What’s the Best NBA Team You Can Build With Second-Round Picks?

During this month of 30 articles in 30 days here at Pace and Space, inspiration can come from some very interesting sources.

Like this delightful distraction from constant political news that hit Twitter last night from Paul Headley:

A starting lineup of Spencer Dinwiddie at point guard, Khris Middleton and Bojan Bogdanovic on the wings in the shooting guard and small forward positions, Montrezl Harrell at power forward, and Nikola Jokic at center.

That’s a lot of scoring—100.2 points per game between the five of them in the 2019-20 season.

How many games such a team would win would depend very largely upon what kind of supporting cast you gave them—if the bench is basically a G-League team, that’s different from if the bench were made up of quality role players selected in the second round.

So let’s go ahead and expand this discussion, checking Basketball Reference’s record of NBA drafts past and finding guys whose production combined with their role on their current teams would make for a plausible 6 through 13 set of guys for a 2020 NBA team.

Our major backups will be as follows:

PG: Monte Morris, Denver Nuggets (51st overall pick, 2017)
SG: Danny Green, Los Angeles Lakers (46th, 2009)
SF: Jae Crowder, Miami Heat (34th, 2012)
PF: Jerami Grant, Denver Nuggets (39th, 2014)
C: Ivica Zubac, Los Angeles Clippers (32nd, 2016)

To keep this simple and because we’re not going to worry ourselves overmuch about garbage time, just take any 3 guys who put up zero VORP or close enough thereto and that’s the roster.

You may notice there are obvious omissions here. If the goal was just to put up the best roster without worrying that they were going to complain about their minutes or their role, Malcolm Brogdon and Draymond Green would be on this team. So, in all likelihood, would guys like Joe Harris and Dwight Powell.

But the goal here is not “let’s put together 13 starters and then ask how they’d do over 82 games?” There’s also the salary cap to consider; I haven’t gone in and counted up the money here, but I’m betting we’d have a plausible route to an acceptable salary total; if not, we could move one of those guys to take on a bad contract in order to get to the salary floor.

That’s not within the scope of Paul’s original question though, so let’s examine the entire statistical pile here.

We’re going to prorate minutes based on the following assumptions:

First, the starters will be playing 34 minutes a game and the backups will be playing 14. Garbage time will remain unaccounted for because in the grand scheme of predicting 82 NBA games, it doesn’t ultimately matter.

Therefore, each position will have a Win Shares and VORP projection that is a simple sum of 70.83 percent starter, 29.17 percent reserve, prorated from each player’s minutes relative to 34 and 14, respectively (so if a bench player played 28 minutes a game over 82 games in real life, their total would be halved, for example.)

Next, we’ll apply formulae that this site has already calculated in previous articles for the line of best fit for Win Shares and VORP to equal wins over a full NBA season.

For Win Shares, that’s (WS*0.94)+3.14. For VORP, that’s 20.34+(2.07*VORP).

We’ll average the two numbers and that, at last, will be my answer to Paul’s question.

The raw data, noting that minutes per game here is per TEAM game, not per player appearance:

Dinwiddie: 4.8 WS, 1.9 VORP, 27.7 MPG. Adjusted total: 5.9 WS, 2.3 VORP.
Middleton: 7.9 WS, 2.9 VORP, 25.4 MPG. Adjusted total: 10.6 WS, 3.9 VORP.
Bogdanovic: 4.2 WS, 1.1 VORP, 28.9 MPG. Adjusted total: 4.9 WS, 1.3 VORP.
Harrell: 7.0 WS, 2.2 VORP, 24.3 MPG. Adjusted total: 9.8 WS, 3.1 VORP.
Jokic: 9.8 WS, 5.5 VORP, 32.0 MPG. Adjusted total: 10.4 WS, 5.8 VORP.

So we can already arrive at a short answer here. These five guys, even if the rest of the team contributed a complete and utter garbage fire (minus-3.1 WS and about minus-10 VORP, consistent with adding absolutely nothing of value to a team that wins even one game), would by themselves account for about 42 wins by WS and closer to 48 wins by prorated VORP.

That’s enough to drag even the worst roster imaginable to the playoffs. Even the 2012 Bobcats weren’t as bad as that theorized bench unit.

But let’s give those five guys some real teammates and see how good this gets. A note on Crowder: He split his season between Memphis and Miami; both teams played 73 games during the season so I’m just going to keep it simple and prorate his minutes per game in 65 games played out to a 73-game season.

Morris: 3.8 WS, 0.8 VORP, 22.4 MPG. Adjusted total: 2.4 WS, 0.5 VORP.
Green: 3.4 WS, 1.1 VORP, 23.8 MPG. Adjusted total: 2.0 WS, 0.6 VORP.
Crowder: 3.8 WS, 0.9 VORP, 25.7 MPG. Adjusted total: 2.1 WS, 0.5 VORP.
Grant: 4.4 WS, 0.7 VORP, 25.9 MPG. Adjusted total: 2.4 WS, 0.4 VORP.
Zubac: 6.6 WS, 1.6 VORP, 18.4 MPG. Adjusted total: 5.0 WS, 1.2 VORP.

Whoa. Zubac for Sixth Man of the Year if he plays on this team!

But here’s what we end up with for a projection on a team based entirely on second-round picks.

By Win Shares, we project a 55-27 record. In the last 82-game NBA season in 2018-19, that would’ve grabbed the 2 seed in the West and the 3 seed in the East.

By VORP, we get an incredible 61-21 record, which in 2018-19 would’ve been the best record in the league, good for home court through the Finals.

Splitting the difference, we get 58-24, the same record as the eventual champion Toronto Raptors.

And we probably could, if we were being extra-generous, eke out one more win from those guys in garbage time statistically.

Now granted, there are actual NBA game-related questions left to be answered. This team would be a disaster area on defense, but they’d probably outscore a lot of opponents as well. In other words, they’d have a lot in common with the 2020 Dallas Mavericks, a team with a historically great offense, a lousy defense, and a prorated-to-82-games 54-28 record.

So there’s your answer! 55 wins, a 2 or 3 seed, probably conference finals as a ceiling if not a tough second-round loss in 6 or 7 games.