Doug Collins Hates Advanced Statistics, Goes By Gut

Doug Collins recently told Mike Jenson of the Philadelphia Inquirer that he prefers to rely on his gut instinct rather than statistical analysis.

"No. If I [were an analytics guy], I'd blow my brains out," Collins said about the subject.

"My analytics are here . . .and here" Collins said, pointing to his head and to his gut.

That's an interesting statement, considering that just a day before Joshua Harris mentioned that the Sixers were close to finding their new director of analytics.

"We are in final negotiations with a candidate," Harris said. "We think we are going to get something done there in the next month."

My first reaction to this was to make it a big deal. To use it as proof that the Sixers are behind the curve. But they're hiring a guy. Tony DiLeo and Joshua Harris have both been bringing up how much they value statistical analysis at every opportunity. The organization values it.

Certainly, Doug Collins can ignore whatever advice the "analytic" gives him, and that's a concern. But until we see them consistently making moves that fly in the face of statistical analysis it's right now only a concern, not something to jump off the Benjamin Franklin Bridge about.

I've also been in a media huddle where Doug Collins has referenced plus-minus. I've heard Collins talk about basic sabre-principles, like forcing teams into long range two pointers and running them off the three point line. Even further on down the Inquirer article Collins talks about getting 40 points from the three point line and foul line combined, which is beginning to sound like he's finding ways to improve their true shooting percentage and free throw rate.

I've seen Doug Collins rattle of traditional box score stats probably every game I've been credentialed for. Anything from "our opponent averages X turnovers per game" to telling you exactly how many free throw attempts Thaddeus Young had in game 2 of the Eastern Conference Finals last year off of memory. The man is practically a walking encyclopedia with a seemingly photographic memory, and he looks at each stat sheet after every game.

I also know Doug Collins can tell you where everybody on the team shoots the best from, where everybody on the opposing team shoots the best from, and what sets each player prefers to operate out of. There is no doubt whether or not he does his homework, whether that is using "PPP" or gut.

He might not know what PER is an acronym for, or the calculation that goes into figuring out a players win shares, but I think Doug Collins probably understands the tenets of what statisticians speak.

I also think there's a fairly good chance Collins doesn't blow them off as much as he projected. Knowing Collins and his attitude somewhat, he was probably just as likely to be making a joke about his lack of desire to perform arithmetic operations himself as he was talking about completely ignoring defensive rebounding percentage in lieu of defensive rebounds per game.

What I don't get is essentially what Rich Hofmann said last week when he soundly thrashed an article in the Philadelphia Inquirer about it being good the Sixers avoided an "analytic" in their general manager search. Why is there a divide? Why is statistical analysis and scouting viewed as mutually exclusive?

As somebody who is a college basketball scout, I watch game tape. A lot of it. Probably not as much as Doug Collins does, but I would bet more than most in the media do. I've watched more basketball from the 2011-2012 college season during the last two months than most people did while the 2011-2012 season was going on. Virtually every night for the past two months I have watched hours of college basketball. What was I doing this morning at 2 am? I was watching Nevada vs Portland from last December. Synergy Sports is both the greatest thing to happen to me and the bane of my existence.

I scout because I love watching basketball. Any notion that I prefer true shooting percentage over field goal percentage so I don't have to watch the game is patently absurd.

"Head and gut" guys are not the only ones who "watch the game", which is how people who don't believe in advanced statistics commonly portray those who do believe.

I believe this is a very dangerous topic to stick your head in the sand about, especially as we get better at isolating cause and effect.

What gets me is that, even the most old school of thinkers, the guys who live and die by their guts, the guys who "watch the game", will cite box score statistics. "Yo man, Monta Ellis scores 20 points per game". "Andrew Bynum could be that 20/10 guy the Sixers need."

Yet the moment I use defensive rebounding percentage I no longer watch the game.

It's a fallacy that gets me every time. If I say "Spencer Hawes pulled down 12 rebounds last night", I still retain my "watch the game" credibility. If I say "Spencer Hawes had an 18.4% rebounding percentage last night", I should stick to my excel spreadsheets and let scouts handle basketball activities.

Yet we're measuring the same thing, and we're both using statistics. I'm just using the statistic I think is more representative of the skill set we're talking about. I would bet if I sat down and was able to explain it, I could get most "watch the game" guys to realize why rebounding percentage is more indicative of the impact a players rebounding had than rebounds per game is.

And that's all "analytics" are trying to do. They're trying to find stats that are more representative of impact than the basic statistics available in the box scores. Sure, there are some statistics that try to look at the entirety of a player, but that's a very small subset of the advanced statistics field. The vast majority of the stats used every day are simply trying to find the flaws in traditional stats and find ways to make them more representative of impact.

Rebounds per game doesn't tell you how many opportunities a player had to collect those 8 rebounds. Assists per game doesn't tell you how much a player dominated the ball to generate those 6 assists. Field goal percentage doesn't tell you how many "extra" points a player got from three's or free throws.

We're not trying to reinvent the game. We're not trying to replace watching the game, which is probably where certain media types and basketball people draw their fear from. We're simply trying to find a way to take Doug Collins saying "we want to get more points from the three point line and free throw so we can get more points in the same amount of possessions" and turn that into something quantifiable.

In the mean time, if the new "analytic" wants to be listened to, rather than telling Doug Collins a lineup with Jrue Holiday and Dorell Wright had an Offensive Rating of 107 and a Defensive Rating of 98, he would best be served to state it as "we tend to go on runs when Holiday and Wright are in the game together".

At least until people get over their fears.