iggymcfrack wrote:KG only got ONE first-place vote in the new voting? That's absolutely wild to me. Before the vote, I was wondering if he'd be unanimous this time. He was so dominant in impact stats. 2nd all-time for Engelmann, had a bigger lead over 2nd than 2nd had over 9th in the 97-14 RS+PS PI. His raw on/off was actually higher than LeBron's during the season even though he was lifting a +5 team to +16 while LeBron was lifting a -8 to a +2.
The project's certainly been unpredictable. I wonder what more surprises are in store. Maybe Kawhi or Curry over LeBron in 2017? AD over LeBron in 2020? If I've learned one thing from the 2008 results, it's that ANYTHING IS POSSIBLE!!!
Yeah, I think what’s particularly wild about it is that, given the team results, the argument for LeBron basically has to be that he was clearly the league’s best player…but individual data doesn’t really tell us that. For instance, EPM has LeBron 4th, behind two of the major POY candidates (Garnett and Chris Paul), and he was behind Chris Paul in playoff EPM too. He’s behind Garnett, Chris Paul, and Kobe in PIPM. He was 2nd in RAPTOR—well behind Chris Paul and not much ahead of Garnett. He was 4th in DPM, including being behind Garnett. He was 2nd in Win Shares, behind Chris Paul, and he was 5th in win shares per 48 minutes behind Chris Paul and Garnett. He was 1st in BPM, so that’s one in his favor. In Thinking Basketball’s BPM, LeBron is behind Chris Paul, Kobe Bryant, and Kevin Garnett. In AuPM/g, he is 7th, behind all the major POY candidates. One-year RAPM is super noisy, but FWIW LeBron is behind Garnett and Kobe (among a bunch of others) in that as well (and this is true whether we use TheBasketballDatabase’s raw RAPM or use Engelmann’s PI RAPM). Based on the data, LeBron is definitely in the mix for being the league’s best player, but he’s probably not even the one with the best case for it.
Of course, a major thing going on here is that the voter pool is largely made up of OhayoKD’s discord buddies so there’s definitely just an idiosyncratic voter pool, but if we are going to ascribe some general logic to it, I think a big thing motivating people is probably that LeBron was so good the next year. I do think performance in nearby years can sometimes inform how we see the data for a player in a given year, but it’s a little curious here because (1) the main reasons 2009 looks so good—winning 66 games with that Cleveland team and having incredible playoff numbers and amazing on-off—really aren’t actually present in 2008; and (2) many of the same group of people will generally argue that LeBron’s prime only started in 2009 while his peak was also 2009. The latter sentiment would strongly suggest he took a huge leap in 2009, which would make basing an opinion for 2008 on how good he would soon be in 2009 not particularly sensible. On this point, I do personally think LeBron was already in his prime in 2008, so I guess I wouldn’t characterize the leap he took as being quite as large as some of his biggest fans here would, but I do think 2009 was meaningfully better, and, if we’re looking at surrounding years to contextualize the 2008 data, I also think 2007 is at least as relevant to 2008 as 2009 is, and I’ll note that LeBron isn’t actually ahead of Garnett in 2007-2009 three-year RAPM (using either TheBasketballDatabase or NBArapm), so it’s not like using surrounding years gives him a clear-cut case anyways. If we looked at five-year RAPM, we can get LeBron in 1st if we start far enough forward (i.e. if we go through 2010), so that’s a point in LeBron’s favor, but Garnett is close anyways.
Overall, I do find it a bit curious. We basically have several players that the data has in the mix for being the league’s best player, and the voters chose one who probably doesn’t even have the best data case for best player and also had the least team success.
