My fellow pro-basketball fans, it’s time once again to de-mythificate some of the data we’re being subjected to by writers and talking heads. As usual it’s up to us to find the baggy shorts of truth hidden beneath the warmup suit of infotainment and misdirection that the big news outlets have snapped and zipped across our eyes (and pelvises).
Let us address a notion in the media this week about the complexity of NBA defenses and the Gordian Knot that team offenses have to untangle in order to somehow get the ball in the basket. Take this excellent article by Beck Masely or Mase Beckson… Pardon me, Beckley Mason… over at Truehoop http://espn.go.com/blog/truehoop/post/_/id/55656/its-smart-to-be-fun. Great write up, and the George Karl quote in particular is fascinating if not that surprising:
“Coaching has now gotten so technical and scientific and there’s so much of it and there’s so much video and and there are so many statistics, that basically the reality of coaching is when you play 5-on-5 basketball it’s very difficult to beat the defense and the scouting reports and the preparation and the tendencies that we know teams have. So what we’re trying to do is play before those things can be settled in to.”
Now I’m not going to tell you that George Karl is wrong. Running is his strategy, and it works great for the Nuggets with their 9 man rotation and home court advantage. But the truth of the matter is that offense is at least as dominant a force in the game today as defense is. Maybe the analytics and videography of the modern game are having a significant impact and maybe not, but if so it appears to be at the scoring end of the floor as well as on defense. My Basketball-Reference research shows the league averages of Dean Oliver’s 4 Factors most crucial to winning in basketball across the years since 1980. As you can see, in 2013 the league average effective field goal percentage or eFG% (FG% adjusted to account for the additional value of made three pointers), is 0.7 points higher than average for the full 34 year time span and is exactly average for the 8 years since the no-hand check and defensive 3 seconds rules took effect. Back in the 1990s and early 2000s before the rule changes and the improvement in scouting and analytics, the average effective field goal percentage was 1 to 2 points lower than it is today. THAT was a defensive league.
What has changed drastically in recent years has been the average offensive rebound rate and free throw rate which are both down significantly. Of the four factors, these are the least significant to winning games, but it is interesting that as offensive efficiency has increased in terms of shooting percentage, it has slipped in these other categories. This is particularly confounding because more offensive rebounds historically means more interior scoring which usually boosts shooting percentages, and more free throws means fewer missed field goal attempts which again boosts shooting percentages. We should expect to see eFG% drop off in correlation with the decrease in offensive boards and free throws, but we don’t. The opposite is true.
For eFG% to remain constant or improve while ORB% and FT rate drop off, some other factor must be at play. At first blush I assumed this was just increased three point attempts and percentages, but as I dug through the statistical muck, I hit on a more complex and probably more accurate hypothesis. I believe the perimeter power forward, the player we commonly refer to as the stretch 4 and the swing forward, has fundamentally altered the way the game is played at both ends. The best supporting evidence I have, ironically, comes from guards and driving forwards, in particular the free throw rates of playmaking perimeter stars. League-wide overall free throw rates are way, way down, 3.5% lower than average since 1980 and 5% lower than the peak. However, free throw rate for high usage perimeter players today is MUCH higher than it was for perimeter players in the 1980s, 1990s, and even the early 2000s.
Michael Jordan, who purportedly got every call in the ‘90s, and is in fact the highest rated of his generation of wing players, has a career FT rate of .297. Since his retirement 9 high usage (25% or higher) wing scorers have a FT rate over .300, and the two players most similar to MJ, Wade and Kobe, have FT rates of .370 and .327 respectively, much higher the Jordan’s. Further highlighting the difference, every single one of those current players who outranks Michael takes a significantly larger proportion of 3 point shots than Jordan did. AND Jordan has a higher career usage rate than any of the players in this comparison. If we expect fouls to be drawn principally in the paint, these players all had fewer chances to get fouled, and all got fouled at a higher rate than Jordan, the player most often associated with getting favorable “star calls.”
So does that mean that the stars of today get even more favorable calls than the G.O.A.T. did? Is this a side effect of the no hand-check enforcement? Probably more the latter than the former. I would hope anyway. But I think there’s a better explanation than rules or favoritism.
