By the Numbers: How Valuable Have Timberwolves Draft Picks Been?

Dec 3, 2022 - 5:16 PM

Minnesota Timberwolves v Detroit Pistons" src="https://cdn.vox-cdn.com/thumbor/08yVxD3WOQinYDGMBsIw-xsCICE=/0x0:4928x2772/1920x1080/cdn.vox-cdn.com/uploads/chorus_image/image/71705287/1238169889.0.jpg" />

Last time, I came here and said I was going to prove Bill Simmons wrong. I said that there was no reason that one team’s picks should be inherently more valuable just because of the logo on the front of the jersey. Past failures should not prohibit to future success, because if it was we’d live in a retro-futurism world of basketball Gattaca.

While I exaggerate a little, what Simmons is essentially saying is that bad teams stay bad, which isn’t completely antithetical to what the league preaches, but let’s see how that applies to draft picks.

Part 1: The Math

How are we calculating draft pick value? Well, I wanted to use a metric that allowed me to measure the expected performance of picks as well as their actual performance. This led me to two stats initially: WAR (wins over replacement) and VORP (value over replacement player). In other words, how much better are you than a decent starter? While this was a good initial idea, it led to some issues.

First, VORP and WAR are answering similar questions, but use drastically different equations and as such need standards to be compared to each other, so I’d have to choose one. Second, both VORP and WAR are measured cumulatively, and not by year, and as such give older players a lot more credit than younger ones. Why, I hear you asking, don’t you average those stats by years played and find your yearly answer that way? I did that, it resulted in Ben Simmons being the 13th most valuable player drafted since 2010 and left Andre Drummond top-30, which is not a conclusion I will accept (someone once said that data science is just shifting stats around until you get to where you want. That man was my statistics teacher. Thanks Joe!).

So what is there left to do? Well, let’s start by finding a good scale to find simple draft pick value. I used BPM (box plus minus). Now BPM has its own issues, most notably that it fails to account for role. Good players on winning teams will get higher scores than great players on bad teams. You want to know why I’m using it though? Because BPM gives us this benchmarks:

A BPM of 6.0 is All-NBA level. A BPM of 4.0 is All-Star level. A BPM of 2.0 is a good starter, while 0.0 is a good bench player, and -2.0 is a fringe bench player (Lazar Hayward, eat your heart out).

Using that, I can spit out this equation: 8*0.94^(Draft Pick Number-1)-2. I can explain all the intricacy that went into it, but here’s a graph to show you what you need to know.

That was a whole lot of words to say, here are my expectations for picks numerically. The first three picks should be All NBA caliber. Four through 10 should be All-Star or very good starters. Late lottery through early 20s should be good role players, starters or bench players. And lastly, the last five picks of the draft are wild cards. Expectations are near zero, go out there and find a gem.

Now that we have all of that set up, let’s get to some actual content.

Part 2: The Analysis (My Own Worst Enemy)

So before I say anything whatsoever, let’s start with some graphs. This is pick value generated, which takes the above graph and applies it to whatever pick this team would have had if trades didn’t exist (Kahn training wheels).

As you can see, losing teams generate more value. The Wolves made the playoffs twice since 2010. The Sacramento Kings did not make it at all. My favorite part of this is the San Antonio Spurs, who generated a mere 5.23 BPM worth of expected value in 12 years, which would be beaten out by the Magic’s singular attainment of the No. 1 overall pick this past year, which they used to select Duke’s Paolo Banchero. But, this is nothing new. Losing teams generate high picks. You want to see pick value generation next to losing percentage, here we go.

Now that we’ve seen enough shiny charts to know that losing begets high picks, let’s talk about this. There is no reason why certain team’s picks are worth more that has anything to do with the team’s name or location or whatever. Picks are about the team. The Cleveland Cavaliers are top five in this statistic. They also won a title in the middle of the stretch and have one of the brightest futures in the league after trading away multiple firsts. Whats more, they are literally the team that caused the Stepien Rule to be enacted because they traded so many picks. It’s simply a lazy and simple take to make fans like me mad enough to build 600x30 celled spreadsheets. Here we are doing exactly that, and, maddeningly, giving him evidence to use.

Part 3: Is it Time to Go Home Yet?

We know that gaining draft picks isn’t the only path to success. We know from last time that the Minnesota Timberwolves don’t have the best track records with trading their picks, and we know from the eternal memory of the Joe Smith disaster that they don’t have the best history when it comes to free agent deals. Wolves fans haven’t had a lot, let’s be honest. But what do they have? Well, they hit on the picks they had to. Anthony Edwards and Karl-Anthony Towns are enough to build a very good team around. They invested 12 points worth of BPM into those picks and have gotten good returns. Jaden McDaniels is already better than his -0.5 expected BPM. Draft picks only matter when they become players. Cam Reddish was the future pick included in the Luka Dončić/Trae Young deal and we still think of it as a straight swap. These picks could be in the mid 20s and become stars, they could be in the top three and produce statistical benchmarks like Anthony Bennett’s gargantuan -10.8 BPM over expected. That depends on the Wolves being a losing team for the next 10 years. And that usually doesn’t happen with a team this talented.

Epilogue: You’re still here? Go home already!

I’m sure you all want to see who the best drafting teams are so here’s a final graph I threw together. No color coding for this one, just straight numbers. Team on the x axis, draft value over expected on the y. Enjoy!