Looking for positive lineup combinations
by Jonathan Macri
Every year I’ve been teaching, my wife and I pick one show to binge during the summer. This year’s choice was The Office.
(Three years ago it was Veep, two years ago it was Thrones, and last year it was Mad Men. Breaking Bad is on deck for next year, but I’m up for suggestions. We’ve watched almost nothing else from the last 12 years, so the choices are plentiful. Tweet away.)
In general, I think there’s lots of room for cross-pollination between Dunder Mifflin and the Knicks (Would Michael Scott have performed better as GM than Isiah Thomas would have as Regional Manager? Does Thomas sign Kevin Malone to a five-year, $30 million extension after a particularly productive week? Would Michael have called a team meeting the minute things began to bubble up between Larry Brown and Stephon Marbury? We need answers…), but I recently thought of one episode in particular when I was struggling with research for this week’s newsletter.
“New Leads,” from season six, features Michael accidentally throwing out several important sales leads which then inevitably end up in the local landfill. We later find Michael and Dwight digging through piles of garbage in an effort to find the leads - which are, to be clear, single sheets of paper amongst literal miles of refuse - which then prompts Dwight to insult Michael by saying “I hitched my wagon to a horse with no legs” (I could see Mike D’Antoni having said the same thing to conclude his exit interview).
This is the episode that came to mind as I attempted find meaningful lineup data from last year’s 17-win boondoggle. I was trying to mine said data in an attempt to write about potentially strong player combos for the year ahead. At one point in my process, I thought Michael and Dwight had better odds of achieving their goal.
In the Knicks’ defense (famous last words) last season’s stench was at least partially by design. Facing the prospect of a holdover year until the summer where they could try to mold the roster as they wished, Scott Perry and Steve Mills attempted to turn other franchises’ trash into their own personal treasure. That led to the likes of Enes Kanter, Emmanuel Mudiay, Mario Hezonja and Noah Vonleh receiving meaningful playing time.
This wasn’t all bad. We had Vonleh’s first two months, Hezonja’s last two weeks, Kanter’s nipples, and Mudiay’s…shit, who am I kidding. Other than a few blips on the radar, any positive play was so vastly engulfed by mounds of rotting trash that trying to parse out combinations that worked is harder than finding those misplaced leads.
The good news is that I’m a glutton for punishment, and after much time beating my head against the wall, I was able to see a few cracks eventually. Any positive player combinations obviously only include guys who were here last year, and three fifths of the roster is brand new, so even if something worked, there’s at least a chance it’ll be upended by something better.
Also, judge accordingly. Most teams feature several lineup combinations that have massive scoring margins in small sample sizes; for the Knicks, anything above an even net rating (the same number of points scored per 100 possessions as points allowed) counts as a minor miracle.
Almost all numbers are courtesy of Cleaning the Glass (if you aren’t subscribing, what the hell are you doing), which filters out garbage time, of which there was a lot, so this is quite helpful.
Without further ado…
Mitchell Robinson, Damyean Dotson at SF, no Kevin Knox: +8.5, 462 possessions
Why it matters: For any combination of last year’s Knicks to post this high a net rating in this significant a sample size is meaningful.
What it tells us: Aside from the fact that Mitchell Robinson is really good and Kevin Knox was really bad last season, it lets us know that Damyean Dotson is a lot more effective when he can play up a position. This makes sense in part because he’s an elite rebounding wing, but also because for as much as he can get bullied by bigger assignments at times, this pales in comparison to his struggles chasing smaller shooters off of and around screens. On the offensive end, his floor spacing is more of a difference maker at the three than it was at the two, and although Knox should improve from deep (and Marcus Morris is a good shooter in his own right), so should Dotson.
When we’ll see it: For Dot to see time at SF, even in small sample sizes, Morris would have to spend some time at the four, which may or may not be feasible depending on Taj Gibson’s role in the active rotation. Perhaps this is a back pocket lineup move for Fiz late in the year if and when Morris gets dealt to a contender.
Frank Ntilikina, Damyean Dotson & Kevin Knox: +7.9, 256 possessions
Why it matters: Take my analogy about finding the leads in the dump, and now light the dump on fire…that’s how hard it is to find positive lineup data for Kevin Knox last year.
What it tells us: Kevin Knox played almost all of the minutes in the above combination at the four, which is probably his best position long term. That said, he was equally putrid from both forward spots last year (-15.3 at SF, -13.5 at PF), so it’s not like putting him there unlocked some magic spell (him at the four with Mudiay, Dot, Hezonja and DJ was a “I have to laugh to keep from crying” -28.0 in 203 possessions). He’s beefed up a bit this year, but opponents rebounded 30 percent of their own misses and got to the line 23.4 percent of the time with him at the four last season, which ranked in the 6thand 14th percentile league-wide, respectively. These numbers will once again be bad.
With the Frank/Dot combo on the floor with Knox, the Knicks actually did even worse in those two areas, but more than made up for it thanks to the improved perimeter defense (opponents’ effective field goal percentage during these possessions was a mere 48.2, in the 97th percentile) and potent offense.
