Nicholas_T, Creative Commons
Something is happening here but you don't know what it is
Way back in the winter of 2011, the Hornets were dead in the water. No owner, no lease deal, no players of real import, with Chris Paul adamant that he wanted to be a Knick or a Laker, or, wait, why not, a Clipper. The following months brought much discussion here on the best ways back to contention - tank for the lottery? build a "winning culture"? re-sign Aaron Gray to relive the glory days? - that proved, at times, quite contentious.
One of the things I was most interested in then was defining what "contention" truly meant. People call teams things like "contenders" or "title favorites" or "darkhorses" all the time, but it's all very vague. And it's confounding because if the Luis Scola-Lamar Odom-Goran Dragic Hornets would have been "pretty good" while the Eric Gordon Hornets were "terrible", it's tough to have a real discussion. So last December, I laid out a method of assessing team quality a bit more scientifically - through offensive and defensive efficiency differentials.
Very simply: offensive efficiency is simply points scored per 100 possessions, and defensive efficiency is points allowed per 100 possessions. What's a possession? Dean Oliver defined it in Basketball on Paper as starting and ending whenever the ball switches hands - on turnovers, defensive rebounds, made free throws and field goals. Anything else is part of the same possession, and so that includes things like off-ball fouls and offensive rebounds.
The simplest way to think about it: every time a team crosses halfcourt, a new possession has begun. The average team uses around 92 possessions a game, the average offensive efficiency last season was 104.6 (points scored per 100 possessions), and by definition, since all point scored by one team are also points allowed by another team, the average defensive efficiency last season was also 104.6 (points allowed per 100 possessions).
Offensive efficiency differential, then, is just the difference between a team's offensive efficiency (points per 100 possessions) and the league average for the season. If a team scored 110 points per 100 possessions (as Oklahoma City did in 2012) and the league average was 105 per 100 possessions, the team offensive efficiency differential would be +5. The same goes on defense, allowing for the fact that a lower number is a positive on that end of the floor. So a team allowing 100 points per 100 possessions (like Miami in 2012) playing in a league with an average of 105 points per 100 possessions would be a +5 on defense. Add offense and defense together, and you get the total team efficiency differential.
And that's where the notion of "+6" enters the picture. Here are the total (offense + defense) efficiency differentials of the past 20 NBA finalists:
Miami Heat (+6.4)
Oklahoma City Thunder (+6.6)
Dallas Mavericks (+4.7)
Miami Heat (+8.2)
Los Angeles Lakers (+5.1)
Boston Celtics (+3.9)
Los Angeles Lakers (+8.1)
Orlando Magic (+7.3)
Boston Celtics (+11.3)
Los Angeles Lakers (+7.5)
San Antonio Spurs (+9.3)
Cleveland Cavaliers (+4.2)
Miami Heat (+4.2)
Dallas Mavericks (+6.8)
San Antonio Spurs (+8.7)
Detroit Pistons (+4.4)
Detroit Pistons (+6.6)
Los Angeles Lakers (+4.2)
San Antonio Spurs (+5.9)
New Jersey Nets (+5.7)
On average, Finals teams have a total efficiency differential of about +6. A +6 differential isn't a guarantee a team will make the Finals; plenty of great teams have fallen short. On the flip side, a smaller differential doesn't mean a team can't make the Finals. If you take a look at the list above, teams in the high +3s and low +4s have made it, though the common thread through each of them was an experienced core that had had higher efficiency differentials in the past. Essentially, teams like the the '09 Celtics knew how to "turn it back on" come the playoffs. After all, these total efficiency differentials are simply a measure of regular season performance.
Another important takeaway is that how you arrive at +6 - elite offense with average defense, elite defense with average offense, or a good balance between the two - really doesn't matter all that much. The teams in the above list serve as evidence. Last year's Thunder were very offensively skewed, collecting about +5 of their net +6 via offense, and just +1 via defense. The Heat, meanwhile, were better defensively, going about +2 on offense and +4 on defense.
And so, at the very beginnings of the Anthony Davis era, that is now the goal the team shoots for. Last year? New Orleans was -2.7 on offense and -0.5 on defense for a net -3.2. There's a lot of work yet to be done.
One final note before I put away the numbers for the day: maybe the best part of analyzing teams through efficiency differential is that the same numbers can be broken down to an individual level. We can calculate just how much Xavier Henry or Jason Smith or Austin Rivers are contributing to the +6 cause as the season progresses. An individual player's offensive contribution is simply the player's Offensive Rating (more here) differential (player's Offensive Rating minus league average points per 100 possessions, as described above) multipled by the percentage of his team's possessions the player used.
Star players use around 20% to 25% of total team possessions in a given season. So if a player produced 120 points per 100 possessions (his Offensive Rating) while using 15% of his team's possessions, and if league average efficiency was 105 points per 100 possessions, the player would have contributed (120-105)*15% = +2.25 to his team's offensive efficiency differential. And so forth.
Last year, Jarrett Jack had an Offensive Rating of 109, the league average was 104.6, and Jack used about 13% of the team's total possessions, meaning he contributed (109-104.6)*13%, or approximately a +0.57 differential to the offense. LeBron James? Offensive Rating of 118, 27% of team possessions used, resulting in a ridiculous +3.57 differential. In other words, LeBron James' offense alone takes a team more than halfway towards title contention. One player on one side of the court! In related news, LeBron James is insane.
Defense is a lot more complicated to break down to the individual level. I very much distrust stats like 82games' "counterpart production" data as well as Synergy's points allowed/possession numbers for bigs (who are, in many cases, forced to eat the mistakes of their perimeter teammates based on the way Synergy catalogs video). Multi-year regularized adjusted plus-minus is the way to go defensively if we want things on an individual basis, but that data is generally hard to find regularly updated. (If anyone's interested, Jeremias Engelmann has 10-year regularized adjusted plus minus numbers from 2011 for offense and defense which are fascinating).
The larger point is that individual efense is a conversation for another day or month or year or whatever it ends up being. As we take a closer look at the Hornets from an efficiency differential standpoint, we'll break down the players individually on offense but look at defense more holistically. In a sense, great defense is far more system-based than offense is regardless.
And that, in a nutshell, is +6 revisited. It'll form the backbone of my NBA-wide team previews here over the next couple weeks as well as inform our final Hornets preview before the season finally begins. Thoughts?