Villanova Basketball Advanced Stats: The Big(s) Question

From: youngBUCK
To: Corollary

What sort of production do we really need out of Daniel Ochefu and JayVaughn Pinkston to compete at a high level? More importantly, what can be expected?  What will we do when Chief declares for the NBA draft?

Corollary's Aside: Quick piggyback on your last post/response, regarding Villanova's 3 point shooting for the year ahead. I've written about the 3 point shot EXTENSIVELY, guys, and if you don't agree with the points I made in articles before yB jumped on, and the same things we've hammered since he started, I've probably insulted your intelligence. Sorry, it's a bad habit. I'm working on it.

Threes are good, Darrun Hilliard and Kris Jenkins will crush it from deep, and I'm betting one of Josh Hart, Dylan Ennis, and Mikal Bridges will make the jump to be a respectable 3 point shooter - and maybe even all 3 will be fine, on lower volumes. Ryan Arcidiacono needs to dial it back a bit, but he did improve from deep from his freshman to his sophomore year. Another jump would push him past below-average territory, and I'd finally be OK with him shooting. Hopefully, the off-ball move will help him. On to the question.

From: Corollary
To: youngBUCK

We've already seen what we need from Chief and JVP, the only two nominal ‘bigs' on the team (JVP gets an include because he lives on inside, not because he's a ‘big' guy), to compete at a high level.  A 29-5 season with a Big East championship, a handful of top-25 wins, and a return to national prominence sure sounds like we've already been there.  The knock on last season, of course, is the 2nd round tournament exit. 40 minutes of sub-par basketball probably isn't the right data point to use as season-, program-, and philosophy-judging , but, hey. That's where we are as sports fans - recency rules.

The contributions from our big man rotation were an integral part of last year's success, and should be even more important as the team looks to make a real run at national relevance for our university and the Big East at large.  What do we need? More of the same - offense, rebounding, and defense - with ample space for growth.

Quick note - the database we can now pull from includes shot charts for all teams and players from major conferences last year, as well as lineup data from over 300 other NCAA teams. Expect the stats to get deeper and more ubiquitous, guys. Basketball's back, and it's exciting times.

Offense

First, let's take a look at the contributions each of these players made last year.

Where JVP and Ochefu really make their biggest contributions to the offense is setting picks for the guards, the true stars on that end.

Kidding, kidding. It's around the rim. This is something I went into rather in-depth in an earlier shot chart article for JayVaughn, so I'll just quote it here:

"The essential takeaway for me [...] is the value of the exact shots JVP is taking for Villanova. [...]One of the ‘value shots' we don't talk about enough in our quest for the perfect shot selection profile are those taken from the free throw line. A possession ending in free throw attempts, whether it's 2 or 3, is almost always the most efficient in basketball. Anyone shooting over 60% on FTs will score 1.2 points per possession or more on 2 free throw attempts, which is well above shots taken from anywhere else on the court.

Almost 90% of shooting fouls in the NBA game occur where JVP [and Ochefu!] takes nearly all his shots from, according the 82games.com article. The four zones (11-14 in the article, correspondingly) see significant jumps in shooting foul percentage. While only shots at the rim (13 + 14) carry appreciably better offensive ratings, according to their charting, it's tough to encapsulate the importance of drawing free throws, getting towards the bonus, and getting opposing bigs in foul trouble with literally any stats - one of the many limitations of traditional stats professionals and (us) amateurs use to help define the game."

The rest of the article can be found here.

These two bigs are responsible for a large portion of the team's shot attempts at the rim. In our sampled games last year, they accounted for 217 of the team's 521 shots near the basket (41.6%), and 84 of the 189 (44.44%) shots from the closest mid-range areas in Regions 3 & 4. They also hit these at a clip MUCH higher than Villanova's perimeter guys - an approximate hike of nearly 10 percentage points near the rim, and ~4-5% from the close mid-range areas.

These changes can be seen graphically, and in some tables describing that graphic action, below. The larger shot chart at the top of the combination is all of Villanova's shots last year - the bottom left is the perimeter guys, and the bottom right is our interior workhorses. The difference is pretty obvious - without the big men, the rim area goes cold.

Continuing to attack these areas is what the team needs from Chief and JVP for the team to succeed.  In an ideal world, we'd get all of our players to prefer to attack the rim - but what's most likely is the lion's share of this duty falling to the guys who are (and play) biggest within our scheme.

Here's a quick summary of what I expect to see, and what I hope to see, from each of these guys on offense this year.

