Designated HitterApril 13, 2006
Can Money Buy Love in Baseball?
By David J. Berri

One of the prizes in baseball's free agent market was Johnny Damon. Expected to re-sign with the Boston Red Sox, Damon shocked the Red Sox nation when he left the team he helped win a title in 2004. What was even more shocking was his destination, the dreaded New York Yankees. Note only did Damon break the hearts of the Red Sox faithful, for the extra millions the Yankees paid, Mr. Damon was required to dramatically alter his personal appearance, cutting his trade-mark hair and shaving his beard to join the men in pinstripes. The Damon story suggests that money can buy looks. But can all that money buy another World Title for the Bronx Bombers?

We are told over and over again that money can indeed buy the fans love in baseball. Supposedly teams that spend the most win the most. As is so often the case, the numbers tell a different tale.

Our story begins in 2001. That year the Yankees, Red Sox, and Los Angeles Dodgers became the first teams in Major League Baseball history to spend more than $100 million on player salaries in a single season. The following year the Yankees pulled away from the pack, becoming the first team to clear the $125 million barrier. In 2003 the spending continued with New York's payroll surpassing the $150 million mark. The next season another record was set, with Yankee salaries rising above $180 million. In 2005 team payroll actually cleared the $200 million barrier. Among all teams excepting the Yankees, player payroll has increased only 7% in the last five seasons. Over the same period the Yankees have increased payroll by more than 85%.

Year Yankee Payroll Yankee Wins
2005 $208,306,817 95
2004 $184,193,950 101
2003 $152,749,814 101
2002 $125,928,583 103
2001 $112,287,143 95
TOTALS $783,466,307 495

Source: Payroll Data is from USA

Over these past five years the Yankees have spent more than $780 million on player salaries. What did all this spending buy? The Yankees have won more regular season games than any other team in Major League Baseball. New York has averaged 99 wins per season, while the next berst teams, the Oakland Athletics and the St. Louis Cardinals, have averaged 96. Yes, over five years, those $780 million Yankees have only won fifteen more games than their closest competitors.

This neck-and-neck competition on the field should have been reflected in the payrolls of each team. The numbers defy this expectation. Over these five years Oakland has spent less than $240 million on player salary, or less than a third of the Yankees payroll during the same period. St. Louis has been a bit more spendthrift, paying its players more than $410 million. Still, St. Louis and Oakland combined have spent over $100 million less to achieve almost as many wins as the Yankees.

What else did the Yankees gain from all this spending? They did manage to win the same number of World Series titles as Oakland and St. Louis combined: zero. A massive payroll, more massive every year, and in each season the last game the Yankees played ended in a loss. To be fair, the Yankees are not alone in their failure to buy a title. Of the eighteen teams in baseball history that have spent more than $100 million on players in a given season, only the 2004 Boston Red Sox managed to win the World Series.

This lack of success leads us to wonder, given the amount of money the Yankees are spending, why have the Bronx Bombers not won the World Series recently?

Contrary to popular perception, payroll in professional sports is not strongly linked to wins. A $100 million team does not win twice as many games as a $50 million team - not even close. Our own work has shown that only about 18% of a team's regular season wins can be attributed to its payroll. In other words, more than 80% of a team's regular season record cannot be tied to team spending. We would add that this is what we see when we look at teams in Major League Baseball from 1988 to 2005. In other words, the lack of a link between spending and wins is not a recent phenomenon. Across time more spending is not an elixir that leads automatically to success on the field. As the saying goes, games are not won on paper. Moreover, they are not won just because you spent a pile of paper.

The Yankees, however, both spend and win. The Yankees have consistently placed near the top of the league in regular season wins and at the very top in payroll. But near the top is not actually ascending to the ultimate height, a point principal owner George Steinbrenner often notes when he frets about his team's lack of a World Series title over the past five years.

Steinbrenner should well fret. A number of factors can conspire to bring down even a team packed with all-stars. The three most salient roadblocks are: performance inconsistency, the ravages of time, and luck.

Baseball is a game of inherent inconsistency. Remember, we are talking about grown men hitting a round ball with a round bat. Even the very best hitters are unsuccessful more often than not. And even the best cannot simply create success whenever they like. Take the Yankees most recent addition, Johnny Damon.

Year Damon's OPS Ave CF OPS Difference
2005 0.805 0.776 0.029
2004 0.857 0.816 0.041
2003 0.750 0.799 -0.049
2002 0.799 0.805 -0.006
2001 0.687 0.781 -0.094
2000 0.877 0.807 0.070
1999 0.856 0.799 0.057
1998 0.779 0.779 0.000
1997 0.723 0.798 -0.075
1996 0.680 0.819 -0.139
Ave 0.781 0.798 -0.017

Note: OPS stands for on-base percentage plus slugging percentage.

