What can we learn by breaking individual stats from the 2019 season into thirds?

Earlier this week, I wrote a shot post on how Marquette’s late season collapse in 2019 looks more like a reversion to the mean than the bottom dropping out unexpectedly.

This graph, in particular, blew my mind. Despite an abysmal record, the underlying stats weren’t that much different from the middle to the end.

But the eye test is tough to overcome. Particularly when it involves a candidate for Player of the Year shooting his way out of contention on a bum wrist.

A Twitter conversation with Phil Bush, one of the Scrambled Eggs Podcast hosts, particularly brought up a significant point. The eye test said that individually, most of Marquette’s players were worse in the final 11 than the middle, so how were the combined team stats so similar?

I’m not one to leave intriguing questions like these untouched, so I went back in to TRank and split up the season into thirds again, but this time, measuring just the individual player performance.

The results (again, leaving the NCAA Tournament game out of it) fit much better with the eye test than the chart at the top of the post. Markus was much more inefficient. Joey couldn’t break an ORtg of 100. Sam couldn’t stop turning the ball over.

Let’s start off with a simple, unadjusted Offensive Rating, broken down into thirds.

Ugh. You don’t need to get much further than the top 3 to see the stats matched the eye test, Marquette’s best players were significantly worse on offense down the stretch than at any other point in the season.

But ORtg needs the usage context, because Matt Heldt’s eye popping stats only came on a handful of possessions, so don’t come close to making up for the Big 3’s dip.

A couple things here:

• I can’t believe Markus’ usage actually went down, compared to where he was in the middle of the season. Don’t get me wrong, 37.5% usage is still really high, but it is a bit different from the tale I’ve been telling on Twitter.
  • Also, the fact that he had a 39.4% usage for an 11 game stretch and matched it up with a 115.9 ORtg still doesn’t get the love it deserves. He was Steph Curry good for 1/3 of the season against much better opposition. The injury killed what was looking like an all-time great campaign.
    • Of course, the fact that his usage was so high could be a cause for the injury/injuries towards the end of the season. The groin, the back, the wrist. One thing could have primed the other.
• Joey Hauser’s usage bumping up to 19.3% was not something I expected to see. It seemed like he was much less involved in the offense late in the year, not more. And, yes, I do realize 19.3% is not a very high usage, but it was more than his brother Sam.
• I can’t help but notice that as Theo’s usage went down (from 14.3 to 12.4%), his ORtg went up (from 111 to 129). It’s been proven that there is causation in this correlation, but the extent of the correlation seems intriguing.
  • Relatedly, as Morrow’s usage went down (17.9%), his ORtg also went up (from 110 to 113).
• On the other hand, the fact that Anim’s usage stayed level while his ORtg climbed up to 112.5 is a pretty big deal. He wasn’t a high usage player at any point, but going from an ORtg of 87.1 to 112 in the span of 20 games solidifies the growth the eye test showed. He is going to be a huge piece this season.
• One final thing that puzzled me, how did almost everyone’s usage go down without a corresponding step up from other players? I think it has to do with the way minutes were allocated as the rotation tightened. So Sam was still using 1/5 of possessions at 30 minutes a game as when he bumped up to 35 minutes.

And speaking of minutes:

Big spikes from Markus, Sam, Ed and Brendan. Noticeable drops from Joey,  Theo and Chartouny. Nothing too surprising here, but a good chart to be able to pair with the previous two as a reference.

This one is pretty self explanatory, too.

Howard’s 3pt% reverted to the start of the season percentage, but as his minutes and usage had increased, those misses were more noticeable.

Despite Sacar’s accuracy, the volume didn’t make up for the dips Markus, Sam and Joey experienced at the same time. I think this is one of the clearer charts that captures what went wrong down the stretch.

And so is this one.

Contrary to my (and popular) opinion, Markus didn’t start turning the ball over at a much higher rate. Everyone else did, though. Sam’s spike in particular, while not “high” per say, was particularly shocking. It’s just like nothing we’ve ever seen from him for any stretch of his career.

But then add in everyone else’s increase, and it makes perfect sense why the offense felt so flat.

Which just gets us back to the original question, of how the individual advanced stats can be so disparate from the collective team stats for that same period of time? Because if I had just shown you these graphs without ever mentioning the team ORtg at the top, you would feel validated in thinking Marquette (particularly Markus, Sam and Joey) collapsed at the end.

For starters, Sacar’s increasing efficiency helped stem some of the tide. He took on a huge burden every night of guarding the opponent’s best player, and yet still was able to become an efficient and productive player on the other end.

Bailey’s increase in minutes also helped, as he took some from Joey and Chartouny, both of whom slumped at the end. And while shooting 35% from 3 isn’t going to garner much praise, doing that after shooting 17.6% the previous 11 games is a pretty large leap to make during a season. For the Bailey stans of the world, this, and his flashes defensively, show why he could be a critical piece in 2020.

Less Chartouny and Cain didn’t hurt, seeing as any of their possessions taken by anyone else was a net win down the stretch. Again, we focus on individual usage rate as a standalone number, but it’s always a tradeoff, because someone else isn’t getting that possession. Ideally, Sam would have taken a good chunk of these possessions himself in those extra 5 minutes, but even if it was Markus at a lower efficiency than before, it was a net win for the team.

This is specifically true for the bigs. Theo John and Ed Morrow were 2 of the 3 most efficient members of the rotation down the stretch, but this also came at a time when they were getting fewer possessions than ever before. Ed’s ORtg went up 19 points from the starting 3rd to the finishing 3rd, but his usage also fell from 24% to 18%. Similarly John just wiped the floor of his early self, going from an ORtg of 88.4 to 129.5 at the end, but that was accompanied by a similarly large decrease in usage of almost 10% points.

I bring this up because it has a huge bearing on Marquette’s success in 2020. If Morrow and John could maintain ORtgs in the 105-115 range for the season while playing 40%+ minutes and using 20%+ possessions, I’d be all for Wojo’s idea to play longer stretches of game time with 2 bigs.

But Ed’s career high in ORtg came last season at 105.3. And Theo’s did too, at 105.9. That this came with 3 elite shooters at the floor most times makes it even more alarming.

Yes, players get better. Yes, changing the offensive focus to be more paint touch oriented could work wonders for them. Yes, including Jayce in the rotation could create plenty of size mismatches to exploit. But as we saw last year, and I think the numbers in this post have made quite clear, both of these players performed much better at a lower volume. It will be hard to lower their combined volume if you are doubling their presence on the court.

Adding a penetrating guard in Koby and seeing a more pass-first from Markus should make it easier for the bigs to score, even if it’s just on offensive rebound opportunities. But one last time, more opportunities wasn’t the answer last season.

I do realize there really was no other alternative in the transfer market that would have covered what Sam provided even halfway decently. I just want to reiterate once more before the season comes charging in through Marquette Madness tonight, Wojo is going against a season’s worth of evidence in giving the bigs a more prominent offensive role.

Here’s to vitriolic tweets and egg on my face come March.