbed by Jeremiah Gambrell.
As far as big men, yes, Carson Williams is the only post player that can score on a consistent basis. I totally agree. Here’s a thought: Cozart played eight minutes against MTSU. Why couldn’t he play 15 in a game like this? Like we talked about, this game was out of hand for over 28 minutes.
Matt Horton played a little more than one minute, and Pat Murphy came in under a minute remaining. Was this not an opportunity to play some young guys? The closest margin of the entire second half was 14 points. Can you not throw in a couple of guys and see how they do with no game pressure? Let those guys get some experience. If MTSU goes on a 5–0 spurt or more, call a timeout and take them out. Anyone in the universe would agree with that.
If it’s true Matt Horton is in the dog house, great. There is still not much of an excuse to not let your players sit in what is still a long season left to play. Those minutes will add up and put more wear on the body down the stretch. There’s no way Savage, Williams, and Hollingsworth do not have a little less in the tank than they would if they were rested.
Another example of an opportunity for rest would have been Charlotte and Belmont. Charlotte was out of hand for the last seven minutes or so. Up 15+, WKU was well in control. The Tops did not sub out most of the starters until about two minutes remaining. Is this egregious? No, but again, it’s an opportunity to save someone from getting injured.
Belmont was another opportunity to try something different, to give the guys a blow. WKU shot 35 percent from the field. Of the six guys in the real rotation, only one shot 50 percent or better (Josh Anderson). Everyone else was 40 percent or under.
With Belmont dropping the ball in low and feasting on Carson Williams’ lack of height, could it not make since to play some height to help Williams in the post? This was at a time Horton was the first big man off of the bench. Since he has traded his minutes with Cozart. However, Horton was 2-of-3 from the floor in this game in eight minutes.
The point is not who specifically plays at the end of the day. What really matters is taking a few minutes a game off of the workload for each player over the course of the season. Then, when it really matters, those players could be at peak performance.
Every other team in C-USA has less talent than WKU without question. No other team has multiple Mr. Basketballs. No other team has multiple four-star players. Few other teams have a 6'11" player, period. Yet those other teams find ways to play their bench. Yes, the quality goes down of the individual play when moving down the bench, but those teams will be much more prepared when it’s time for those young men to step up in a big moment.
Again, this team is not in disaster mode. Clearly, they have got their identity figured out. Winning 70 percent overall and standing a game out of first place is a good spot to be nearly halfway through. However, the identity would not change by reducing their minutes by a few minutes per game.
It would allow confidence to flow throughout the lineup, and those that do come off of the bench would know without a doubt that they will be rewarded if they do well.
Stansbury needs to think long-term. Each game is important, but trying to win in March is the ultimate goal.
Think about it this way: If WKU is to win in March, they have to overcome the loss of an All-American, wear and tear on the few that get to play, no productive size, and a tendency to be inconsistent within games. The best way to alleviate a few of these issues is to spread out minutes.
You no longer have an All-American, so don’t play everyone All-American caliber minutes. You have fewer bodies, so protect the bodies you have, lest you have even less available when you need it. You have size on your bench, but it doesn’t play. When you need size against, say, Marshall or ODU, wouldn’t it be great to turn to a big man with some experience and give him 12 minutes to clog up the middle and take some time off of the clock while Williams, Savage, and/or Anderson rest? With more players, inconsistency would be minimized. As I mentioned a few times, playing only six players requires everyone that plays to produce something significant. If you play more players, there is less pressure on everyone to produce night in and night out.
Observing the Tops up close in Murfreesboro, Hot Springs, New Orleans, Birmingham, and Frisco over the years, the teams tended to be worn out in the second game. If the team had a bye, they then had enough to perk up and play for a championship usually. But if the team particularly did not have depth, everything was so much more difficult because the best players were being asked to play 110 game minutes in three days. That’s a whole lot of strenuous minutes.
All of this is to simply say, “Be cautious.” Burning your players out now means they may just not have it in March. That remains to be seen, but without question, Rick Stansbury is the outlier in the area of substitution in C-USA. So far, this preference has not produced a championship despite being heavily favored in one and appearing in two. This season, he will likely be expected to make another deep run, and his team will have a realistic chance to win in March.
Will they have the energy?