In these times of economic difficulty, shrinking player payrolls, and elephantine superstar salaries, all it takes for a team to handicap its ability to add salary to address a need-say that key hitter that will put the club over the top in its divisional race-or even to shed salary to lessen the blow of a lost season, is one key mistake. As in all businesses, getting more for less is the name of the game, but baseball has an added dimension of difficulty for teams: they need to maintain flexibility while dealing with long-term guaranteed contracts. Some clubs are very good at managing their payrolls to meet these imperatives, while others stumble.

Maintaining payroll flexibility is much like the venerable Magic 15 puzzle, in which one tries to order numbered tiles while moving them around a constricted space. Once clubs have defined that space in terms of what they consider an affordable payroll, they have to weigh the value of each piece of added salary very carefully. For a team with an $80 million payroll-and 16 teams spent less than that in 2008-adding a $15 or $20 million superstar carries very high risk. That player is going to take up a fifth to a quarter of the overall payroll, leaving very little breathing room for other star-level players, and despite players legendarily “carrying” teams, no team has won a pennant with only as much quality as a single 10-win player can provide. Last year, the defending NL pennant-winning Rockies spent $68.7 million on salaries, nearly one quarter of which went to Todd Helton. Helton got hurt and contributed less than one win above replacement, and the rest of the club, starved due in large part to his massive salary, lacked the ability to compensate.

The poster child for payroll inefficiency and inflexibility may be the Houston Astros, with a solemn nod in the direction of the Seattle Mariners, who last season spent $117.7 million to win only 61 games. Houston’s problem is one of having a few too many jumbo eggs in a very small basket. Their 2008 Opening Day payroll, above average at $88,930,414, has contracted a bit with the deletion of Ty Wigginton, Brad Ausmus, and a few of their cheaper players, but remains inflated by several immovable objects. Four players dominate Houston’s payroll: first baseman Lance Berkman, who made $14.5 million in 2008 and will do so again in 2009; shortstop Miguel Tejada, whose $13 million 2008 salary will carry over to this season; starting pitcher Roy Oswalt, who will see his $13 million 2008 salary rise to $14 million this year; and left fielder Carlos Lee, who made $12 million last year but will see his salary jump to $18.5 million this year and remain there until 2012. That latter figure for Lee would have ranked as the sixth-highest salary in baseball in 2008. These four players represented nearly 60 percent of Houston’s payroll last year and will consume and even larger slice of the pie this season.

While the rapidly declining Tejada’s contract expires with the 2009 season, Berkman, Oswalt, and Lee, all players on the wrong side of 30, will remain under contract-Berkman through 2010 or 2011, depending on whether his option is exercised, Oswalt through 2011 or 2012, depending on his option, and Lee through 2012. Of these, it’s Lee who represents a particular problem. Though he was having a career year before a Bronson Arroyo pitch shattered his pinky on August 9 (forcing him to the DL for the rest of the season), his previously established career rates of .288/.342/.499 fall squarely in the good-not-great category, and even those numbers are severely compromised by his weight-hampered defense. In an age of contracting payrolls, Lee will remain an unshrinkable mass for the Astros through his 36th birthday, doing less and less while taking up more and more payroll space. For a team operating under small- or medium-market constraints, this is a huge handicap.

The Astros have averaged almost $102 million in player salaries over the past three years while winning an average of just 80 games. The other end of the payroll efficiency scale is dominated by the likes of the Twins, Indians, and Angels. Over the last three years, the AL Central rival Cleveland and Minnesota teams have averaged about $67 million in payroll each, while averaging 85 and 88 wins per team, respectively. This is even more impressive in the light of the bad contracts both teams have to live with. In Cleveland’s case, that would be the four-year, $57 million contract the team signed with designated hitter Travis Hafner in July of 2007, one which cost them $8.05 million last season and will ding them for $11.5 million this year. The Twins have made smaller-scale errors, anchoring themselves to replaceable talents like Michael Cuddyer and Jason Kubel through 2010 for a combined total of $22.1 million, not counting the price of 2011 options for both players. The Twins will also be paying $3.1 million to Milwaukee Brewer Mike Lamb this year.

The Twins and the Indians have made just two post-season appearances between them over that span, but the Angels have gone to the big dance in each of the last two seasons. Their 2008 Opening Day payroll ranked just fifth in the AL, well behind that of perennial spenders Boston and New York. They’ve also ranked sixth in the majors over the last three years, and yet have averaged more wins than any other team, having won 89, 94, and 100 games from 2006-2008.

Perhaps, though, the most efficient team in baseball last year was the Rays, with their $43.8 million opening-day player payroll, lowest in the AL, their 97 wins, and a World Series appearance. (The Marlins were even more efficient than the Rays, posting a .522 winning percentage with a payroll of just under $22 million, but this was more a fluke event than an item for future study.) The Rays showed that intelligent drafting and strong farm system production can provide an inexpensive roster base which can be augmented with intelligently spent free-agent dollars. The true test of this method will be their ability to maintain a relatively low payroll as success and experience push their players up the salary scale.

