Polian thinks Moneyball won’t work in NFL


The Bills plan to import Moneyball concepts to the NFL.  Plenty of NFL people are skeptical.

Our take was/is/will be that baseball’s pitcher-vs.-batter nature is more conducive to such measurements.  Football, in contrast, is 11-on-11 controlled chaos, with plenty of good and bad things happening once chaos overcomes control.  Great players who command extra attention make their teammates look better.  Likewise, no amount of film review can account for the question of whether a player properly or improperly fulfilled his actual assignment on a given play, and whether the play was designed in a way that gave the player a realistic task.

But there’s a bigger problem with Moneyball.  As former Bills, Panthers, and Colts G.M. Bill Polian explains it to Buffalo Business First, the primary impediment to the successful use of analytics in the NFL comes from the league’s business model.

“As a practical tool, Moneyball does not work in the NFL because there are very few undervalued players and no middle class because of our salary cap,” Polian said.  “There is no middle class in football because the minimum salaries are so high, and because of the salary cap, a player will reach a point where you can’t keep him.  They go.  They’re going to get big money elsewhere.”

Still, in any given year there are decisions to be made about who should be paid and who shouldn’t be paid.  Some great players get more-than-great money on the first day of free agency for reasons unrelated to football.  The smart teams concede the March press conferences to the big-spending owners, opting instead to get good players at a cheaper rate.

But that has nothing to do with analytics, unless a team chooses to use statistical analysis over old-fashioned film study to identify the best players.  And that get us back to our original point.  Obsessive-compulsive review of stats overlooks the reality that, when the ball is snapped, all hell often breaks loose.  Having the right mix of quality players and veteran leaders and youngsters who don’t get overwhelmed on the big stage  guided by competent coaches is the way to turn unquantifiable factors into the hard numbers of points scored versus points allowed.

43 responses to “Polian thinks Moneyball won’t work in NFL

  1. Moneyball principles thrive on baseball players that excel in specialty stats such as guys who get walked a lot.

    Football demands all guys harmoniously doing their tasks. The only specialty position that really comes to mind is a short yardage back which nearly every team already has.

  2. The analytics football outsiders use are cool. I’m sure if they’re so commercially available the bills are trying something much more indepth though

  3. What does Bill Polian know he only took the Bills to 4 straight Superbowl’s. Russ Brandon is from Sales & Marketing, and they know everything. So who would you believe?

  4. We’ve seen so many players who look great on one team and then get a big free-agent contract for another yet subsequently look bad. It all depends on the teammates, the system, the coaches, etc. I agree with the old fart…

  5. Leave it to Polian to explain that “moneyball” doesn’t work. Yet, his replacement in Indy cut nearly every high-priced player, resigned only two (Wayne and Mathis) below market value and filled the roster with rookies, free agent stop-gaps, undrafted free agents, and $30+ million in dead cap space.
    There might be truth to the fact that “moneyball” doesn’t work longterm, but the GM in Indy has figured out a way to make it work this season and walk into 2013 with some very experienced young players and $40M in cap space.

    Thank you, Mr. Polian. You suck

  6. “Obsessive-compulsive review of stats overlooks the reality that, when the ball is snapped, all hell often breaks loose.”

    Don’t you think the baseball people said something similar??

    Think outside the box, there is more to using statistics than deciding who to sign and who to start.

    I think it would be a GAME CHANGER if used correctly for things such as field position in relation to the decision to punt or go for it…as well as the decision to go for two or not.

  7. Moneyball is finding advantage through undervalue.

    It’s pure BS that it doesn’t work in football: slot receiver was under valued a few years ago and now a big part of teams, same with TEs now. Just because people don’t call it by name doesn’t mean people are not doing it.

  8. I expect the Moneyball approach to yield some useful ideas, at least initially. Sure, it probably won’t make it any easier to evaluate which TE you should draft, but there’s no reason why the approach can’t find areas of the team that are undervalued relative to their impact on winning. For example, is the amount of money put into your special teams equivalent to their impact on your likelihood to win. Or what is the impact of a single premium (read: expensive) CB vs several competent ones paid closer to the position’s average?

    These statistical tools don’t have to replace film review and scouting, but they can point out where you have made poor assumptions about the relative importance of a position.

  9. I dont agree with much he says but hes right on this.

    Too many subjective factors in football to isolate on each position amd get any worthwhile data.

    However, a fool amd his money….

  10. there are a lot of teams that should be courting Bill Polian as GM, Director of Operations, or President of the club. News stories about the Polians over the years paint Bill and his son as kind of obnoxious jerks. However, you aren’t hiring him to be your friend. I would want him as my team’s GM and I believe his assessment on Moneyball tactics. He’s right and the Bills are dumb.

  11. But that has nothing to do with analytics, unless a team chooses to use statistical analysis over old-fashioned film study to identify the best players.
    This is not an either or scenario. Analytics offers a less biased way to evaluate talent. The work that ProFootballFocus offers is an example of how statistical analysis complements film study. While a simply version is “all hell breaks lose” or controlled chaos, it is pretty clear when a right tackle gets bull-rushed into the quarterback or misses his block that he did something wrong.