Pinkston

There's not much I have to add about his offensive game that I didn't already share in his original shot chart article.  He's our 6'5" microwave post scorer, able to generate shots and free throw attempts from nothing more than an apparently threatening pump fake (seriously, why do people keep falling for that? I'm not complaining, just honestly wondering...) and an athleticism advantage over most of the heavy-footed bigs forced to guard him.  He's Villanova's volume attacker of the rim, most reliable source of points (an assertive Hilliard could change this), and an integral cog on the team.

I have been harsh on his game in the past, because it never seemed to suit the team basketball ideal that is Villanova's offense at its best.  He's a bit of a ball-stopper, avoids the extra pass too much for my liking, and really needs to stop taking 3s, period (the team doesn't need any more guys that think they can shoot from distance).  I'm just not a fan of forcing shots, which fuels my distaste (certainly overblown, but it happens).

These are mostly quibbles for one of our most efficient offensive options who has been for years.

My prescription for him on offense?

• Let the game come to you.
• Don't force shots when the matchups aren't there
• Make the easy pass when you get doubled.
• Keep doing the positive things you do.
• Play a bit smarter, but don't change anything else.

He's already at that elite level for the team.

Ochefu

Where to start? Ochefu certainly made strides on offense last year, developing a level of assertion to go with that 6'11" frame. He actually started making the open dunks that came to him, flashed a developing playmaking hand, and even started using some post moves in real games (with varying degrees of success). Sometimes, he even showed the initiative to drive to the basket and create some of his own offense.

That said, his offensive game was often on life support, so to speak - without supply for easy baskets from the team's playmakers, and dunks in transition, he really would have struggled to sustain his own effectiveness. A large portion of that is an offense that's not going to feed him, but, let's be honest - it's not like he earned those touches with a consistent level of play last year. He took way too many mid-range shots for a guy who barely cracked 30% away from the rim. He had one post move - back his guy down, until he either turned it over or took a hook shot (he'd pass out if he got doubled, to his credit).  His turnovers were a real problem - both in terms of offensive fouls (typically for moving screens), and just being overmatched at times handling on his own.

He's a raw big man, and this type of development is to be expected, especially in Villanova's system. And, honestly, a team with the type of playmaking ability on the wing and at guard Villanova has shouldn't be relying on a guy like Ochefu to create offense - finishing on his rolls to the rim and off dump-offs is typically enough.

For him to develop his game, he's going to have to add a wrinkle or two to his offensive stylings. On hook and jump shots last year, Ochefu shot 13 of 33 - just under 40%. In a related note, he only made 23 unassisted baskets the entire year. At this point, he has an unreliable isolation game, mainly predicated on his strength in the post and that aforementioned hook shot.  A short-range jumper, or some additional post moves, would add some variability to his game and increase his efficiency when he isn't being handed easy dunks to finish (which made up most of his assisted baskets).

If he can't reliably add a wrinkle, he really shouldn't seek out his own offense too often - it's just too inefficient (if it doesn't improve relative to last year).

The biggest markers for improvement on offense I'm looking for from Ochefu are: cut out the fouls, it's just unnecessary.  Keep your handle tighter - you're too good on the dribble to be turning the ball over that much for a non-primary ballhandler. Keep muscling your way to the basket, pass out of doubles, and get some more confidence in your overall offensive game. Keep nurturing the natural (and developing) playmaking skills, and don't try to do too much - just succeed to the largest extent possible within the right role.  Oh, make some foul shots, too.

Ochefu + JVP = Spacing Nightmare

Their minutes should be staggered more often than not, though.  Both guys like to operate in the same spaces around the basket, and while they can hook up on some slick interior passes, too often, there's just not enough room to operate.

Take a look at these splits:

Ochefu, who looks to be staying closer to the basket than JVP when the two play together, sees a general overall uptick in his numbers when sharing the floor with JVP. He takes more of his shots at the rim, and hits a higher percentage of them, while moving away from some of his jump-shooting tendencies.

JVP, on the other hand, clearly drifts away from the rim - Ochefu, or the big defending him, is already there, cramping the spaces he operates best in. He takes nearly 17% less of his shots near the rim when sharing the floor with Ochefu (though he does hit a higher percentage of them), and takes FAR too many 2 point jumpers while converting at an absolutely abysmal rate. Despite spending more time on the floor WITHOUT Ochefu than he does on, he took nearly the same amount of 2 point jumpers in the time the two of them spent on the court - and went 6 for 42. That 14.2% clip is scarily bad. JVP also tends to drift outside the 3 point line when Ochefu's in the game, something he's really not good at and shouldn't be doing.

