When advanced metrics lie

A guy like John Hollinger only speaks in numbers. His intention is to display information, and while he may or may not have his own agendas, he provides an entire database worth of information, available sites for the average layman such as nba.com. He’s not an entertainer, and he doesn’t have a nickname like “the Stats Guy.” He’s as serious as serious gets.

A guy like Bill Simmons is an entertainment writer. Yes, he deals in facts as well, but since he primarily writes opinion-pieces for ESPN, under the moniker “the Sports Guy,” we have to remember that whenever he writes, he has an end-goal in mind. In this case, he was writing to debunk advanced metrics as being a perfect measure of a player’s talent (though no one has ever made that claim). Naturally, like any writer with an intent to make a case or convince a reader, he’s going to use evidence that supports his theory.

There are definitely cases where advanced metrics don’t tell the whole story. We know this. The creators of those metrics have conceded this. However, by and large, these metrics generally show you who the winners are. If you go back in history, before the metrics were used, and apply them to the winning teams over those years and to the losing teams, you’ll undoubtedly find a huge trend towards winners having high PER and high WP/WS/EWA/etc. There may be exceptions to the rule because these things are not perfect, but they are pretty darn accurate.

For me, I like to use these advanced metrics in certain situations. When I’m trying to figure out who the guys are who are contributing (however sneakily) to the team’s success, or finding the ones who who are hurting the team, despite having gawdy scoring numbers. These metrics are absolutely fantastic for these uses, and time and again, they’ve identified why Randy Foye may get Mike Smith overly excited, but secretly he’s still hurting the team and should probably be traded while his high scoring numbers confuse the league into thinking he’s valuable.

Contrarily, I don’t like to solely use these metrics when trying to figure out who to trade for. In these cases, advanced metrics should certainly be used to identify producers out of the available free agents, BUT I think that the eye-test should be used to try and see which guys fit best with the team. I think Corey Maggette would be great for a team like the Blazers or the Celtics or the old version of the Spurs. A slow-paced offense that likes to pick at the defense would be fantastic for him. Golden State’s run-and-gun offense? Not so much. Likewise, if you put Marcus Camby in a situation where he has to be a key scorer who has to create his own shot, he’s about as useful as a Pomeranian on offense, regardless of how efficiently he scores. For some teams that have trouble scoring, that’s a huge problem, so Camby really doesn’t fit the need, even if he adds a ton of wins and has a high PER. You put him in a situation where there are other scorers and he can sneak his way into offensive putbacks or open jumpers on slow-rotating defenses, and he’s an incredible rebounder who can give you 8-12 points a night, and he fits perfectly on that team (—> …LA Clippers… <—). So I maintain that you can’t solely use advanced metrics, at least when trying to build a team. There has to be a give and take based on the player’s fit with the team in question.

To the surprise of nobody, Simmons was pushing the Rose for MVP storyline.

A discussed of the merits of looking at advanced metrics in a vacuum versus using them as one tool for evaluation is an attack on one particular person? And such a discussion has nothing to do with the Clippers? I think your bias is clouding your analysis.