Breaking down WSU’s defense in the Colorado/Utah split

We’ve spent a lot of time focusing on defense around here, mostly because we knew that if anything different was going to happen with this season — heck, if anything different was going to happen with Ernie Kent’s tenure — the Washington State Cougars were going to have to somehow pair a competent defense with their competent offense.

It happened in a fairly shocking sweep of the Arizona schools, then took a step back in a loss to the Washington Huskies. It wasn’t a surprise that they lost to UW (everyone is doing that), but it was surprising that they went away from the man-to-man defense that had emerged in the desert. In a game in which they played mostly zone, they gave up 1.13 points per possession — not a great result after giving up less than 0.84 in both games in Arizona. (The median Pac-12 offense is scoring about 1.05 in league play.)

So, I was curious to see how and what WSU would play on defense against the Colorado Buffaloes and Utah Utes, figuring that the offense would take care of itself. I decided to chart each halfcourt possession. Here’s what happened.

A brief caveat about methodology

One of the things that’s always hard to parse out with data such as this is what’s actually due to the defense and what’s due to randomness.

For example, if we take any two possessions, one man and one zone, and each gives up a wide open three to a good shooter, each one has a roughly 60 percent chance of “success” — the shot not going in — even though the defense totally failed. If one goes in and the other doesn’t, we shouldn’t ascribe that to the defense. Yet, a chart like the ones below would show the one scheme as yielding 3.0 points per possession and the other as yielding 0.0 ppp. It would be silly to draw sweeping conclusions based on that, given that both possessions ultimately featured the same kind of shot. The difference is just luck — or variance, if you want to sound like you’re smart and look down your nose at the unwashed masses.

If I really wanted to do a thorough job with the charts that follow, I’d also try to chart how many of those possessions featured a strong contest, or a run-off the line, or whatever, trying to somehow quantify process instead of just results. I can’t stand results-based analysis in most contexts. But I didn’t do that here for a couple of reasons. First, I don’t know exactly what WSU was trying to accomplish with its defensive scheme against each opponent; I’m not in that room when they’re plotting strategy, and I’m not an Xs and Os nerd (even though I’m always trying to learn). Second, I just wanted to try and capture the broad strokes here. The results are definitely noisy — for example, there was a five-possession stretch of 1-3-1 vs. Colorado that was absolutely horrendous, but the Buffs badly missed three consecutive wide open threes — but the data can still give you a bit of a picture of what’s happening.

Zone to win against the Buffs

First, here is a link to the chart for anyone who wants to see the data or maybe even analyze it further. For the rest of you, here are the aggregate totals of the different basic sets vs. Colorado:

And broken down by man vs. zone:

WSU started the game against Colorado in man-to-man — all of those first half possessions were right out of the gate — producing some iffy results. When the Cougs switched to the 2-3 for the remainder of the half, the results got a lot better.

Was that because of the zone itself? I tend to think not. At least, not directly. One thing a zone will definitely do is entice three-point attempts. And against Colorado, that’s a pretty good strategy — the Buffs are shooting just 31 percent on threes in conference play, 11th in the league. They typically mitigate that terribleness by (wisely) not shooting very many of them.

The Buffs couldn’t resist against WSU. Against the man, Colorado hit three of its first six threes out of the gate on its way to 1.20 ppp. Ernie Kent changed to zone, presumably in an effort to switch up the “momentum,” and Colorado kept firing away: They would hit just 6-of-23 the rest of the game as WSU played zone on 36 of its final 41 halfcourt possessions.

To my eye, the zone didn’t look “better” than the man. But the goal of any defense should be to force the opponent into an uncomfortable shot. Against Colorado, getting them to jack threes with the idea that they’ll probably miss a bunch of them because that’s what they do is a smart strategic decision, and it worked out for WSU — even if they still needed a couple of colossal blunders by the Buffs to salt it away.

Sometimes, though, allowing lots of threes is a bad idea!

Again, here is a link to the chart for anyone who wants to see the data or maybe even analyze it further. For the rest of you, here are the aggregate totals of the different basic sets vs. Utah:

And, again, broken down by man vs. zone:

Sooooo ... yeah. A bit to unpack here.

First off, Utah shot 16 of 35 on threes in the game. The Utes are the best three-point shooting team in the Pac-12, hitting nearly 40 percent (!) of their attempts in league play. They also shoot more of them than anyone else. Thanks to that, Utah features the best offense in the league.

This really isn’t a situation where you want the opponent taking a ton of deep shots, and as the game unfolded, it became really, really clear that the zone was going to be a very bad strategy against a team that shoots it so well. In this case, I think the results in the table pretty much tell the story on that one, since it’s certainly easier to make a wide-open shot — and there were lots of them against the zone.

So, the Cougs did what should have logically limited those three-point attempts: they played primarily man-to-man.

Only ... it didn’t work. Utah still shot threes on roughly 60 percent of its attempts, well above its league-leading mark of 47 percent. If there’s one thing we know from watching our own team, it’s that if a team wants to shoot threes, a team is going to shoot threes pretty much regardless of what the defense does. However, I think the attempts being so far above the season average speaks to the kinds of looks Utah was generally getting from beyond the arc — looks it would be silly for a good shooting team to pass up.

Where Utah really excelled was in getting the Cougs out of whack with their switches. This newfound man-to-man for WSU generally runs as a “switch everything” defense. That can be effective, but it also can be vulnerable when you’ve got, say, a 5-11 point guard who routinely gets switched onto a big in the paint, or when you’ve got, say, a 6-7 center who routinely gets switched onto the point guard on the perimeter. (Hypothetically speaking, of course.)

The Cougars cleaned some of that up in the second half, but the Utes still regularly put WSU in compromising positions, then used that to get the Cougs’ floor balance all screwy. Ahmed Ali and Viont’e Daniels, in particular, repeatedly got caught either ball watching or sinking off their man in a weak attempt to contain penetration, only to have their man catch and bury a three before they could recover.

This is where WSU really is going to have to work hard to take the next step defensively. The scheme has potential; 1.19 ppp allowed with the man is not good, but it’s also not embarrassingly bad, considering Utah has scored at a 1.11 clip against everyone in the conference. But some of these issues need to be cleaned up, whether it’s getting more discipline out of the perimeter players or anticipating the ways in which teams are going to try and set up and exploit mismatches. The Cougars did a pretty good job of rotating Robert Franks onto the really bad post mismatches; they’ll need to do more of that sort of thing.

I honestly don’t think it will take much to bring the defense up to a competent level for these last couple of weeks. And it really helps that they’re heading off to play the two worst offenses in the conference: Stanford and Cal.

It probably doesn’t matter anyway ¯\_(ツ)_/¯

Look, coaches are who they are. Upon witnessing a defensive performance that produced as bad of results as any this season, Ernie said:

“There was a few times in that second half where we’d cut it to 10, nine, right around in there, and a missed free throw here, missed free throw there, loose ball here, charge here, block. Just didn’t have it.”

Yes, of course — had WSU scored 14 more points, they’d have won the game. But to do that, WSU would have had to exceed Utah’s 1.34 ppp clip to do it, something they have not done even once in 27 games. They came close against Idaho, scoring 1.33; the most they’ve scored against high major team (and I use this term extremely loosely) is 1.17 against Cal.

I want to believe defense matters enough to Ernie to clean up the errors and improve the team’s chances of winning. But when he walks away from that game and says to the media “our offense just wasn’t quite good enough” — even though that game represented the team’s seventh-best performance of the season on a per possession basis — it makes me think that the preceding 1,600 words are probably just wishful thinking.

Oh well. This was a fun exercise. I’ll watch again anyway.