The Browns, analytics, and finding market inefficiencies as an organization

Banner and HaslamYesterday, I wrote about how I thought that the analytics revolution in Berea would be subtle. Today, I’m going to suggest some practical ways that analytics could be used inside the building to help the Browns. More often than not, it will be supplemental reports to help increase confidence levels in decision-making processes in an effort to reduce the frequency of inevitable mistakes. Put simply, there are so many decisions made in the NFL that you’re going to mess up sometimes. Why not do everything you can to mess up less often?

Teams will need scouts forever, so don’t think that scouts will ever be antiquated. Let’s just say that the Browns decided they desperately needed to find 3-4 linebackers in this draft from rounds three through seven, you know, hypothetically1. Wouldn’t it be worthwhile to take all the best 3-4 linebackers, put them in a data set with their heights, weights, various combine stats, NFL years played, NFL dollars earned, and more? Then all of a sudden you’ve applied some filters to an unmanageable list and let your scouts concentrate on a more focused group of players. That’s a no-brainer and most scouting departments do this kind of stuff already, I’m guessing. Add a few more layers of complexity, start getting creative with the variables and that’s where the next competitive advantage comes from. If you’re truly in sync as an organization you’re spending your time becoming an expert with a group of players that you aren’t competing with the whole league to get. That’s what Pete Carroll did.

Pete Carroll seemingly accomplished this at the cornerback position. Pete Carroll has giant physical corners in Brandon Browner and Richard Sherman. Browner is 6-feet-4-inches tall and weighs 221 pounds. He’s a 28-year-old journeyman who was most recently with the Calgary Stampeders before Pete Carroll signed him up. In 2013 Browner will complete the three-year $1.29 million deal he signed. Granted Browner wasn’t drafted by the Seahawks, but Richard Sherman was. He was taken in the 2011 draft in the fifth round. He’s 6-foot-3-inches tall and weighs 195 pounds. His contract is four years and $2.2 million that runs through the end of 2014.

Contrast these guys with guys like Darrelle Revis and Joe Haden who are quick 5-foot-11 guys who make tons and tons of money. Granted, both Revis and Haden were under the old rookie salary structure, so let’s look more recently at the 2012 draft. The Bills took a corner in the first round 2012 when they grabbed Stephon Gilmore out of South Carolina. Even under the new salary rules for draft picks, Gilmore signed a four-year deal for just north of $12 million fully guaranteed. In one player taken highly in the draft the Bills are spending more than the Seahawks have in their two big, physical corners combined. In the salary capped world of the NFL that’s important when constructing the best possible roster of players to compete.

That’s not to say that Gilmore is better or worse than either Sherman or Browner. It is simply to point out that the Seahawks found a market inefficiency between the style they wanted to play and the way the market values cornerbacks.

Pete Carroll wins for now, but it’s probably short-lived. Assuming in the world’s largest copycat league that others now try to copy Pete Carroll’s design, the price just went up on big corners. Those big corners on the draft board with 4.56 40-yard dashes that were lasting until the fifth round might go much sooner and get paid more.

So, this year’s advantage is next year’s over-payment.  That’s where analytics kicks in further in an organization all working together on the same page. Subtle coaching changes to a scheme to fit a certain player characteristic helps the scouting and personnel department focus on getting pieces that are available and useful. The analytics department helps analyze the list of players and hopefully limit the necessary scouting time on players that won’t fit. Maybe the analytics department uses win probability numbers in certain situations to even help the coaches decide on a subtle change to their scheme.

Back to the Browns. Let’s say the Browns want to get off on the right foot defensively this year, but they just don’t see big middle linebackers or elite pass rushers that will be in their grasp? Maybe Ray Horton and his staff decide to run a new variant on the 3-4 that relies on linebackers who resemble big safeties rather than typical linebackers. They decide that they will drop into coverage more often than not. Then all of a sudden, maybe the Browns’ scouts have a mission to find “undersized” speedy linebackers and instead of being a disadvantage, they’re creating the new thing that everyone else wants to copy.

I’m making all this up, of course. I have no idea if that would work. Ray Horton is the defensive coordinator for a reason. I’m just pointing out that this is yet another way an organization is able to stay competitive all the time through innovation other than the necessary drafting and development that every team aspires to do well.

Yet again, another example of where analytics can help supplement an organization by enabling creativity and helping them execute a plan. More coming on all this. I will discuss how we need to separate process from the obsession with outcomes. A guy I met at the conference tackled it a bit already here, but I’m going to have my take on it as well soon.

Now, I wonder if there’s a way to exploit market inefficiencies in franchise quarterbacks? Joe Flacco’s agent thinks not.

  1. Ha. Hypothetically. []

  • mgbode

    Craig, I don’t quite follow the LBer example. If Horton decides to go with small speedy LBers, then we go after them; how is that utilizing analytics? that seems to be straight up “old school” scouting.

    same with the Pete Carroll CB example. He designed a defense built on bump-n-run coverage and then grabbed physical (big) CBs to handle it.

    I don’t doubt there are analytics they could run to tell them if it would work or who to target specifically, but you don’t touch on any of that here.

  • Ezzie Goldish

    Yeah, agree with mgbode – the LB example is actually reverse analytics. IF they had a theory then tested it, and the analytics showed that it was more successful in some/all situations, then that would be great, but that’s not using analytics to make decisions. It’s checking that decisions were right using analytics. (Still useful, just not the same.)

    That said, if they had analytics that said “speedy linebackers who drop back more don’t hurt a team’s rush D as much as it positively impacts the pass D”, then based their scouting accordingly, that would be a great use of analytics.

  • Pete Carroll’s example is an example of market inefficiency. I was just trying to present a sample project that a team could undertake and lean on analytics to supplement their normal jobs. They could use atypical stats to analyze the draft class to see if there’s a big supply of a certain quality at a certain position. They could use atypical stats to figure out if their scheme could be altered slightly depending on successful plays they’ve run in the past. The problem here is that there’s no tape on Ray Horton with this Browns roster. That’s why I wrote yesterday that I thought it was going to be subtle and take some time.

  • The market inefficiency lies in the fact that taller, heavier, cornerbacks are not valued highly for many traditional/scouting reasons, i.e. stiff hips, inability to fluidly change directions, too slow etc. These traditional or market concerns may or may not be valid, but the point is that you can make some technical or schematic alternations and seize the value in those players that are being overlooked.

  • mgbode

    Ok, I read again and think that the word “analytics” in the title along with the recent discussions here in general biased my reading to looking for the concrete example of analytic usage. This is more about the market inefficiencies in general (and extending that analytics can help). Apologies.

  • Lunch

    A few questions remain.

    1. How strong are the analytics department of the Browns compared to other NFL teams?

    2. How many personell within the Browns organization will listen to analytics?

    3. Will Jimmy Haslam build a “Mike Lombardi, The Redeemer” statue in front of FirstEnergy Stadium, if all of these analytics, and consensus decisions work? (You know Lombardi will get the credit for the Browms draft selections.)

  • Alan

    They brought in Scheiner, I would hope they are ready to listen to analytics, or what would be the point?