    Individual NFL teams have a lot to gain from getting smarter in how to draft players. Think about the Ryan Leafs of the world compared to Alfred Morris or Geno Atkins.

  12. “Moneyball” is not a specific system that can or cannot be adapted across different sports.

    It is simply a guiding principle that says one should attempt to quantify player value in terms of production and determine value based upon results, not perception and that the amount paid to a player should be solely based on what he brings to your team, not what other teams are bidding or what would please your fan base.

    I think it is more important in a salary cap league because it helps you determine when to give a big contract to a player or replace him with someone who approximates his production at a lower price.

    “Moneyball” proponents I have read have criticized the contracts given to Kevin Kolb, Matt Flynn, Mario Williams, Chris Johnson, and DeAngelo Williams on the day they were signed, NOT in hindsight.

    Who turned out to be right and who turned out to be wrong?

  13. The issue isn’t that Polian is wrong; it’s that he’s not really looking at the grand scheme of MoneyBall. You take under-valued assets and give them opportunity. In football, you have players who haven’t been given a shot or have skills that are being utilized incorrectly. Couple that with an idea of at what specific age a position generally establishes themselves as being good and you have the ability to look at free agents or players who undervalued by other organizations who can come into a new role and succeed. (Chris Houston was a CB who Atlanta gave up on; Detroit traded a late round pick for Houston, who ended up starting and playing decently in the Detroit secondary. Typically you can see how good a CB will be after 3 years)

  14. Wow, reading through these comments it’s amazing how many people are confused on what the “moneyball” concept actually is.

    Perfect example of this is Wes Welker (and slot recievers). When the pats traded to get him from the Dolphins it was because he was looked at as an expendable player, too small, etc to be a classic WR. However the Pats expanded the role of the slot player and Welker very quickly became one of the most elite pass catchers in the game. All at a bargain price because the patriots identified his skill set before it was exposed and signed him to a long term deal.

    It’s the same principal that was making 2nd round picks more “valuable” than first round picks (after the top 5) because of the guarenteed money that was given to 1st rounders. The teams got better bang for their buck at pick 33 than at pick 16-32.

    Moneyball is essentially identifying skillsets that you can get on the cheap, before they become trendy and thus expensive.

    There are plenty of places to use analytics in football and to claim otherwise is just silly. Does that mean abandon scouts? Of course not. That’s like throwing out all your hammers because you bought screwdrivers. You need all your tools to build a house.

  15. If anyone is using Moneyball, it’s the Patriots. Moreover, they often go from Moneyball to Lowball (see, e.g., how they treated Wes Welker regarding a new contract).

  16. Polian’s analysis has merit. Mostly because baseball is a collection of individuals competing collectively. One’s fielding and batting have very little dependence on the play of one’s team mates.

    In football it’s much easier to have good stats as a receiver if you have a good passing QB, easier to be a good running back if you have a very good offensive line etc.

    I’m fine with advanced metrics, but they will be much harder to apply to football than baseball.

  17. ——————————————-
    anotheryx says: Jan 3, 2013 4:16 PM

    Moneyball is finding advantage through undervalue.

    It’s pure BS that it doesn’t work in football: slot receiver was under valued a few years ago and now a big part of teams, same with TEs now. Just because people don’t call it by name doesn’t mean people are not doing it.

    I have to agree with this.

    It used to be a big deal to have a “shutdown corner” so you can cut the field in half. This was fine when everyone ran 2WR almost all the time. Now it’s almost more of a game of how good are your nickle and dime guys versus the opponents 3rd/4th receiving options.

    Now I kind of doubt it will be as effective as baseball. There’s no minor leagues, so for reserves all you can look at is college, and whatever limited action they have in games.

    Additionally, the fact that there is a cap has already forced some of this to happen naturally. There are some teams that will not pay a RB period. They’ll draft and take whatever they can get for less than a million for 3 to 4 years.

  18. Moneyball may not work in the NFL without some outside the box thinking. For example, some offensive linemen may be great run blockers, but struggle in pass protection, or vice versa. Those guys likely make less money than a well rounded lineman. But, if you were willing to use some sort of OL rotation, that had the road grader in on 1st down and short yardage, and the pass blocker in on long yardage, maybe your OL could perform better than with just either one playing all of the snaps. This wouldn’t be too different from the way some teams use their RBs with 3rd down or short yardage backs. No team that I know of currently does this, and maybe it wouldn’t work due to “cohesiveness” issues. But, it may take that kind of outside the box thinking in order to try to make Moneyball work in the NFL.

  19. MoneyBall doesn’t agree with Mr. Polian, in fact the current state of the teams you linked him to should prove one thing, you don’t do what he thinks you should do. Hey is this some kind of strange reverse psychology through mass media to actually bring moneyball into the NFL?

  20. It is possible, but the amount of data that would need to be gathered for each play would be astronomical. It would also require a subjective assessment of each of the 22 players on the field to determine if they completed their assignments in a manner that either had a positive/negative/non impact on the play.