In the olden days, the long long ago, teams played two big men, and the big men both played in or near the paint. On offense this means that driving lanes were harder to come by, but offensive rebounds were easier to secure (more shots came from big men in the paint, and more big men were in the paint to fight for boards). On defense this meant that there was a tall defender taking up space in the paint even when the center was involved in a pick and roll. Today teams often only play one big man in or near the paint, which means that when a driving player gets past the first line of defense, the paint is his oyster. Err.
This is a tremendous difference and it becomes very apparent when two teams playing big line-ups meet. If no stretch fours or driving power forwards are in the game, then the rotating player on a high or side pick and roll is going to be the second big man with the weak side wing defender hedging. This means that a driving wing or guard whether coming off the initial pick and roll or as a secondary penetrator after a kick out is going to meet size at the rim and have a tough shot attempt. Think of Memphis with Gasol and ZBo or the Pacers with West and Granger (in fact most of the top defenses in the league still use the double big man line-ups such as Perkins-Ibaka, Duncan-Splitter, Boozer-Noah, and Blake-Jordan). In contrast teams that go small at the four with guys like Carmelo can’t defend the paint as successfully, and offenses that have shooting specialist big forwards like Ryan Anderson can nullify their opponents’ size because big forwards can’t leave Anderson to rotate at all.
The Miami Heat take the concept of the stretch 4 to the extreme because they don’t play a center at all. Their primary big man line-ups are various combinations of LeBron James, Chris Bosh, and Udonis Haslem. Haslem and Bosh are primarily jump shooters, and LeBron is a slashing wing who happens to be 6′ 8″ 265 pounds. The Heat’s best offensive line up is Chalmers, Allen, Wade, LeBron, and Bosh. In this line-up James is playing power forward, and generally, speaking teams will play small to cover James with athleticism. The paint is effectively empty when LeBron or Wade try to drive. Even a very disciplined defense like the Celtics can be drawn to empty the paint in the pick and roll situation. In a pick and roll where Bosh is the player setting the screen, the Heat can pull the only big man defender on the other team (Garnett) out of the paint. This means that when James or Wade gets into the lane, any rotating players will be perimeter “small” players.
But while the Heat are an extreme case, this is a league-wide trend of perimeter slashers getting good looks more often than ever before. Looking at the top 20 playmakers of the last decade by eFG%, you get an average of .511 which is 2 points higher than the total league average including big men and shooting specialists. Looking at the top 20 from the 80’s, 90’s and early 2000’s the eFG% is .485 which is just below league average for that time frame. And focusing just on 2 point shots, in 2013 the top 20 players are averaging .511 on 2 point shots. 10 years ago in 2003 the same group shot .470 on 2 point shots. 20 years ago in 1993 the top 20 perimeter slashers / play makers shot .475 on 2 point shots. What that tells us is that perimeter scorers are getting more layups and shooting less midrange shots than they did 10 or 20 years ago.
What’s interesting is that while drivers’ shooting is at an all time high, the loss of the second big men on offense and defense has severely reduced the offensive rebounding and overall free throw shooting. While perimeter scorers are putting up prettier per possession numbers than ever before, overall team efficiency is more average than excellent.
An interesting example of a team that maximizes the value of the stretch 4 line-up and the traditional double big man line-up is the Thunder. OKC’s best offensive lineup (by FAR) is Russell Westbrook, Kevin Martin, Thabo Sefalosha, Kevin Durant, and Nick Collison. Just like Miami’s best offense, they put star perimeter forward Durant at the big forward and masquerade power forward Collison at the center. OKC’s best defensive unit is the starting five of Westbrook, Sefalosha, Durant, Ibaka, and Perkins. Both of these units are elite in the league today.
So is the stretch 4 changing the game? It looks like it is, and for fans who like to watch slashing perimeter players like LeBron James, Dwyane Wade, Russell Westbrook, and James Harden get to the lane and throw it down unopposed or over undersized backline rotations, that’s a very good thing.
Tags: advanced stats NBA, Chris Bosh, dean oliver, Dwyane Wade, fastbreak, field goal percentage, George Karl, gordian knot, home court advantage, kobe bryant, lebron james, Miami Heat best offensive unit, Michael Jordan, NBA, Oklahoma City Thunder best defensive unit, Oklahoma City Thunder best offensive unit, Power forward, run and gun, Russell Westbrook, sports, stretch four