I have not had time to fully watch and digest this 90 minute video yet, but I want to get the discussion started.
The Sloan Conference is the preeminent gathering of sports nerds, executives, and math geeks. This particular panel discussion was called:
"What Geeks Don't Get: Limits of Moneyball"
I'll try to absorb this in the next day or so. When I do, I'll post my reaction here. For now, this will serve as the home base for this particular discussion. I can't wait.
UPDATE: I've watched the clip, and I have to say, it was well worth the time. Polian commented on a wide variety of topics. Here are some of the highlights.
On use of advance statsitics in the draft: Polian said that the Colts use a variety of advanced stats in talent evaluation. Their entire model is built around finding players that are undervalued by other teams. Shockingly, he said the Colts have metric they use and plug all potential draftees into it. If a scout says one thing and the metric says another, the Colts go with the metric. I found that shocking, but insanely cool. Later Johnathan Kraft of the Patriots talked about how New England and Indianapolis are two of just four teams that use a high priced psychological consulting service to evaluate potential draftees. The reason why the Colts are so 'odd ball' when it comes to the draft is that they are absolutely not paying attention to the same things that everyone else is. Keep that in mind this April and as you are reading mock drafts.
On use of advanced statistics in games: Polian emphatically said they are all crap and useless. He had especially harsh things to say about any stat that models how teams should behave on downs and distances. This was the comment that caused no end of consternation at FO. Polian's point is that football is too complex with too many variables to rely on averages across the league and across seasons to make decisions based on statistics. What's interesting is that he then used some statistics to show why the Pats were right to go for it on 4th and 2. It was a little inconsistent.
While that is obviously true to an extent, I think the stats can help you change your default starting place for making tough calls. For example, if your default is "Going for it on 4th down is a risky play", then your end decision will be colored in that direction, even after taking into account game scenarios. If, however, you realize that going for it on 4th down pays out over the long haul, you can still take game situations into account, but you are starting from a better default position. For instance, in Cleveland in 2008, the Colts went for it on the goal line on fourth down. At the time, I was actually against the call because of the game scenario (points at a premium, inside of 2 minutes to play in the half). However, as everyone knows, under almost every other scenario I'm in favor of going for it.
As for other kinds of measurement of players, the big problem is that without access to coaches tapes, outsiders will never be able to analyize play as well as coaching staffs.
On the next frontier of analysis: He wants better ways of measuring the effect of different officiating crews on games. He clearly doesn't like the randomness. He also wants to see a way to compare players across scheme. How much would Vince Wilfork be worth to the Colts who run a totally different scheme, for instance.
On 'the cutting edge': He claims that NFL coaches already push the envelope to the limit when it comes to data collection and analysis. The games are so tight and complex that most coaches want as much data as they can gather. He says any new metrics would be immediately accepted if they proved valuable, because coaches just want to win. I found this surprising. He clearly has a higher opinion of NFL coaches than I do. As with any disagree I have with Polian, I can only say, "He must be right. I must be wrong...but it sure doesn't feel like it".
Other highlights included Mark Cuban's game of "Protect the Moron", Bill Simmons providing comic relief, and the auro of "we're just better than you" given off by Kraft and Polian toward the rest of the NFL. There's little wonder those teams dominated the decade.