JaMarcus Russell: Concorde of the NFL

With Jamarcus Russel’s recent benching, there’s been a lot of talk about when it’s time for a team to cut its losses on a failed quarterback. I don’t have hard numbers at my fingertips, but I’d be fairly certain that if a QB isn’t playing above average football or there hasn’t been steady improvement, by the end of his second year, it’s time to move on. [Edit: Here's a good look at that very question at PFR.] There’s no question teams tend to stick with struggling QBs well beyond their expiration date, even when better alternatives exist. The real question is, why?

Let’s say you’re an out-of-town Bills fan, and before the season began you were understandably optimistic about the team’s prospects. You bought prime tickets to the January 3rd game hosting the Colts, including parking and a hotel room. Altogether the bill comes to $400. In August, this feels like a great deal.

As the season wears on, it becomes clear the Bills aren’t contenders. The coach is fired, and the upcoming Colts game is not looking promising, as the Colts appear likely be playing for home field advantage in the playoffs. Everything points toward a humiliating blowout. What’s worse, as the game approaches the weather isn’t looking good. Bills fans are always the hardy type, but the foercast is beyond bad—snow, wind, freezing rain, and bitter cold. You’re not exactly excited about the prospect of going to the game.

A few days before the game, your friend invites you to watch the game at a party to inaugurate his new palacial home theater. You’d really rather do that than actually go to the game, but you’ve already sunk $400 and it’s too late to sell the tickets. Naturally, you can’t let those tickets go to waste, so you suck it up and go to the game.

But this is completely irrational. It’s called the sunk cost fallacy.

To understand your mistake, think of your options in terms of costs and benefits. We’ll call the $400 you spent a cost of -4. Actually going to the game would be a benefit of +0, since it doesn’t appeal to you. And going to the party is a benefit of +2. Here are the options for January 3rd in matrix form:

Cost-Benefit Matrix

Go to game-4+0-4
Go to party-4


Going to the party is the better option. You’ve spent the $400 already—it’s out the door no matter where you spend your Sunday. Economists would call those $400 ‘sunk’ because there is no way to retrieve them. You’re better off ignoring the money you spent and making the best of your day.

Sunk costs play tricks on our minds. People will forego alternatives belieiving that previous costs will somehow be ‘made good.’ We all have a stubborn loyalty to our past mistakes. Studies have shown that people honor sunk costs more when they believe they have responsibility for them. Other studies have shown that people become overly optimistic about a project’s outcome when they are responsible for the sunk costs.

A famous example of the sunk cost fallacy is the Concorde supersonic passenger jet. Planned and built as a joint effort between the British and French, early in its development it became obvious it could never make money. Decision-makers pushed forward in spite of the projections because they had already invested so much in the project. But those investments were irrelevant. They’re sunk costs whether the program continues or not. Only future costs and benefits matter.

Jamarcus Russel is the Concorde of the NFL. Russel was the overall #1 draft pick in 2007. For three seasons now, he has given every indication, statistical and otherwise, that he will not become a decent quarterback anytime in the near future. Russell cost the Raiders the #1 pick, over $30 million in guaranteed money, plus many millions more in base annual salary. All of these costs, except for his future base salary, are sunk. They’re already gone and they’ll never come back, no matter how good or bad Russell turns out to be.

If Bruce Gradkowski, who makes about $400,000 per year, is the better option at quarterback, then the past investment in Russell shouldn’t matter. Starting Russel is like going to the Bills game. You thought it was going to be awesome and you invested a lot, and you just can’t bear to let that investment ‘go to waste.’

Russell certainly isn’t the only top pick who was kept under center too long. Just about every team has had a similar experience in recent memory. General managers and coaches are the ones least willing to cut their losses with bad players because they’re the ones most attached to the sunk costs. The importance of responsibility is why it makes some sense to periodically replace senior management, whether at corporation, a government agency, or professional football team. New managers are not beholden to their predicessors’ sunk costs, and are freer to make rational decisions.

You don’t have to be an NFL GM to fall victim to the sunk cost fallacy. In fact, I’d bet you’re doing it right now—go take an honest look at your fantasy roster. We fall prey to sunk costs all the time in real life too, whether it’s with a crazy girlfriend, a falling stock market investment, or a dead-end career.

There are a number of theories that may explain our vulnerability to sunk costs. Our strong aversion to losses, whether in the future or the past, might be part of the cause. Another theory is that people are sometimes acting rationally when honoring sunk costs, especially in political settings. It may be better for a decision-maker in the short-run to deny making a mistake for as long as possible.