In 2002, Baseball Prospectus’s Doug Pappas proposed the concept of marginal dollars per marginal wins. Pappas suggested that a team playing entirely replacement-level players would win about 30 percent of its games, or 48.6 wins in a 162-game season. This team would be paid the major league minimum ($390,000 in 2008) and was assumed to have a 25-man roster, plus three players on the disabled list. In 2008, that team’s payroll would have been $10,920,000. Every dollar a team spends over that minimum is more than it “has” to, but those marginal dollars do also earn the team marginal wins-victories they would not have earned had they only spent the minimum. The question is, how efficiently did the teams spend to get those extra wins? Here are the marginal dollars per marginal wins figures for the majors in 2008:

Team        EstPayroll     MargPayroll   WPct   MargWins    M$/MW
Yankees    $207,896,789   $196,976,789   .549    40.3     $4,883,157
Red Sox    $133,390,035   $122,470,035   .586    46.3     $2,643,314
Blue Jays   $97,793,900    $86,873,900   .531    37.4     $2,321,466
Orioles     $67,196,246    $56,276,246   .422    19.8     $2,847,412
Rays        $43,820,597    $32,900,597   .599    48.4       $679,231

Team        EstPayroll     MargPayroll   WPct   MargWins    M$/MW
Tigers     $137,685,196   $126,765,196   .457    25.4     $4,984,084
White Sox  $121,189,332   $110,269,332   .546    39.9     $2,766,971
Indians     $78,970,066    $68,050,066   .500    32.4     $2,100,311
Royals      $58,245,500    $47,325,500   .463    26.4     $1,792,225
Twins       $56,932,766    $46,012,766   .540    38.9     $1,183,456

Team        EstPayroll     MargPayroll   WPct   MargWins    M$/MW
Angels     $119,216,333   $108,296,333   .617    51.4     $2,108,820
Mariners   $117,666,482   $106,746,482   .377    12.5     $8,557,518
Rangers     $67,712,326    $56,792,326   .488    30.5     $1,864,734
Athletics   $47,967,126    $37,047,126   .466    26.9     $1,377,626

Team        EstPayroll     MargPayroll   WPct   MargWins    M$/MW
Mets       $137,793,376   $126,873,376   .549    40.3     $3,145,257
Braves     $102,365,683    $91,445,683   .444    23.3     $3,919,997
Phillies    $97,879,880    $86,959,880   .568    43.4     $2,002,945
Nationals   $54,961,000    $44,041,000   .366    10.7     $4,119,061
Marlins     $21,811,500    $10,891,500   .522    35.9       $302,845

Team        EstPayroll     MargPayroll   WPct   MargWins    M$/MW
Cubs       $118,345,833   $107,425,833   .602    48.9     $2,195,770
Cardinals   $99,624,449    $88,704,449   .531    37.4     $2,370,382
Astros      $88,930,414    $78,010,414   .534    37.9     $2,057,888
Brewers     $80,937,499    $70,017,499   .556    41.5     $1,688,308
Reds        $74,117,695    $63,197,695   .457    25.4     $2,484,772
Pirates     $48,689,783    $37,769,783   .414    18.5     $2,045,147

Team        EstPayroll     MargPayroll   WPct   MargWins    M$/MW
Dodgers    $118,588,536   $107,668,536   .519    35.5     $3,034,797
Giants      $76,594,500    $65,674,500   .444    23.3     $2,815,265
Padres      $73,677,616    $62,757,616   .389    14.4     $4,352,727
Rockies     $68,655,500    $57,735,500   .457    25.4     $2,270,013
D'backs     $66,202,712    $55,282,712   .506    33.4     $1,656,560

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
Interesting analysis, but I have a problem with your methodology. The way you have framed it, the 60th win counts just as much for a win above the "margin" as the 90th win does. This is obviously invalid, both from a competitive standpoint and from a marketing one -- the crowds ain't gonna come any more enthusiastically for a team that wins 60 games than for one that wins your replacement-level 48.6.

How does this analysis change if, say, wins above 80 count double, and wins above 90 triple, or something like that? Because those are the wins that it's really worth spending payroll bucks to achieve.
There is some small value to any win. Including from 49 to 50. But the gentleman's overall argument is entirely valid.

