Beyond the RPI

Jim Van Valkenburg’s creation of the Ratings Percentage Index in the fall of 1980 marked an analytic and administrative triumph. Van Valkenburg was working in an information economy of near-total deprivation, with little or no supporting data at hand beyond wins, losses, and points. Nevertheless he was given time (six months), staff, and an office roof over his head in Kansas City by Walter Byers and told to come up with a rating system that would make the NCAA tournament’s selection and seeding processes something more than a rote parroting of the AP poll.

And, after a fashion, Van Valkenburg’s RPI did exactly what it was intended to do. Part of the impetus behind creating a rating system in the first place was the possibility that the NCAA might choose to give automatic bids to only a portion of Division I.

It never came to that. Instead, the NCAA expanded the field to 52 teams in 1983, and to 64 in 1985. By then the selection committee had already made some relatively daring at-large choices that appeared to be fueled, at least in part, by the RPI. At the same time a rating system that had been created to shed badly needed light on the game’s balance of power was beginning to change how the game was scheduled. Continue reading

Tuesday Truths: “Grayson-free” edition

Welcome to Tuesday Truths, where I look at how well 120 teams in the nation’s top 10 conferences are doing against their league opponents on a per-possession basis.

If Virginia’s just pretty good on defense, what does “Virginia” even mean?


His hot shooting has helped his team overcome its defensive shortcomings. Wait, what?

Through games of January 16, conference games only
Pace: possessions per 40 minutes
PPP: points per possession   Opp. PPP: opponent PPP
EM: efficiency margin (PPP – Opp. PPP)

ACC                       W-L   Pace    PPP   Opp. PPP    EM
1.  North Carolina        5-1   76.7    1.13    0.97    +0.16
2.  Florida State         4-1   74.4    1.11    1.01    +0.10
3.  Miami                 2-2   63.9    1.06    0.96    +0.10
4.  Notre Dame            5-0   65.7    1.11    1.04    +0.07
5.  Duke                  2-3   74.2    1.13    1.06    +0.07
6.  Virginia              3-2   62.2    1.10    1.05    +0.05
7.  Louisville            3-2   68.4    1.03    0.99    +0.04
8.  Syracuse              3-3   65.9    1.13    1.11    +0.02
9.  Clemson               1-4   67.4    1.06    1.11    -0.05
10. Virginia Tech         2-3   74.4    1.07    1.13    -0.06
11. Boston College        2-3   72.5    1.02    1.09    -0.07
12. Wake Forest           1-4   71.0    1.04    1.12    -0.08
13. Pitt                  1-4   65.7    1.03    1.12    -0.09
14. Georgia Tech          3-2   70.2    0.98    1.07    -0.09
15. NC State              1-4   75.3    0.97    1.13    -0.16

AVG.                            69.9    1.06
Acceleration since 2015:        10.6%
KenPom rank: 2
% of games played: 28

An average ACC defense will allow the opposing team to score 1.06 points per possession, and thus far on the young conference season Virginia has held opponents to 1.05. This is not what we’re used to seeing from Tony Bennett’s guys. Continue reading

Forecasting the next scoring revolution


Dan D’Antoni may be on to something. (AP/Garry Jones)

Scoring is up this season, thanks in part to what can only be termed a sophomore-breakout season from the 30-second shot clock. Surely that clock will remain a fixture for the foreseeable future, and it’s therefore reasonable to assume that the scoring boost it has provided may well plateau. By next season, after all, players with a working memory of the previous clock will already be a decided minority.

So what happens to scoring now? Glad you asked….

Teams in Division I are making about 49 percent of their twos and 35 percent of their threes, meaning 100 attempts of each type of shot will net you, on average, 98 and 105 points, respectively. This seven-point margin perhaps holds the allure of a green light from the hoops gods.

All the usual caveats apply, naturally, and in particular Josh Pastner and Kim Anderson are hereby given permission to tackle any of their players that are about to attempt a trey. Not to mention the math here can be boosted in favor of our old friend the two-pointer through the simple expedient of shooting fewer jumpers inside the arc and getting more chances at the tin.

North Carolina has never shot threes, never will, and will always be hegemonic at basketball anyway. And, whether you’re speaking of D-I, a conference, a team, or a player, more three-point attempts can mean less accuracy. There are complexities intrinsic to this question, to be sure, and what follows is pitched at the level of the whole beach and not the grains of sand. Continue reading

Tuesday Truths: “Far too early” edition

Welcome to the season’s first installment of Tuesday Truths, where I look at how well 120 teams in the nation’s top 10 conferences are doing against their league opponents on a per-possession basis.

