#PreviewSZN in full swing (Hoop Vision Weekly: 10/4/19)

No hot takes on California's legislative branch to be found here...

It’s a Friday and coaches are in press conferences talking about “creating a culture,” so you know what that means:

Another edition of the Hoop Vision Weekly.

This is also a special week for the Weekly, for a few reasons:

  1. We’re celebrating one week of Hoop Vision PLUS (HV+ for short), our new premium subscription product for the 2019-20 season.

  2. It’s October, which means we’re in the month where real, officiated, televised college basketball games (albeit exhibitions, this month) will be played. Our long offseason is coming to its end.

  3. Amidst the buzz of HV+ and a new season, we have LOTS of new subscribers reading this email. Glad to have you — please keep all hands and feet inside the ride.

In this week’s edition, we have some extended excerpts from Week 1 of HV+ — deep dives on Virginia’s offense, Texas Tech’s defense, and the virtues of a “hot start” for coaches at new programs.

For those who were with us last year, we also have a similar sight: a roundup of Hoop Vision content from the week across platforms, along with some notes, observations and favorite reads from around the sport and around the web this week.

Next week, we’ll start turning a more critical eye toward this season’s early slate.

Featured HV+ Excerpt

On Tuesday, HV+ examined a few questions which seemingly never fade from the college basketball discussion at large:

  • Is a bad program always bad? Or can a coach find a way to turn it around?

  • Which coaches have “turned around” their programs in recent memory?

  • Is a turnaround permanent? Or do programs always revert to the mean?

In fact, this “turnaround” idea is one we explored a bit in one the original editions of the Hoop Vision Weekly, back in early January.

Nearly a year later, we have more concrete data on the 2018-19 season, along with a more developed idea of just what we should be looking at. For a sneak peek at what to expect from Hoop Vision PLUS, here’s an extended excerpt on that concept:

(Full HV+ post: “Measuring Coaching Performance”)

…last year we debuted a very simple concept for quantifying coaching performance. The formula for the idea goes like this:

[Adjusted efficiency taken from the invaluable kenpom.com]

(Season Adj Efficiency) - (Program Average Adj Efficiency) = Coach Performance

To explain this with an actual example, let’s use Chris Beard. In the KenPom era (now all the way back to 1997), Texas Tech has been 9.2 points per 100 possessions better than the average D1 team. Last season, Texas Tech was 30.0 points per 100 possessions better than the average D1 team. So the formula would look like this:

(30.0 pts per 100) - (9.2 pts per 100) = +20.8

In other words, Chris Beard performed 20.8 points per 100 possessions better than the historical standards of Texas Tech.

That puts him in the top five for best single season performances this century:

  1. Keno Davis (+22.4) - Drake 2008

  2. Chris Beard (+20.8) - Texas Tech 2019

  3. Jeff Bzdelik (+20.6) - Air Force 2007

  4. Casey Alexander (+20.2) - Lipscomb 2019

  5. Tony Bennett (+19.8) - Washington State 2008

Add all of a coach’s single season performances together and you have an attempt at “career coaching performance”.

The most obvious flaw in this method is for a coach that has spent decades at a single school. For example, Mike Krzyzewski’s career coaching performance is (by definition) zero. Because Krzyzewski has been head coach of Duke since 1980, his performance is inseparable from Duke’s history in the KenPom era. For the Krzyzewskis, Boeheims, and Fews of the world, their long-term program building isn’t being captured by this (admittedly simple) metric.

Where the metric does work is with newer coaches….


Just like we did for Chris Beard in the previous section, we can calculate coaching performance for all head coaches. To qualify, the coach must have started his head coaching career after the year 2000 and coached in all four seasons directly following the initial hire.

389 total coaches qualified. The top 10 coaches are displayed in the table below.

Two coaches that just finished their fourth year this past season made the list: Chris Beard and Steve Forbes. There are three more coaches from that same class (started D1 careers in 2016) that have a current rating above five points per 100 possessions: Nate Oats, Eric Musselman, and Mark Pope.

Purchase an HV+ subscription now (button below) then read the full post, including a closing section on whether early success is an accurate predictor of long-term excellence.