More to the point, you guys do have a table somewhere showing the marginal value of wins, don't you? How about meshing that with the data in this article?
You're right, of course, and that's why in BP's book "Baseball Between the Numbers" Nate Silver reconceptualized the metric to take into consideration how close a team was to the "sweet spot" in which spending a relatively few more bucks might produce a very substantial payoff by putting a team over the hump and into the playoffs.
I don't know if there is a way around this, but using the payroll for each team may not adequately account for the internal accounting of the team. For instance, the Braves payroll last year is listed at just north or $102 million, but a large portion of this was Mike Hampton's 2008 salary, which was not reflective of the amoritzation that was done on Hampton's salary when the Braves acquired him. Based on that, the team was amortizing the contract for about $8 million a year, including the years that the Rockies and Marlins were paying most of his salary. With this method, his entire $15 million shows up for 2008, not the $8 million the Braves were actually accounting for. Not to mention the insurance money that was received for the times he was on the disabled list. This is still an informative chart, I just don't think enough information is available to make it an accurate representation, unless it's done over a 3-5 year period (offsetting the some of the internal accounting effects).
As far as I know, BP does not post marginals wins/$$$ anywhere in particular on the site. It just pops up in articles every so often. I've emailed BP several times over the past few years requesting that the data be presented for each year, and have been given the typical "oh, that's a good idea, we'll put that on the list" response. PLEASE BP, make this info available. As harpago17 points out, spread out over time, this is a good tool and really provides a great metric for measuring front office performance. Based on just a single year, it doesn't tell you nearly as much.
Hmmm. I must be doing something wrong, as I don't even get the "we'll put that on the list" response to my inquiries. Hmmm. I wonder which of the many things I do wrong it is?
Yeah, that was actually back in the good ol' days when someone would actually respond. These days I get nothing at all.
While the inflation-adjusted numbers are not posted here, the concepts that many of you are discussing are considered at much greater length in "Baseball Between The Numbers" ( ), which is an excellent read (really, I don't work for BP).
So that's it! They won't come out with the temperature/runs-scored data 'cuz they want me to instead shell out the money for whichever book it's in! And they're bright enough not to mention that to me 'cuz they know how cheap I am. Darn that intelligence of their's!
Great read though. It's like my baseball bible. I can always review it when I need to better imform a friend about the game. I probably read it as much as my The Bible, if not more.

I also want to say, even if the Marlins were better than their Pythagorean record, isn't it still amazing what they (and the Rays) are able to do with such a small payroll!
I wouldn't have a problem with calling the Rays something like "the most successfully efficient" team, as there can certainly be some debate over whether there is much to be by being, say, the best team to miss the playoffs and the worst in baseball, whereas the revenue from simply making the playoffs, even if you're swept in the opening round, makes it demonstrably better than being the best team to miss. But if you're measuring all 30 teams and defining efficiency by M$/MW, it seems a bit out of hand to then dismiss one result as a "fluke" when it doesn't jibe with the desired results or accompanying analysis.
That should, of course, be "...debate over whether there is much to be *gained* by being..."
Considering the Marlins have been perennial overachievers even with the way Loria skimps on payroll, isn't it about time they are dismissed as a fluke? They've done a good job at fielding a competitve team for years without getting any #1 overall picks. Contrast that to the Pirates who don't spend much, still get #1 picks, and still can't break .500.
This is a useful metric for efficiency. I think a more useful metric is how is the team efficient. What are the tiers of players salary? I think: fully controlled (they get the minimum), arbitration eligible and FA.

The first two are a function of drafting and trades, while the latter is FA selection. The reason I would like to see it separated is that the marginal value per win implies you can go buy that win if you wanted to. Well, if the $ in the marginal win are not available because the players are controlled by the teams, you cannot access that talent.

So really need to see the marginal value of FA VORP. That, theoretically, is the talent you can go buy. The other stuff you have to draft and hope it pans out.

And previous posters comments that the value of marginal wins when you have a chance to get to the playoffs are >> greater than generic marginal wins. The Yankees spending probably makes more sense. They are not trying to win marginal games, but trying to consistently win 90+ games. How much do you need to have a high probability of that? Apparently a lot.
I can't imagine a $10.9-million squad winning 49 games on any sort of regular basis; the pitching, in particular, would not hold up. But, beyond that, the entire concept is flawed and the results irrelevant. Neither a $10.9M payroll nor 49 wins is a feasible approach on which to base or maintain a successful MLB franchise. Marginal dollars per marginal wins might have some use as an indicator if the bar were set at the average wins and average payroll of those teams below .500.
Steve: Many thanks for a highly interesting article. I'm in my own way (as a Yakee fan!) as sceptical of this as a diagnostic as anyone, but you have got the issues out there and being intelligently (for the most part) discussed. Long term thought experiment to set for myself: Pick a payroll number- probably at the MLB midpoint. Assemble a group of VORP neutral core players (sort of like the way MLB "stocked" an expansion team?). As the seasons turn, assign yourself theoretical draft picks. Never offer arbitration, never sign a free agent, live with what the "core", the draft and development do for you. If you were the type to be entirely candid with your players, advise them to assume they would never remain with the club beyond their arb-eligible or FA date; you will never be able to afford them if they are still on your roster at those dates. And see what happens. I don't think this is the Rays, or the Marlins- it's a thought experiment, but I wonder what you think about the utility of playing it out?

Love my BP.
"The Rays showed that intelligent drafting and strong farm system production can provide an inexpensive roster base which can be augmented with intelligently spent free-agent dollars."

Yeah, after having the benefit of the first draft pick (or close to it) becuase you've been God-awful for 10 years. Lets get real here.