This last part is interesting; the Knicks often couldn’t score worth a damn with Knox at PF, but most of those minutes came with point guards who, umm…maybe weren’t so good with the whole passing thing. For whatever reason, Frank unlocked something in Knox on the offensive end. In 682 possessions Knox played with Frank, the Knicks had a 109.4 offensive rating – just under league average. When Knox played without Frank? 101.6 – good for 4th percentile league-wide (or, in Office terms, the face Michael makes when he finds out he must undergo six hours of counseling with Toby).
When we’ll see it: In my dreams? Ha! Kidding…(not really)…Umm, I suppose there’s a realistic chance this trio is among the first five off the bench on most nights if David Fizdale doesn’t take my advice and starts RJ Barrett. Frank, Dot, Knox, Payton and Portis is a bizzaro-world combo that features more shooting at the 3, 4 & 5 than the 1 & 2, but hey, who knows? It just might work.
Mitchell Robinson & Lance Thomas: 107.6 defensive rating, 599 possessions
Why it matters: 1) it gives me an excuse to mention Lance Thomas in this column, and 2) even though Lance is no longer on the team, it teaches us something about an ideal frontcourt pairing for Mitch.
What it means: Lance caught a lot of shit for making it seem like the offense was playing 4 on 5 when he was on the floor (and there’s some validity to that…for instance, this pairing netted an abysmal 101.2 offensive rating). He was, however, the closest thing this team had to smart, savvy defender in the frontcourt. This defensive rating, which would have been good enough for 8th in the league over a full season, helps confirm that. Robinson was fairly raw last season, and will have more growing pains in the year ahead. Giving him a frontcourt partner who thinks the game like Thomas did makes sense.
When (and with whom) we’ll see it: Taj Gibson would seem to most closely match this description, but he’s not the spryest sort at this age. The Wolves – who, granted, were a train wreck on defense – sported a 113.4 defensive rating when Gibson was on the court last year. Most of those minutes came with Towns (who is very much not of Robinson’s caliber as a rim protector) so maybe this could work. Personally, I’d like to see how he does with Marcus Morris, but at PF, not at SF.
Dennis Smith Jr. & Allonzo Trier: +1.7, 311 possessions
Why it matters: David Fizdale is the coach, and based on a year’s worth of evidence, DSJ and Iso Zo are Fiz types to a “T,” so it’s interesting that they played pretty well on the floor together, even if this is small sample size theater.
What it tells us: Maybe the Fizdale ethos (“Be…aggressive…be, be aggressive!”) isn’t as silly as us Frankophones make it out to be. I even defended Fizdale’s seemingly simple offensive philosophy many times last year, because in theory, with the right personnel, constantly attacking the basket and getting to the line should yield positive results. The Knicks (are you sitting down? You should sit down for this…) did not have the right personnel last year, and yet they still finished eighth in free throw attempt rate after being 28th in the league the previous two seasons. This obviously wasn’t enough to prop up their offense by itself, but you could see the vague outline of something that might one day be coherent with better players.
With Smith Jr. and Trier on the floor together, sure enough, New York’s free throw rate was 23 percent, which was in the 86th percentile of all 5-man units across the NBA. They also took care of the ball (12.2 TOV%, 90th percentile), which is unsurprising given this is a strength of both guards. The best part? Units featuring these two scored effectively (113.2 offensive rating) without shooting it all that well (just a 51.8 eFG%).
When we’ll see it: A lot, if I had to guess. Fiz loves Trier, and despite the noise that Elfrid Payton will get a chance to earn the starting job, I’d still be shocked if Smith didn’t have a prominent role on this team.
Best bet: these two are deployed with Mitchell Robinson more often than not. In an even smaller sample size of 177 possession, the Knicks ran opponents off the floor when the two ball-dominant guards played with Mitch to the tune of 121.5 points per possession. They rebounded over 30 percent of their own misses during these minutes – an insane number but not that crazy when you consider the blueprint: let Smith Jr. and Trier get a defense on its heels, and have Robinson clean up the misses when they’re still off balance.
That said, who would I be if I didn’t’ end with…
Frank Ntilikina & Damyean Dotson: +0.3, 943 possessions
Why it matters: the fact that two players on this team last year could see nearly 500 minutes of court time together and outscore opponents during that time is astounding.
What it tells us: Sometimes guys just work well together. Neither of these dudes have been anywhere close to complete players, or good players, for that matter. Frank couldn’t hit a shot to save his life, was tentative more often than not, and his defense plateaued. Dot’s defense is bipolar at best, and while he improved by leaps and bounds as a pick and roll ball-handler, he’s not in any way a creator on offense. Units featuring both of them should not have been able to score effectively…and they didn’t, netting just 107.5 points per 100 possessions. But they made up for it on the other end, limiting teams to 107.2. With Mitch, that number dropped to 101.4 (with Kanter it was 110.1. Ha. Oh, and they scored 97.1. Kanter’s awesome.)
Point is, something just clicks with these two, and the Knicks should be careful about throwing that away.
When we’ll see it: Like I said earlier, maybe they come off the bench together. At this point, I just want the organization to show a commitment to the combo even if it may not be in line with how they want this roster to look and feel. Good teams fit in good players all the time despite not being a perfect fit. There’s a very real chance that both Frank and Dot are going to be good, and might be even better together. Let’s give it a chance.
Thanks for reading, talk to you next week!