What else you missed this week on HV+

Monday: Virginia & Continuity Ball Screen
”Deconstructing Virginia’s Offense”

1) There’s not much room for improvisation in Virginia’s offense — especially for big men

Tony Bennett teams simply thrive on elite execution. The opponent knows exactly what’s coming, but Virginia’s precision and attention to detail are what makes them so difficult to play against.

2) The most efficient options for big men last season were largely thanks to Kyle Guy and Ty Jerome

Kyle Guy running off a pin down being set by Mamadi Diakite is much trickier to defend than Kihei Clark running off a pin down being set by Mamadi Diakite. Guy’s shooting ability is what helped set up the potential “pin & pocket” play — where Diakite scored 20 points on 77% eFG%.

With Guy and Jerome now in the NBA, Casey Morsell and Tomas Woldetensae’s ability to shoot on the move (and pull defenders towards them) will be crucial to Virginia’s success running Blocker-Mover.

Tuesday: Coaching Analytics
Measuring Coaching Performance (excerpt in section above)

Wednesday: Texas Tech, Part I
"How Do You Score Against Texas Tech?”

The big question that’s been on my mind in the off-season (which I also asked Texas associate head coach Luke Yaklich in a recent episode of Solving Basketball) about Texas Tech’s defense is simply:

Should you just take what they are giving you and drive baseline, or resist the urge?

In an attempt to inject some data into the conversation, we watched every spot-up three-pointer allowed by Texas Tech last season and charted the root cause.

Theoretically, middle drives are the kryptonite of the Texas Tech defense. The scheme isn’t built to effectively handle the ball getting into the middle of the paint. Middle drives lead to Red Raiders being forced to help one pass away — which leads to catch-and-shoot threes.

The problem for opponents, however, is that they can’t get into the middle in the first place. Texas Tech allowed just 1.8 points per game on threes generated via middle drives.

So maybe my initial question is a bit misguided. Regardless of if an opponent wants to drive baseline, Texas Tech is going to force them to do just that.

Thursday: Texas Tech, Part II
"Should More Teams Drive Baseline?”

The no-middle component of Texas Tech’s defense has received most of the love (I’m certainly guilty of that), but the switching component is equally as important. Switching — and taking away an offense’s initial action — is another crucial way of reducing randomness. A typical Texas Tech defensive possession went something like this:

  • The offense runs some type of motion or ball screen action…Texas Tech switches

  • The offense spends 10 seconds with tunnel vision trying to post-up the mismatch from the switch…Texas Tech doesn’t let the ball in

  • The offense is forced to drive baseline with less than 10 seconds on the shot clock…Texas Tech performs “help-drop-zone” rotation

The switching aligned perfectly with the force baseline and rotate style. Inevitably, Texas Tech would stall out ball screen and motion actions. And inevitably, opponents would be forced to drive baseline in late clock situations.

The Weekly Roundup: Favorites from Hoop Vision and Beyond

If you’re new here, take a look back at our Summer Review edition, sent a month ago. It links out to every HV Weekly of the summer, along with a quick breakdown of each.

Some other snacks from the week:

1) Video look at Wichita State’s UCLA series:

2) Really well-done feature by Andy Wittry of Stadium, getting some of the Big Ten’s top returning players to draw up a play they’d run with a game on the line.

The scenario presented to 11 stars on Big Ten Media Day?

There are five seconds left in the game. Your team is down by one point and you have the ball under the opponent’s basket, setting up a final baseline out-of-bounds play in a do-or-die situation.

What play are you drawing up?

Click/tap in for the flashy names and fun concept, stay for the look inside players’ minds and the not-awful play diagramming ability of said players.

3) Want more UVA? Well, we’ve got a Jay Huff breakdown, as he looks to step into a leading role for the defending national champs:

4) An intriguing research paper on the use of — and strategy surrounding — timeouts at the Division I level from data scientist and college hoops Twitter fixture Luke Benz. The full paper linked above is a quick and meaningful read, but his main takeaways can be found below:

5) Bart Torvik — a past guest on the Solving Basketball podcast — has a (somewhat?) surprising team atop his preseason “T-Rank” projections.

That’s it for this week!

Were you forwarded this email by a friend, colleague or coach? If you enjoyed it and would like to receive original research, insider access, and strategic analysis of college basketball on a regular basis, please tap/click the button below to subscribe to Hoop Vision Weekly and/or Hoop Vision Plus.