Kentucky Basketball: Insight from Average Possession Length

Ken Pomeroy made another addition to his site this week to give some cool information about teams that you can't find anywhere else. He used play-by-play data to determine the average length of an offensive and defensive possession for each team since the 2010 season and then looked at how that length was related to offense, defense, and overall tempo (possessions per game). You should read the whole thing because Ken includes a number of great graphs showing relationships between the various quantities, but here's a quick summary of the results:

• The year-to-year consistency is about the same for offensive possession length (oAPL) and tempo and both are considerable higher than defensive possession length (dAPL)
• Overall tempo is more strongly related to offensive speed than defensive speed
• Teams that draw out defensive possessions (have a high dAPL) tend to be better defensively
• There is no systematic relationship between how fast and offense plays and how fast (or slow) a defense plays.

None of those things are ground breaking discoveries of course, but that's good because it lends credibility to looking at teams this way.

With that in mind, here is what Kentucky has done the last four years. By way of comparison we'll also look at what a couple of former UK coaches are doing at their schools.

The John Wall-led Wildcats have easily had the fastest offense under Cal. No surprise there as Wall was a 1 man fast break and Eric Bledsoe wasn't far behind. The speed is especially impressive given the Cats prowess on the offensive glass that year. Led by DeMarcus Cousins and Patrick Patterson, the Cats were the 5th best offensive rebounding team that year. All those rebounds tend to increase time of possession which really emphasizes just how quick Wall and Bledsoe were on the break and how often they pushed the ball.

By contrast, the 2011 team was very deliberate on offense, particularly at the end of the season. I remember Glenn writing about this particularly as it related to the transition of the offense from playing through Terrence Jones to the Brandon Knight-Josh Harrellson pick and roll.

Defensively, Cal's teams don't stray much from average. That isn't a surprise as his defense isn't focused so much on making it harder to get a shot in the way that zone or pack-line does. Cal does press, but it usually is more of a stall press than one designed to speed up the opponent or force a lot of turnovers the way VCU's Havoc or Oliver Purnell's myriad presses do. That results in a defense that time-wise doesn't stray far from average, regardless of how quickly the offense plays.

Let's look at how Cal's style compares to Mr. Tatoo.

This is really interesting and I think it offers a perfect illustration of where time of possession is an incredibly useful tool for determining the style of an opponent. As Pomeroy mentions in his article, up until now possessions per game was the best measure we had for describing the playing style of a team, but it can be very deceiving, especially for teams that play fast on offense and force time on defense. Louisville is one such team.

In the last 4 years Louisville has ranked 170, 103, 123, 126 in pace - not exactly slow but certainly at odds with Rick Pitino's reputation and in contrast to the sense you get when watching UL play. Breaking down possessions into their offensive and defensive components helps explain this dichotomy though. The Cardinals do play fast on offense, but their pressure defense is zone-like in its ability to slow down opponents.

How about another National Championship winning, former UK coach?

Not much surprising here. Tubby Smith's squads play deliberately on offense and they are always good on the offensive glass which further increases the time they spend on offense. On the other side of the court his ball-line defense effectively slows down opponents, although I think it's interesting that it does so less than Pitino's full court pressure. Oddly, Tubby's worst season at Minnesota was also the one in which his offense played the fastest. I don't know if those two things are related or not, but I thought it was notable.

That's a look at how further separating possessions into their offensive and defensive components can help us understand how teams want to play their game. In this case it conforms to a lot of what we already know about the above coaches and that should give us some confidence when applying it in the future to other teams that we aren't as familiar with.