What to make of Harrison Barnes

Harrison Barnes has one of the largest gaps between his perceived value by traditional analysts and viewers, and his value as suggested by advanced stats. This has been the case for his entire NBA career, including when he was a draft prospect, and yet even a poor performance in the NBA Finals couldn’t extinguish the demand during free agency. He’s essentially being replaced by Kevin Durant in one of the most lopsided exchanges in NBA history, but I don’t think people understand, or will understand, how lopsided this is. However, this effect will be neutralized by a dose of regression to the mean that will likely hit the Warriors and Durant for various reasons, and in some sense this idea that Barnes is a lot worse than many people think doesn’t functionally exist.

More from Nylon Calculus

How poorly Harrison Barnes does across just about every metric is his most impressive attribute. His PER last season was 12.6, which is below the average of 15 and not far from the dreaded replacement level mark of roughly 11 (this is from John Hollinger, and since his replacement level study was from a few years ago I thought I’d round it all to 11); his Box Plus-Minus was -0.2, and that takes into account his floor-spacing via his three-point volume (BPM adjusts for your team’s rating, so it’s surprising a player on a historically great team rates as below average when the consensus is the opposite); his Real Plus-Minus was -1.28, and that should at least halfway decently measure his defense (his RPM wasn’t far from the replacement level mark of -2 once again); similarly, he’s at -1.44 in RAPM (that’s the multi-season version weighed for recency. The single season version has him at -0.36); and his PIE was a mere 7.9 (what’s PIE? Oh, who cares). The only popular metric where he’s above average is Win Shares, and it’s an old metric where he’s just barely above average. At best, when looking at his other seasons and being optimistic because of his age, he’s near average, and at worst he doesn’t perform much better than the players you can theoretically get for the league minimum. Additionally, there is a fear about what his stats will be one he’s away from his historically great shooting and passing teammates, because Barnes is not talented at creating his own shot.

Kevin Durant’s stats are, obviously, gold-plated and deluxe, but he’s a scorer who has to share the ball with Stephen Curry and Klay Thompson now, as well as point forward Draymond Green. His efficiency will probably climb, but his overall total value could fall. On the other hand, he played with Russell Westbrook, a ball-hog in the best sense of the word who dominates the ball more than Curry. I think Durant will be fine, but the remaining Warriors will have a tough time following up on what is (for the time being…) the best regular season ever in terms of wins. Curry, in particular, by any reasonable, objective, sane projection system will not have the same efficiency next season on that number of attempts. In most universes, the Warriors, sans changes, were going to be worse next season anyway; it’s an incredibly difficult feat to pull off twice.

For instance, if Stephen Curry reverts to his previous gold-standard level, i.e. his 2015 season, then his BPM would fall by 2.6 points per 100 possessions. That’s the difference between an average or mediocre player and a solid starter. In other words, the perception disparity of Harrison Barnes equals the expected regression from Curry (FiveThirtyEight’s CARMELO system has Curry regressing to his 2015 level). But advanced stats don’t matter as much as one simple stat — wins. That’s how people will judge this team in the regular season. But last season, the Warriors actually outperformed their expected wins by a whopping eight games. It’s tough to repeat that performance, and most of Durant’s value may be used to cover that deficit. They could in fact be much better and at the same time not repeat their 2016 performance.

Of course, Golden State may retain its close game prowess, as when crunch time happens that can unleash the Mega Death lineup. People may still lament the loss of Barnes and his defensive credentials, which mostly amount to him guarding Zach Randolph — I don’t think the threat of Zach is enough to cause any worry in this exchange. Barnes has been tempting people with his potential since high school, where he was a number one recruit, and though his stats suggested a guy you might want to take outside the lottery he was selected seventh by Golden State. The Warriors displayed some of the most blatant tanking ever to get him, and I don’t really think it was worth all the hassle. Draymond was their gem, and he was taken much later. But given the ionospheric-sized expectations, the Warriors may disappoint people a little and thus reinforce the belief that Harrison Barnes was a significantly above average player for some great Warriors teams. He’s getting paid elsewhere, so this is one basketball mystery that will disappear in the chaotic wake of the 2016 free agency period.

Next: Does Durant make Dubs a better shooting team?

Home/Nylon Calculus

The Wake of the Black Falcon

By Justin | Jul 12, 2016