Payroll Adjusted Dollar Per Win Figures

If you haven’t yet, read Eric’s post below about the need to add context to contract analysis, depending on a team’s ability to contend. I don’t know that I agree with the idea that bad teams should never improve themselves through free agency if it doesn’t result in a playoff berth, but I did want to build off of Eric’s point about teams having different marginal win/dollar rates at which a contract makes sense.

Intuitively, everyone knows this is true – a contract that makes sense for the Yankees doesn’t necessarily make sense for the Rays, simply due to the massive differences in payroll. Alex Rodriguez can be an asset for New York while he’d simultaneously be a liability for other teams simply due to the context of the specific organizations.

However, I don’t know that I’ve ever seen anyone break down the marginal win/dollar rates for each franchise, based on projected payroll, so that’s what we’ll do this afternoon. For those unfamiliar with the marginal win/dollar metrics, they were first made famous by the late Doug Pappas, and his work has been continued on by countless others. The concept is built around the belief that a team of replacement level, freely available players would finish with something like 50 wins, and due to their freely available nature, they shouldn’t cost any more than the league minimum.

Basically, we’re saying that a team that isn’t trying to contend could win 50 games on a payroll of about $12 million, which assumes a $400,000 contract for 30 players. That team would be paying about $240,000 per absolute win, and since they’re the baseline we’re building off of, obviously they wouldn’t be spending any marginal money or accumulating any marginal wins.

To apply this to a team from last year, the Seattle Mariners spent $117 million and won just 61 games. In other words, they spent $105 million to win an extra 11 games over what they could have won without even trying to contend, so the Mariners spent $9.55 million per marginal win. That’s bad. Really bad. A team with that marginal win/dollar rate would have needed a payroll of about $382 million in order to be a 90 win team, which would put them in contention while not guaranteeing a playoff spot.

Hopefully, you have a pretty good grasp of marginal win/dollar rates now. Now, using 2008 team payrolls, adding a 5% markup to adjust for modest salary inflation, and assuming that every team’s target should be 40 marginal wins added, here are the team specific dollar per win rates for 2009 – obviously, for some teams like Seattle, these won’t reflect a change in direction from potential contender to rebuilding.

The Yankees can spend $5 million per win across their entire roster. Remember, we’ve been saying that wins on the free agent market cost just over $5 million per win – if there were enough free agents available, the Yankees have a large enough payroll to build a contender from scratch just by signing free agents. Once you add in the presence of cost-controlled players like Joba Chamberlain, Chien-Ming Wang, and Robinson Cano, where they’re paying quite a bit less than $5 million per win for players developed from within, and the Yankees can actually justify free agent contracts up to $7 or $8 million per win. They can spend $15 or $16 million on a league average player in order to fill out the roster and have it not be a real problem. That’s a remarkable financial advantage.

On the other end of the spectrum, you see teams like Tampa, Florida, and Oakland, who have to try to contend while spending less than $1 million per win. The only way to do that is to build from within, which is why these organizations covet major league players in their 1st-3rd years of service time – you can get huge production for no money, which is absolutely vital to trying to build a 90 win team for less than $1 million per marginal win.

These payroll specific dollar per win rates are a good step in the right direction to evaluating whether a contract makes sense for a specific team. However, there’s still a pretty huge missing piece of information here, and that’s the amount of wins currently on a roster. A team needs to understand how many wins they really have to add and how much money they have available in order to evaluate a specific move, and we’ve only calculated the latter half in this post. In a future post, we’ll take the next step and factor in the wins already in place.

Dave is the Managing Editor of FanGraphs.

Newest Most Voted
Inline Feedbacks
View all comments
14 years ago

I would like to see the list with projected wins. Whether that be PECOTA or using the fielding bible and wOBA on an entire roster to compute the theoretical wins.

From the list you can easily see a theory in which a team like the royals can trade Meche up the list to the Yankees and have both sides come out winners. (that is if the Yankees needed a SP). It also provides a point at which a GM should reject a trade as the player they are receving costs too much per win (Royals w/ Jacobs).