I think the most likely explanation has a psychological basis. Almost everyone perceives themselves as rational, competent decision-makers. For example, in survey after survey the vast majority of people claim to be above average drivers, which is obviously not possible. When a situation occurs that challenges our self-perception as competent, such as when we invest in a failing project, we need to somehow reconcile the difference between our perception and reality. This difference is known as cognitive dissonance, which produces stress, anxiety, and a number of other negative emotional consequences. Honoring sunk costs is a way of avoiding the admission of error and the psychologically painful reconciliation involved.

So what should we do when faced with a sunk cost situation? The best approach is to ask yourself what you would do if the asset fell to you for free. What if those Bills tickets landed in your lap a few days before the game? You’d almost certainly not go.

Imagine that this past off-season Jamarcus Russell suddenly showed up on the roster of some other team, say a team with an unsettled QB situation. Would that team’s coach think, “We gotta find a way to make this guy our starter?” No, of course not! The coach would rightfully think, “If he’s not as good as his $5 million base salary, he’ll be cut before I can say ‘Get me Jeff Garcia’s phone number.'”

  • Spread The Love
  • Digg This Post
  • Tweet This Post
  • Stumble This Post
  • Submit This Post To Delicious
  • Submit This Post To Reddit
  • Submit This Post To Mixx

19 Responses to “JaMarcus Russell: Concorde of the NFL”

  1. Anonymous says:

    however, right or wrong, those are not the only payout factors going into the decision. Many of the managerial people responsible for JmR being selected, in this case this includes the owner, have additional 'negative' benefits (as opposed to the costs, which are sunk and fixed) to "going to the party". These some of these costs: the immediate effects of negative perception of a failed pick in Jamarcus (as opposed to discounted in the future), lost opportunity to 'rectify' these failed picks with future selections that appear successful, forfeiting any possibility that JmR turns it around and the vindication of the past losing seasons as 'necessary' for his development. Unfortunately, more is at stake for these managerial decision makers on the success/failure of a #1 overall pick than a couple extra wins fof a bad team (5-11 instead of 3-13).

  2. Andy says:

    My wife believes that Russel is the smartest, most rational player in the NFL. Call it the "sunk benefit" strategy. He received 30M guranteed, plus a couple of years' salary. He has enough money to be extremely wealthy for the rest of his life, and to pay for his great grandkids' education and healthcare. There is little or no marginal value to a dollar after 30M. On the other hand, if he continues to play, there is a relatively high probability that he will suffer head injuries and other health problems that will negatively affect his life. Let's assume JaMarcus doesn't love football a la Favre. Why should he give up 10 years of freedom, and risk dementia, for a pile of money that he will never be able to productively spend? So the best play for Russel is to do exactly what he did: take the 30M and then try to get fired or at least benched. The Raiders may have paid for their own sink.

  3. Jeff Clarke says:

    I might talk about Russell later. Now, I'm more interested in the example. The biggest question I have for the Bills fan is why he didn't use scalpers. The public perception about scalpers is that they are selling tickets for 3 or 4 times face value. Thats occasionally true. Far more often, they are buying tickets from guys like this for 50% of face value and selling them for 75% of face value. Quite often the percentages are even less. If you don't mind missing the first two minutes of the game, you can get into most sporting events at less than half of face. Scalpers know their inventory is about to become worthless, and they have huge markdown sales.

    When you buy season tickets, you have to remember that "face value" isn't face value. In order to buy the regular season games, you need to buy the preseason games which you don't really want. Out of the 8 regular games, you will probably have something else you want to do for two of them. So you are buying 10 tickets when you really only want 6. If you throw in the various taxes and fees, you are actually paying more than twice face value per ticket that you really want to use.

    Go to the game and spend 10 minutes negotiating in the parking lot and you can get a far better deal the vast majority of the time.

    Are there counterfeit tickets?

    I've probably bought tickets to over 100 sporting events and I haven't gotten a fake one yet. You are taking a risk but its a very small one. If you're paying thousands of dollars for Super Bowl tickets, you need to be very careful. If you are going to a random game for less than face value, the risk is really small. Teams like to overstate the risk because knowledge of the secondary market hurts the primary market. They make a lot of money selling to people that don't use the tickets.

    What is the going rate?

    It sure isn't what you heard quoted on the news. They like to take the most absurd quote they can find and say "tickets are selling for XXX". The real price is often 10% of XXX.Go to eBay in the morning and look up completed listings for the game. Look at the prices that people are actually agreeing on and reduce it by 30%. Thats your opening bid. You can probably haggle it down to 10% off eBay prices. It goes without saying that what the scalper says is the price isn't the price.

    One more thing, never ever walk up to a scalper and say "I want good seats". Walk up and say "I want the cheapest seats you have" even if you want good seats. Often he only has good seats and will be willing to negotiate. If he actually has bad seats, you can always segue into the good seats after setting the initial tone for being price conscious.

  4. Chase says:

    Jason Lisk looked at when teams should give up on young QBs; click on my name for the link.