It's not a combination Jay Wright should lean on heavily, as it gunks up the spaces and mismatches JVP-at-his-best exploits, and cramps the spacing of the entire Villanova offense.

Rebounding

Despite rostering just one player over 6'7" in 2013-2014 (and playing without that one in about half of each game), Villanova ranked just outside the top 50 in defensive rebounding percentage (or, preventing the other team from securing offensive rebounds). How is that possible?

The most important thing to remember about rebounding is:  it doesn't matter who grabs them. Rebounding totals are often conflated with rebounding skill, and it's a mistake easily made - rebounds are one of those stats most familiar to casual basketball fans, as they've been around forever, are consistently mentioned as a marker of contribution, and are often the cheap part of double & triple-doubles.  But it's consistently been shown not to be the case.

Even the rebounders who flash most in the box score - guys like Kevin Love or Dennis Rodman, in much earlier days - may not contribute much to their teams success by ripping down 15-20 boards per game.

Rebounding totals on a team suffer from the concept of ‘diminishing returns' - players compete for rebounds not only with the players on the other team, but also the four teammates they're sharing the floor with.  Pulling down a rebound does not always mean you're taking one away from the other team; a relatively large percentage of the time, you're just taking it away from a teammate.  If you're pulling down 15 rebounds a game, but 10 of them, under normal circumstances, would have just bounced to another teammate (not necessarily realistic numbers, just providing an example), the net contribution of that rebounding total is not as large as the number 15 would have you believe.  Several different advanced statistical studies, adjusting for this effect, have concluded that the rebounding of Kevin Love may be worth just 3 points a game to his teams - the numbers look gaudy, but the overall positive effect can be relatively small.  Just to note - you don't have to 100% agree with such studies - especially because the links for them online have expired - but there is mounting evidence that rebounding totals don't capture the overall picture of a player's rebounding skills - especially for bigs.

The most important part of rebounding for a big man is how well his team rebounds when he's on the floor. It seems like a simple concept - a team getting more rebounds is a good thing! - but it's often overlooked because of how easily countable rebounding totals for individual players are.

The best example of this in today's NBA is Pacers big man (and former Hoya) Roy Hibbert. The man is 7'2", and plays the majority of minutes at center for a great defensive (and middling-to-poor offensive - look at all those rebounding opportunities!) team, yet has only averaged 6.8 rebounds per game for his career. Conventional statistics paint him as an awful rebounder - but they're simply not right.

Roy Hibbert's teams have rebounded better with him on the court over his entire career. Even during the Pacers' post All-Star game freefall last year - when Hibbert went from averaging 11.8 points, 7.7 rebounds, and 2.5 blocks per game to 8.9 points, 4.7 rebounds and 1.8 blocks per - Hibbert's teams consistently rebounded at an elite level with him on the court. During this plunge, his team grabbed 52.5 % of all available rebounds, as opposed to 51.2 percent when he was off the court. Again, it's been a theme throughout his career - his low totals don't adversely affect his team.

A lot of this is due to scheme, and the excellent size/rebounders Indiana had at the other positions (hey, Villanova parallels!). Hibbert would diligently box out the other team's best rebounder with his enormous, difficult to move frame, and let everyone else go nuts.  But that's the point - as the tallest guys on the court, a big and his defender (and vice versa, of course) are the most likely people to come away with rebounds.  Preventing an opposing big from doing that, especially when most teams are loathe to send multiple wings/guards to the offensive boards, often frees a rebound to fall to other members of your team (and away from the best rebounder on the other team). Jason Collins, who did play basketball for over a decade before outing himself, built his career around doing this sort of grunt work around the basket, and allowed his teams to consistently grab rebounds at a higher rate when he was on the court despite posting mediocre individual totals.

Grabbing a high number of rebounds isn't necessarily bad, but ensuring your team grabs more than the opposition is what's important.

If you're asking 'Was an 800 word introduction to the stats you're going to show us in one subsection of this article really necessary?' you can go f%&$ yourself.

Yes, it was.

Stats

JayVaughn and Daniel, typically rotating at the team's nominal center position (though they did spend a fair amount of time on the floor together), were tasked with Roy Hibbert's job - box out the other team's best rebounder and grab the boards that come to you. Otherwise let the onslaught of tall, athletic-ish guards/wings snap up the rest. And it worked!