    One of the beautiful things about Bill James style of baseball analysis is that it relies solely on empirical data and removes the judgment errors that come from eyeball evaluation process, and there’s no way to do that in football.

  21. The Patriots have been practicing Moneyball since BB came to the team. I believe he was an economics major at Wesleyan College so this sort of analysis is right in line with way his brain clicks. Anyway, they play Moneyball but I think they do it as a matter of using statistics as one piece of the player evaluation and decision making process as opposed to the statistics dictating the decision making process. The Patriots are always assigning a ‘value’ coefficient while making player decisions so in that sense they are using Moneyball. I just don’t think they let the statistics dictate the decisions to the extent that teams do in baseball. But then again….who cares about baseball….yawn……

  22. Moneyball concepts will work in the NFL precisely because most people in the current power structure think it won’t work.

    I laugh at the NFL often for being stuck in the stone age for analysis and married to conventional wisdom even more than baseball was at it’s worst.

    Remember they said short QBs wouldn’t work in the NFL? LOL. Pete Carroll found Russell Wilson sitting there waiting to be picked.

    College football is an even richer environment for using money ball analysis than HS baseball or even college baseball. College football is merely one level below the NFL, just like Triple A is for MLB.

    Ted Thompson seems to add 12-15 new rookies each year, incredibly to a Super Bowl caliber team…..the Packers are always one of the younger teams in the league….Thompson basically has a “farm team” built into his roster every year and seems to find at least one new starter or two from the FA ranks, often above average too.

    Younger players = better players Thompson knows this.

    Polian IS conventional wisdom.

  23. The NFL is full of dinosaurs when it comes to statistical analysis. One day a gutsy coach will stop punting, go for it on every fourth down and always go for two after touchdowns. Remember that football players and by osmosis the coaches and GM’s that they become are not exactly the brightest bulbs in the bunch. Peyton Manning is considered a genius in the NFL. His SAT score of 1030 puts him at the 50 percentile of all males who take it. If he wasn’t a great QB he’d be lucky to be a bank teller.

  24. Doesn’t this basically happen every year by every team? Rookies are kept and vets are cut all the time. It is more a question of volume than concept.

  25. From what I gather sabermetrics is not just using stats to determine the value of a player….as I’m pretty sure teams have figured out that tactic. It, according to the doctrine, states there is an objective method to evaluate a players value. Now I’m not a gm, but I’m gonna go out on a limb and say a team is sometimes greater than the sum of its parts…I think the value of a London fletcher, Brian dawkins, ray Lewis, tom Brady, peyton manning, etc… Goes beyond how fast they can run or tackles aquired on third and short. There is an obvious effect these players have on the quality of play of their teammates…
    Unless you maintain that the giants were lucky twice in a short period of time, or the patriots won their superbowls every time because they had statistically the best players at the most positions, you’re gonna have to concede that something about how players have to interact on the field, call it qb wr chemistry, call it a rb working well with an oline (just ask shaun Alexander or priest Holmes), or a linebacking core that meshes well, you have to concede there is a value that goes beyond an objective statistical measurement.
    Therefore according to the doctrine of moneyball….it isn’t applicable to football.

  26. Also, in baseball there are SO many players in farm systems and internationally that it makes sense to seek out diamonds in the rough.. in the NFL we get 7 rounds and a chance to sign free agents from schools who played against high school level teams. I’m not buying into it….. yet.

  27. Okay, I’ve looked into this and believe the Bills mention of ‘Moneyball’ may be less about players and more about play calling… google “How Oregon Coach Chip Kelly Can Spark ‘Moneyball’ Revolution In NFL” There was an article written in November – maybe the Bills have knowledge of this article.. and combined with their desire to interview Chip Kelly………………..

  28. Ralph Wilson’s new football czar getting off to a rocky start with gimmicks instead of just doing the basics right: a serviceable QB who doesn’t throw the game away (bye Fitz) and a HC who can get the most out what should be a top five NFL D. Don Zimmer Cincinnati?

  29. It is going to depend on what analytics are used. Classically you look at the number of sacks, knockdowns, and hurries you get from a DE to see what his value as a pass rusher is but if you include the measurables of holding/false start penalties generated, sacks generated on the other side of the line while on vs off the field, and frequency of double teams beaten into the equation you might be able to better see their value. You can also take a look at the stats of the other side of the line and how it impacts the player you are looking at. You don’t want to put Mario Williams money into a guy that is only picking up stats because they play opposite Aldon Smith.

    You can make a bit more sense out of the chaos if you take a look at a bigger picture than individual stats and performance.

    It also doesn’t mean that you need to replace film study with the Compubot 5000. You can use it to supplement your film study and perhaps to point you in the direction of taking a look at some film you otherwise might have ignored.

  30. If moneyball would have prevented the Bills from signing Fitzpatrick to his deal then it is a start in the right direction.

  31. How many here have read Moneyball? Half the posts here are incorrectly describing the concepts.

    Also, the team that made it famous hasn’t won much have they? Being fiscally responsible is smart; you don’t have to call it Moneyball every time a GM is saving money.

Leave a Reply

You must be logged in to leave a comment. Not a member? Register now!