Learning From The Past: Free Agents The Cowboys Should Avoid

The NFL has just released its tentative schedule for 2011. Provided the players vote to ratify the CBA today, these are the key next steps for free agency.

Saturday, July 23: Starting at 2:00 PM ET, clubs may negotiate with, but not sign, undrafted rookie free agents, free agents, and other clubs’ UFAs, RFAs, and franchise players.

Sunday, July 24: Starting at 2:00 PM ET, clubs may begin to sign undrafted rookie free agents.

Wednesday, July 27: 2011 league year commences at 2:00 PM ET, provided NFLPA has ratified CBA. Free agency signing period begins. Clubs may sign free agents and other clubs’ unrestricted free agents.

And as the desperate scramble between teams for the best UDFAs and free agents begins, it's a good time keep in mind that desperation often leads to dumb decisions in the NFL.

Bill Barnwell, formerly with Football Outsiders and now with new ESPN offshoot Grantland.com, wrote an article over the weekend in which he looked at 18 years of NFL free-agency to come up with a list of free agents that teams should avoid. We look at that list after the break.

Barnwell identifies four archetypes that have proven to be particularly bad investments in free agency, and which teams should avoid at all cost, but probably won't.

1. The veteran running back

Barnwell argues that spending big money on a veteran running back has mostly been a fool's errand. This is a lesson the Cowboys have learned the painful way when they made Marion Barber one of the highest paid running backs in the league. Thankfully, the Cowboys are nowhere near repeating that mistake. This year.

2. The second/third wide receiver in an effective passing offense

Barnwell even calls this the "Alvin Harper Rule", arguing that the performance of this type of receivers is based more on the offense they're playing in than on their talent.

After three years as the starting receiver in Dallas across from Michael Irvin in the mid-'90s, the Buccaneers gave Alvin Harper a deal worth $2.65 million per season — just a smidge under the $2.9 million per season that Irvin was earning in the deal he had signed earlier that offseason. Harper caught a total of 65 passes and scored three touchdowns in two seasons before getting cut.

Care to take a guess which recent Cowboys signing was the second receiver in a (moderately) effective passing offense?

In 2007, the Lions were the 9th ranked passing offense in the league. The leading receiver on the team was Shaun McDonald and a rookie wide receiver called Calvin Johnson was working hard to relegate the previous year's number one wide receiver, one Roy E. Williams, into third place on the depth chart. The rest, as they say, is history.

The Cowboys may or may not cut Roy Williams this year. But if they do, there has been talk of bringing in a veteran wide receiver as the number 2/3 guy. Particularly with WR coach Jimmy Robinson joining from the Packers, there was the occasional muttering about getting somebody off the Packers roster to replace Williams. If the Alvin Harper Rule is anything to go by, that may not be the best option for the Cowboys. If anything, they should get a good receiver playing in a bad offense.

3. The veteran Steelers defender

In a similar rationale as in point number 2, the argument here is that Dick LeBeau's scheme benefits the players much more than vice versa. The Steelers franchized OLB LaMarr Woodley, their top free agent, but there are a couple of other defensive players due to become available. The Cowboys must resist the urge to splurge here.

4. The defensive end coming off a big year

The argument here is that as enticing as high sack numbers for defensive ends may be, for the most part they're likely to regress to the mean in 2011. Here's an example using the ten 3-4 DEs with the most sacks in 2009 and how their sack levels changed in 2010:

Of the ten players on the list above, only two were able to improve on their sack total from the previous year, two were able to maintain their level and six saw a drop in their sack totals. Overall production from this top ten group dropped by 21%. For the six players who weren't able to at least maintain their sack total, production dropped from 36 to 19 sacks, a decline by a whopping 47%.

You can do an exercise like this for almost any stat and end up with similar results. It's called regression to the mean and it occurs in almost all data sets that compare one period to another. Cullen Jenkins had his statistically best year with seven sacks last year at the age of 29. It is unlikely that he will repeat that feat a year later in Dallas.

The key heading into free agency is to find players whom you can pay for potential instead of past performance (which they are unlikely to repeat).