Why Jefferson matters


Not a happy tableau if you’re a Duke fan. (

Through games of January 9, conference games only
Pace: possessions per 40 minutes
PPP: points per possession   Opp. PPP: opponent PPP
EM: efficiency margin (PPP – Opp. PPP)

ACC                       W-L   Pace    PPP   Opp. PPP    EM
1.  Duke                  2-1   75.3    1.23    1.01    +0.22
2.  North Carolina        2-1   77.5    1.07    0.90    +0.17
3.  Florida State         3-0   70.8    1.14    0.98    +0.16
4.  Notre Dame            3-0   64.9    1.13    1.07    +0.06
5.  Syracuse              2-1   65.2    1.16    1.11    +0.05
6.  Virginia              2-2   61.4    1.08    1.04    +0.04
7.  Miami                 1-1   63.5    1.07    1.05    +0.02
8.  Louisville            1-2   64.4    0.97    0.97     0.00
9.  Pitt                  1-2   62.3    1.14    1.14     0.00
10. Clemson               1-2   67.5    1.09    1.10    -0.01
11. Boston College        1-2   76.3    1.07    1.11    -0.04
12. Wake Forest           1-3   70.0    1.00    1.09    -0.09
13. Virginia Tech         1-2   77.5    1.05    1.17    -0.12
14. NC State              1-2   78.4    0.95    1.13    -0.18
15. Georgia Tech          1-2   70.7    0.86    1.12    -0.26

AVG.                            69.7    1.07
Acceleration since 2015:        10.3%
KenPom rank: 2
% of games played: 17

Duke plays at Florida State tonight, and the Blue Devils will contest the issue in Tallahassee without the two guys that I, personally, would rank 1-2 in terms of importance, even above Luke Kennard or Grayson Allen. Those absent guys are: 1) Mike Krzyzewski; and 2) Amile Jefferson. Continue reading

Tuesday Truths for an accelerated world


His team is playing much faster this season, a bit like Division I since 2015.

Next week on a day that I trust requires no further specification here, Tuesday Truths will return for an eighth season.

This rite of the bleak midwinter has now appeared at three different sites and been situated amidst wildly varying levels of graphic support. Yet here it is, still standing. It is the Tubby Smith and/or Lon Kruger of college basketball features.

For 2017, Tuesday Truths will track every possession in conference play for the ACC, Big 12, Big East, Big Ten, Pac-12, SEC, American, Atlantic 10, Missouri Valley, and WCC. This is more or less the top third of Division I, and the idea is to cast a net large enough to either: a) spot a nascent national champion early; or b) offer grounds for proper and informed surprise when a dark horse shocks the world come April.

College basketball is strange in leaving approximately 35 percent of a team’s schedule purely to its coach’s discretion. I certainly don’t ignore that 35 percent — I’ve written lots of words from November up to now — but I have found in addition that there’s much to be gained from looking at the 1200 or so possessions that a team records in conference play. Continue reading

Grayson Allen seems sincere, and maybe that’s the problem


Adrenaline is not Grayson Allen’s friend. It’s not Fran McCaffery’s friend either, of course, but at least the Iowa head coach didn’t give North Dakota’s Brian Jones a leg-sweep in the handshake line on Tuesday night.

Instead, McCaffery petulantly walked off the floor rather than shake hands. Dumb move from a guy old enough to know better, certainly (McCaffery apologized almost immediately), but one that can be classed as the standard-issue wreckage caused by male adrenaline.

What’s strange about Allen (who also apologized almost immediately) is that his adrenaline swells so unerringly and with such velocity in his right foot, of all places. I’ve played and watched basketball for a really long time, and I’ve seen every kind of scuffle and altercation imaginable. Scuffles and altercations are endemic to hoops, truly.

But I can honestly say I’ve never seen a player trip an opponent intentionally twice, much less three times. The particular form that Allen’s on-floor stupidity takes is sui generis. I have no idea where it comes from. This will be for a biographer to figure out, and if Allen indeed turns out to be a seven-time NBA all-star and/or wins a presidential election someday (don’t laugh), we may get answers. Continue reading

Shot volume index


Masters of the SVI genre. (AP)

For the time being I’m taking what I’ll call a shot volume index out for a test drive to see if it can be of any use. The SVI can be thought of as the number of shot attempts a team would record in 100 offensive possessions with average shooting accuracy (determined collectively by 75 teams in major-conference play in 2016), and, most improbably, zero free throws.

I need hardly add that I’m far from the first observer to look at and measure this aspect of the game. Consider this merely one more dish at the buffet.

Last February when I was juggling Tuesday Truths and other stuff, I whipped up a little shot volume casserole in the microwave, and it was, I trust, passable. But with a bit more time to prepare, I’ve come to prefer the slow-cooked SVI and its fair degree of accuracy in predicting how many points your team will score. Also, allow me to extend a big thank you to Svi Mykhailiuk at Kansas for cheerfully loaning his name to this undertaking.

Great offensive rebounding teams that commit a turnover before they attempt a shot don’t get a chance to display their greatness. Conversely teams that excel at taking care of the ball but place a transition-D-focused ban on offensive boards see their shot volumes suffer relative to competitors with identical or even significantly higher turnover rates. The SVI proudly carries this brand of sequential flow-charting horse sense in its DNA. Continue reading