Hitting performance can be measured many different ways. When measured via OPS, the Damon story illustrates our story of performance inconsistency. Some years, like 1998, 1999, 2000, 2004, and 2005, Damon's performance at the plate exceeded the average hitter at his position. In other years, like 1996, 1997, 2001, 2002, and 2003 Damon performed below the average centerfielder. The Yankees will be paying Damon $52 million over the course of four years - and hope he will be above-average each season. Yet Damon has never had a four-year span of consistently above-average play, and there is no reason to expect he will have one now. Given his past performance, at least one of these four campaigns will likely be below average for the Yankees' new centerfielder.

Damon's career performance is not anomalous. Our analysis indicates that less than 40% of player performance in any given season can be attributed to what a player did the previous season. That is past success - or past failure - is a weak predictor of future outcomes. Hitting success will naturally fluctuate - due to chance, injury, schedule, diet, and perhaps hundreds of other factors - just as weather fluctuates in July. July may be hot and sunny on average, but any given day can be rainy and chilly. Neither Damon or his teammates has demonstrated the ability to overcome the reality of statistical variation, any more than anyone can guarantee that everyday at the ballpark in July will be sunny and warm.

Inconsistency is not the only challenge facing the Yankees; the passage of time is an enemy as well. The Yankees management has tended over the past half-decade to sign or trade for well-established players with strong reputations. Unfortunately, the nature of free agent rules has resulted in a Yankees line-up replete with players over the age of 30. Yale economist Ray Fair demonstrated in a recent study entitled "Estimated Age Effects in Baseball" that the performance of baseball players will peak at age 28 and slowly decline thereafter. Of the players projected to start for the Yankees on opening day in 2006, only Robinson Cano is under 30. The remaining players have crested the statistical hill and will all tend to decline rather than improve on their previous feats.

Still, the end is not so close that the Yankees roster is ready for retirement. Over the last five years the Yankees have employed older players and still crafted the best winning percentage in baseball. The World Series title, though, has proven elusive. Hence we come to our third road block to success. Beyond inconsistency and time, teams must also contend with luck.

Year WS Champion Season Wins ML Wins rank
2005 Chicago White Sox 99 2nd
2004 Boston 98 3rd
2003 Florida 91 7th
2002 LA Angels 99 4th
2001 Arizona 91 Tied-6th
2000 New York Yankees 87 9th
1999 New York Yankees 98 3rd
1998 New York Yankees 114 1st
1997 Florida 92 4th
1996 New York Yankees 92 3rd

How often has the team with the best record in the regular season won the World Series title? In the past ten years, this has happened once. In 1998 the Yankees led Major League Baseball in regular season victories and also won the last game they played. In 2000 the Yankees posted the 9th best record in baseball and took home the trophy. Although each Yankee team since has outperformed the 2000 edition in the regular season, the parade each year was held someplace else. As the Yankees learned in 2002 and 2003, being the very best from April to September does not guarantee happy days in October.

Certainly a student of Yankee history would see a different story. In the forty years beginning in 1923 and ending in 1962 the Yankees took the World Series title twenty times. In sixteen title years, the Yankees also posted the best record in Major League Baseball.

In the 1960s the frequency of the best regular season team capturing the crown changed dramatically. Part of this change can be explained by how many teams are allowed to participate in baseball's post-season. Until 1969 the World Series was played by the best teams in the American and National League. In other words, only two teams played in baseball's post-season. In 1969 baseball introduced two divisions in each league and post-season participation was expanded to four teams. With four teams participating the fortunes of the best team declined. Across 25 seasons, beginning in 1969 and ending in 1993, the team with the best record won the title in seven years, or 28% of the time.

After the 1994-95 labor dispute was resolved, baseball began allowing eight teams to participate in the post-season. Now the path to the title required the champion to win three different playoff series. Perhaps it is not surprising that in the years since only the 1998 Yankees managed to win the most regular and post-season contests in the same campaign.

It is important to note that post-season expansions are not the whole story. The difference between the best and the rest has changed since the days of Babe Ruth and Lou Gehrig. In the days of Ruth, baseball players were primarily white players from the Eastern United States. The integration of baseball, coupled with the willingness of teams to search the world for playing talent, has greatly expanded the population of available talent.