    As for Russell, I don't think the analogy fits. The fact that your tickets were worth $400 doesn't tell you anything about how valuable those tickets are now; we know those tickets are just about worthless. We know we don't want to go outside to watch the Bills.

    I don't think we know how good Russell will be. Highly drafted QBs that are bad their first two seasons still project to be better for the remainder of their careers than undrafted QBs that are bad their first two years. A QB's draft slot -- especially when it's really high -- does give us some information, even after two years.

    As for Russell specifically, I've always said I think we should throw out his rookie season. He missed all of training camp and he came out as a junior. For all intents and purposes, his rookie year was meaningless.

    His second year -- what I would call, at age 23, his true rookie year -- was pretty decent. In terms of yards per attempt, Russell was close to league average. In terms of adjusted net yards per attempt, Russell practically *was* league average. To put it another way:

    At age 23, in his first real season, Russell averaged 5.3 ANY/A. And was the former #1 pick in the draft.

    At age 23, in his first season, Joe Flacco averaged 5.3 ANY/A. And he was a former first round pick.

    Flacco happened to play on a team that had an incredible running game and defense; he also had better offensive weapons, and by defition, better coaching.

    So coming into this season, I'm not sure if Russell looked any "worse" than Joe Flacco, a player everyone was ga-ga over. Meanwhile, Russell was a higher pick and is 8 months younger than him!

    This year, Russell has been awful. Really bad. But I'm not sure I would give up on him just yet. Bart Starr was pretty good at age 23 and awful at age 24. Troy Aikman was really bad his first two years. Doug Williams struggled on some awful Tampa Bay teams. I don't think the book is closed on Russell, especially if he's on another team in the near future.

  5. Shaun says:

    "Jamarcus Russel is the Concorde of the NFL. Russell was the overall #1 draft pick in 2007. For three seasons now, he has given every indication, statistical and otherwise, that he will ot become a decent quatyerback anytime in the near future."

    I dont usually disagree with your stuff Brian but had to say something about this. Russells "1st season" was ruined by a coach who didnt like him, and admittelly his own holdout. So technically we should be viewing this season as his 2nd, and yes he has been terrible I wont argue that. But your claim that he has given every indication that he will not be a decent qaurterback yet last season he WAS a decent quartrback, 198 of 368 for 2423yds, 13tds and 8 ints, for a QB rating of 77.1. not exactly a superb season but considering how poor the recievers at oakland have been the last couple of years not that bad. he had some stinker games like new orleans and atlanta but against denver, miami, new york, houston and tampa bay he was solid. typical of a 1st year QB really, I personnally think Russell still has the ability and potential to succeed in the NFL.

  6. Alex says:

    An interesting study would be whether "potential" (draft pick number?) is predictive above and beyond what we'd predict based on x years of actual statistics.

    For example, if a top 10 pick and a mid-second-round pick both play the same in year 1, is the top 10 pick expected to perform better in year 2 all else constant? Intuitively one would think the answer was "yes". If so, how about after two years? Three years? Does age or just experience play a role? (etc, etc.)

    If there's some sort of "potential" factor that gives higher-rated quarterbacks more upside even factoring in their nfl stats, that could be a reason to stay with Jamarcus. If not, or if that factor has diminished to unimportance by his experience level, or if his projection even incorporating additional "potential" is below whatever baseline, then it'd be logical to ignore sunk costs and bench him (or cut him or whatever).

  7. bytebodger says:

    I was going to post earlier and I now see that other commenters have helped me to make my point.

    I agree with you that Russell is godawful. Even last year, when he was supposedly having a decent or "average" season, I thought he looked terrible. But that being said, there are two issues to consider with high draft picks like him.

    The first - sunk cost - should be completely forgotten as you have pointed out in your article. But the second - undefined potential - cannot be entirely discounted. There are many examples (as previous commenters have discussed) of QBs who start cold - sometimes very cold - and then go on to have decent or even strong careers.

    Steve Young looked very iffy in Tampa Bay. But of course, at that point in NFL history, EVERY QB looked iffy in Tampa Bay. Ditto for Vinny Testaverde (who was no HOF'er, but did go on to have a very nice career). Brad Johnson didn't really light it up at first but ended up winning a Super Bowl. Same goes for Trent Dilfer. Mark Rypien and Doug Williams both had to rehabilitate their careers before taking the Redskins to championships. And the list could go on...

    So if you're a wise GM (or owner), I can imagine that you might be able to forget about those sunk costs. But they still have to be paranoid that the player they cut today becomes a Super-Bowl-winner somewhere else.

  8. Shattenjager says:

    In paragraph four, I'm guessing that's supposed to say "a few days before the game."

