Stat Stats: Joe Johnson’s New Year’s Resolution

Stat, from the latin statim for “immediately”, Stats, as in statistics.

The impetus for today’s “Stat Stats” comes from Gordon’s comment:

Great stuff. Just throwing it out there, but I’d love to see an in-depth breakdown of how a guy like Joe Johnson would fit with this team.

You got it, Gordon.

Joe Johnson’s on the back 9, and his 2015-2016 stats are the weakest of his 14-year career.  For the first time since his rookie season, he’s posting a negative offensive Box Plus Minus (which uses box score stats to model plus-minus.  Not perfect, but a cool up-and-coming stat).  His PER is at a career low but some of that is due to his career-low usage %.  That second point isn’t necessarily a bad thing, as the Cavs don’t need another high usage player alongside LeBron (6th in USG%) and Kyrie (21st in USG%).  Kevin Love’s usage dropped by more than 20% after joining the Cavaliers.  So, as a role-player, how would Joe Johnson benefit the Cavs?  I’ll remain an optimist for the rest of this post, but not before reminding everyone that Joe Johnson, despite his LeBron-like frame, is  not now, and has never been, a great defensive player.  Is he better than Richard Jefferson on D?  Not sure.  Johnson posted a 2-year Defensive RAPM of -0.63 from 2013-last season.  So, below average, but not terrible.  During that same timeframe, Richard Jefferson posted a -0.57.  So if we assume they are about the same defensively, let’s put on our optimist cap and look at the ways Joe Johnson could greatly benefit the Cavs at the offensive end.  The key, is to ignore everything Johnson did before the New Year.  Let’s see what 2016-Johnson can do for the Cavs.

Since the ball dropped, Joe Johnson has taken exactly 100 3-pointers.  How’s that for nice round number?  When I tell you that he’s shooting 46 per cent from downtown, I literally mean he’s made 46 out of 100 threes.  That’s awesome.  Less than 15% of those 100 threes have been unassisted, meaning, he’s evolved into a deadly catch-and-shoot 3-point shooter.

It’s worth pointing out that, during this same time-split (2016 only), Richard Jefferson has made 19 of 41 3s (or, 46.3%).  Like JJ, less than 15% of those have been unassisted.  On a per-minute basis, these guys are almost identical in their 3-point shooting since January 1st:

JJ: 46 triples made in 826 minutes (or, a made 3-pointer every 18 minutes of action)
RJ: 19 triples made in 323 minutes (or, a made 3-pointer every 17 minutes of action)

(Of course, that Johnson has shot the same percentage despite attempting more than twice as many is more desirable.)

https://www.youtube.com/watch?v=z2gcxeku56Y

Johnson has dished out 109 assists to just 55 turnovers (actually the worst A/To ratio of his career), so that jives with his reputation as being a decent passer.  This playmaking ability as well as JJ’s mid-range game is what separates him from Richard Jefferson.  Jefferson has been, since Jan 1st, almost exclusively a 3-point sniper and fast-break finisher.  Johnson has done a lot more playmaking and only about a third of his shot attempts are 3-point attempts, unlike Jefferson who has taken more 3s (19) than 2s (14) since January 1st.

So a good way to think about Joe Johnson, based on the numbers, is that he embodies Richard Jefferson’s spot-up 3-point shooting but, additionally, he’s a more dynamic offensive threat and he creates more opportunities for his teammates.  He’s shooting 50% from the field on 2-point attempts which is outstanding for a wing that is the primary offensive perimeter threat on his team.  Unlike Jefferson who almost never takes mid-range shots, Johnson is still pretty effective from anywhere on the court. While historically more of a mid-range player anyway,  Johnson has seen a precipitous decline in the number of shots he’s attempted at the rim this season (now down to 6%).

The Cavs play such a LeBron-centric style.  Richard Jefferson is really the kind of guy suited to play with LeBron James, not in his stead.  Part of the reason for this is that it’s easier to recover against more one-dimensional offensive players when LeBron is not drawing all that attention.  This is where Joe Johnson is arguably a better LeBron substitute, while still being as statistically compatible with playing alongside LeBron (the highly accurate spot-up 3 point shooter).  Here are some figures that lend themselves to this argument. These stats courtesy of NBAWOWY.com

Jefferson at the SF with LeBron on the Bench:
Cavs taking 40% of their shots from mid-range and only converting 39% of them.

LeBron at the SF with Jefferson on the Bench:
Cavs are taking 35% of their shots from mid-range and converting 43% of them.

Joe Johnson on the Bench:
Nets are taking 46% of their shots from mid-range and only converting 38% of them.

Joe Johnson at the SF:
Nets are taking 46% of their shots from mid-range and are converting 43% of them.

‘Mid-range effectiveness…’ OK. Ah…let’s back up. k? And let me explain to you how the modern NBA works.

If those stats don’t mean anything to you yet, consider this: LeBron James’ percentage of mid-range shots (26%) is more than twice that of Richard Jefferson’s (12%).  So why would lineups featuring RJ but not LeBron lead to MORE mid-range jumpers as a percentage of team shots?  Doesn’t make a lot of sense, right?  Well, my theory is that it’s the same reason the LeBron 1.0 era always went so badly whenever LeBron sat or missed games.

I remember those days…

The team was/is built around LeBron’s unique skillset.  He draws in defenders and the roster put in place around him is meant to catch and shoot, not create their own offense from scratch.  (Kyrie Irving is the lone, and oftentimes necessary, exception.) When those more one-dimensional offensive players are left to their own devices, efficient offense is harder to come by, and guys that are not as comfortable creating something from nothing are forced into late-clock mid-range Js after they’ve been run off the comforts of their 3-point line.  Thus, the much lower FG% in those cases.

If these mechanics I’m describing are true, a guy like Joe Johnson is a very solid LeBron replacement for those other role players, perhaps even Kevin Love. (Dare I call him a ‘role player’ at this point?  Maybe ‘X-Factor’ is better…)  Look again at the on/off numbers I posted above.  The Nets take the same number of mid-range shots with and without JJ but they increase in accuracy by five percentage points with him.  That’s not trivial.  Johnson is comfortable creating something from nothing and he’s still pretty good at it.  When LeBron sits, the overall offensive execution and role-player effectiveness should increase with Joe Johnson at the SF position instead of Richard Jefferson.  Not only will Johnson draw the attention when he faces up that LeBron does, (not to that extent, of course) but in the event that the offense stalls out, he’s a much better guy to be taking those “we gotta get some kind of shot off” shot than anyone else sans Kyrie when LeBron is sitting.

Of course, in the playoffs, LeBron’s not going to sit very often in tight games.  He averaged over 45 minutes a game in the Finals.  That’s why it’s so critical that Johnson has been such an effective spot-up 3-point shooter since January 1st.  The LeBron/Johnson fit should be not too dissimilar on offense than lineups that feature LeBron and Richard Jefferson (interestingly, those lineups draw the most fouls of any 2-man combination on the Cavs)

Conclusion:  Joe Johnson’s New Year’s Resolution to shoot 48/46/92 would give the Cavs an adequate replacement for Richard Jefferson’s (also very effective) spot-up 3-point shooting.  But his solid mid-range game and ability to create for teammates makes him a huge upgrade as a LeBron replacement during the 5-10 minutes a night when LBJ needs a breather.

For some other analysis on how Joe Johnson could fit the Cavaliers, I strongly recommend today’s “Hey Windy” podcast.  He talks about how Johnson could help the Cavs in a small-ball lineup against the Warriors “lineup of Death”.