Danny Chief pulled down excellent individual rebounding totals - per KenPom, ranking in the top 250 of the country for both individual offensive and defensive rebounding rates - and he also made the team better through his typically good boxing-out skills. And, again, that's the key. The team's rebounding rates fell off significantly when he was off the floor (down 4% on the defensive end, 2% on the other side). The continuity provided by Ochefu and JVP, the team's best options against the other side's center, are the biggest reason I'm not worried about the team dropping off in the rebounding area after Bell's departure. His rebounds will likely just slot to the guys replacing him in the rotation.

JVP really did yeoman's work as the team's 6'5" center (hahaha) last year, as the team rebounded about the same percentage of the other team's misses when he was on the floor as opposed to when he was off.  Combined with JVP's relatively average individual defensive rebounding rate (it sat at 16.3%, well below James Bell at 17.1% and Chiefy at 21.8%), it tells you that he did an exceptional job of walling off bigger opponents and allowing teammates (like the aforementioned James Bell) to convert those defensive rebounds.  While there was a dropoff on the offensive end, it remains an impressive feat for a guy who's really playing above his size down there.

Continuing to do the grunt work, catching the rebounds that come, and allowing rangy, athletic rebounders like Hart, Hilliard, Jenkins, Arch (you can remove the athletic), and Ennis to snap up some of the individual rebounding glory, is what we got (and need) out of these two on both ends.

Defense

And, of course, rim protection! Villanova's defensive scheme is predicated on walling off the areas inside the arc (which was covered as part of a 3-part series I did on transition basketball here) - the team does a great job contesting and preventing rim attempts and two point jumpers, but a typically bad job defending behind the arc.

A key part of any team's scheme to protect the paint is the big men that will be closest to it. The numbers...back this up! Observe.

Team defense was better with both of these guys on the court.  Ochefu, in fact, reversed early season trends to become one of the team's (apparently) most important defensive players - the team played about 4 points per 100 possessions better (on the defensive end) with him in the game vs. him off.

Chief's rim protection credentials stand out the most in this chart. For a quick reference - S 2P ("Short 2-Pointers), or dunks, tip-ins, layups, and all other shot types that basically start at the rim.  L 2P is all 2 point jumpers.  You should know what the rest are. Comparing the stats when Chief was on the floor and when he was off show this effect. Team's took about 3% more of their shots from the restricted area when Chief left the floor (39.42% when he was in vs. 42.39% when he was off), which could point to his presence discouraging teams from aggressively attacking inside. While team's shot a bit better from in close when he was out there, it's really difficult to isolate, from just play-by-play data, whether those shots were taken when he was anywhere near the rim. Obviously, a 6'11" guy who jumps well enough and typically plays solid defense would appear to be difficult to score on when he's in close - but the play-by-plays let us do counting and lineup stats, not recreate the position of every player on the floor. I think his importance can be seen in the improved defensive rating and smaller percentage of shots taken at the rim.

JVP, on the other hand, tries gamely, but probably shouldn't be the team's long-term solution at center. I think he's fine in spurts (the team's second most-heavily used 5 man lineup, which swapped in Josh Hart for Ochefu, actually defended and scored better than the team's starting lineup in a significant amount of minutes), but his lack of true size tells in the rim stats. Team's got off nearly 5% more of their shots near the rim while he was in the game, and converted them at a higher rate. Lineups with JVP as the five should not drop out of the rotation, as they can absolutely blitz teams with spacing, athleticism, and interchangeability on defense, but it shouldn't be a normal state for the team. These lineups should be used to goose the offense, and throw a manic, pressing, turnover-forcing defensive style at the opponent; it shouldn't be expected to be a half-court defensive menace.

These two will generally rotate at center when they're not playing with each other, and I like both fine - but the team's defense could jump a level if Ochefu is ready to shoulder 25-30 minute loads (probably predicated on his offensive leap, or lack thereof). It's something to keep an eye out for as the season progresses.

And, I mean, do I really have to address Chief leaving? The guy's barely on the radar for most draft services, and he's 6'11". If he makes the leap and gets drafted in the first round, so be it. We'll thank him for the Final Four memories, and struggle through a year or two without a serviceable big. Worth it for me. If he makes incremental improvements, much like he did last year? I just don't see him getting a guarantee he'll be drafted, and if he doesn't have that, what's the point in leaving? His size will still be there, and he'll have a degree and an extra year of seasoning before the next round if he comes back. If he were a raw prospect teams wanted to draft on 'potential,' he would have already left. I don't see it happening.

Sorry I went so deep, guys; just working out the cobwebs. Good to have the season back, though.