    If you want to see whether sunk costs are because of cognitive dissonance, I would suggest finding out if there are any studies on it in Japan. As strange as it sounds to most of us, Japanese people don't try to mitigate cognitive dissonance the way westerners do. I don't know if there are such studies (I was a psych major), but they might illuminate whether cognitive dissonance is a big part of the sunk cost phenomenon.

    Very interesting, as always.

  9. Becephalus says:

    Genetic Pool A spends three weeks building an elaborate trap for an upcoming heard of mammals. 2 weeks into it they find a hurt animal consisting of enough food to feed themselves for the immediate future, but they push on and complete their other project anyway. When the original food is stolen or spoils they are later very happy that they continued with the project.

    Genetic Pool B spent 2 weeks building traps, then lost interest when they stumbled across the bonus food. A week later they were very sorry when they discover that animals have found their cache and it is too late to finish the traps.

    Not a perfect example, but I don't have time to craft a better one right now.

    All that is to simply point out that the honoring of sunk costs is to me no mystery. The ability to defer instant gratitifcation and accept immediate costs for future benefits is such a valuable ability that a respect for sunk costs is a natural psychological nudge in the direction of long term planning.

    Is deffering to sunk costs rational in a situation where you understand most of the variables, not in the slightest. But on evolutionary timescales we were little better than animals and had little ability for rational evaluation of options. In such a context erring on the side of continuing large investments could very well have been an advantage.

  10. zlionsfan says:

    Actually, Shaun, he doesn't get a pass on that first season. I'd guess the list of quarterbacks who held out the majority of their rookie season and then went on to become, well, anything is really short.

    Yeah, he's not surrounded by All-Pro receivers in Oakland, but he has backs and TEs to whom he could throw as well, and it isn't all on the WRs. When you have completion percentages of 54.5%, 53.8%, and 47.1% (so far this season), well ...

    To put that into context, right now Russell's completion percentage is 60% of the league average. In the last 20 years, there have been 15 players to post a completion percentage no better than 60% of league average while throwing at least 100 passes. (19 occurrences - Mike McMahon three times - plus two more potentials this season, Russell and Derek Anderson.)

    Of those QBs, you have a lot of nothing. Michael Vick, Kurt Kittner, Stoney Case, Ryan Leaf, Alex Smith ... yeah, Alex Smith might turn out to be a good QB, as might Russell, but that's certainly not the norm for QBs who are that inaccurate. Accuracy doesn't seem to be something that QBs learn in the NFL. You either have it or you don't.

  11. Jeff says:

    Becephalus, your example isn't on point. The marginal benefit of finishing the work in your example is significant. The sunk cost fallacy is premised upon little or no marginal benefit to additional work.

  12. Brian Burke says:

    It's not a perfect analogy, but I get the point. There may be a survival-fitness reason why people honor sunk costs. Like a lot of other built-in biases, it may make sense in an uncertain, unbounded, unpredictable world. But sports is an abstract world, where purely unbiased decisions are best.

  13. Brian Burke says:

    Chase-point taken. Russel was the best high-profile example I could think of.

  14. Brian Burke says:

    Added the link to Jason's article in the main post above.

  15. Becephalus says:


    Like I said the example was kind of thrown together quickly. I was minutes away from having to make dinner for my wife.

    I am sure you can admit there would be scenarios when giving weight to sunk costs would lead to an overall improvement in behavior for an individual/group insofar as it helped override the normal tendencies to take immediate gains over deferred ones.

    I really like your method/research. All I meant by my point was to try to illustrate that most of the mistakes people make with their economic thinking have very simple explanations if you look back into the aboriginal environment. I think sometimes economists are a little too incredulous about people behaving sub-optimally given how different (and dynamic) the modern situation is.

  16. Shaun says:


    with his rookie season included, Russell has had by my count 25 different WR/TE/RBs to throw to, the one constant being zach miller who he does have a good chemistry with. The RBs have been constantly changing wih injuries, bad form, etc. As for the WRs starting 2 rookies isnt exactly the best idea. You find me another QB with so many recievers over so few games that has played considerably better, and has a good chemistry with them. Now im not absolving Russell of everything but when passes are on target you wnat your recievers to catch them, not drop as many as they have.

  17. Anonymous says:

    "There’s no question teams tend to stick with struggling QBs well beyond their expiration date, even when better alternatives exist."

    I don't think this is obvious. Do you have any objective evidence to support this view?

  18. Bigmouth says:

    Great to see you blogging a bit about behavioral economics. I believe the term you're looking for is "bounded rationality."

  19. Anonymous says:

    file under better late than never.

    The individual you hypothesize with sunk costs of $400 will almost never go to the party. Being embarrassed in a crowd of people you know is as big a dis-incentive as any I know.

Leave a Reply

Note: Only a member of this blog may post a comment.