What does this mean for the Yankees? The Yankees may be able to acquire the best free agent talent available, but no team can buy all the best players - there are just too many of them to fit on a single roster. No matter how large the Yankee payroll, the opposing teams will also have good players, especially in the playoffs when the weakest teams have been eliminated. In the end, when two good teams face-off the outcome can be more about luck than about skill. And with the playoffs expanded to three rounds, winning a title in the 21st century might require more luck than ever before. Unfortunately for the Yankees, luck is the one thing money cannot buy.

Even without a crystal ball, death, taxes, and the uncertainty of baseball can reasonably be predicted. The natural inconsistency of player performance, declining productivity of aging veterans, increases in the league-wide level of talent, and the unpredictability associated with a long post-season, all conspire to rob George Steinbrenner of the World Series titles he keeps trying to buy. So although we are sure the Yankees will lead Major League Baseball in payroll in 2006, a payroll title does not guarantee a parade in October. And Damon may look better in 2006. But just like he did in 2005, he may still lose the last game he plays in 2006.

David Berri, Martin Schmidt, and Stacey Brook are authors of The Wages of Wins: Taking Measure of the Many Myths in Modern Sports (Stanford University Press), forthcoming in May, 2006.


I'm suprised the idea of the team that spends the most money should win it all is still prevelant. Spending the most money has never been a guarantee of success-- but it sure helps. This point is made quite clear in "Winners", Dayn Perry's book, which I recently finished reading. Well-written, entertaining, and informative. The book supported most of what I thought went into winning teams in the first place. Good article.

Where does the 18% come from?

"Our own work has shown that only about 18% of a team's regular season wins can be attributed to its payroll."

The age factor may be paramount. In mid-season, 2005, the Yankees were carrying 14 players born in the 1960's, on their 25-man roster. The pitching staff averaged 35, position players and the bench averaged 34, for an overall average of 34 1/2 = far removed from the peak-perfomance years in the late 20's.

The chart above shows the "LA Angeles" winning the 2002 World Series, when in fact they were known as the Anaheim Angels at that time, and will be again when Public Relations dolt Arte Moreno goes the way of the dinosaur.

Hmmm. In your table of last 10 WS champions, another column called Payroll Ranking would probably indicate their expenditures in the top quartile of all teams each year. The highest payroll may not guarantee a champ, but relatively high spending probably does.

Thanks for the comments. Just a quick response from the book: In chapter three of the Wages of Wins we detail the link between wins and relative payroll. Estimating this simply model reveals that relative payroll explains 17.6% of wins from 1988 to 2005 (In response to Cyril, this is the r-squared). If you look at this from 1988 to 1994, relative payroll explains 6.2%. From 1995-99, which we call the Blue Ribbon Panel years, the explanatory power is 32.5%. After this time period, though, from 2000 to 2005, explanatory power falls again to 17.6%. We take this to mean that payroll alone does not explain much of wins. I would add that the low explanatory power of payroll is also seen in the NFL and NBA. This is a topic we discuss in some detail in our forthcoming book.

I don't much care for articles like these, which brush aside the spending inequalities in baseball. Come on. It's a joke. Teams like the Los Angeles Angels and so-on are referred to as "high payroll" teams when they have like half the payroll of the Yankees. Sooner or later, the Yankees are going to rattle off 2 or 3 World Series victories in a row and these articles will have to cease. The Yankees simply spend their money poorly.

Teams like the Yankees have a margin for error that other teams cannot begin to dream about. Look at what happened with Mussina, Wright, and Pavano last season. What would happen to to the A's if that fate befell Harden, Blanton, and Zito? Or the Angels with Colon, Lackey, and Byrd?

A lot of the problems come from the Yankees spending some of their money unwisely. Just wait until some prime free agents come on the market and the Yankees lose contracts like Mussina's.

Somebody ask the 8-1 Mets if they feel money hasn't bought them anything...


When you ran the regression was each team/season an observation, so that you had a couple hundred or so? Or did you try one where you had each team's average annual relative payroll and their average annual winning percentage? So in that last regression you would only get 30 observations.

The regression was run as you first suggested, with team/season as the unit of observation. So from 1988 to 2005 we had 510 observations.

Okay, thanks. I wonder if it would be worth it to try the second way, if that might smooth out some of the year to year randomness. Also, I wonder if a separate regression could be done for each league. Maybe how much a team spends should only be compared to the teams it plays against (although now there are some inter league games). But maybe a regression with just 14 or 16 observations won't mean much, though. I really don't know.

I think the key point in the essay I posted was "Our analysis indicates that less than 40% of player performance in any given season can be attributed to what a player did the previous season. That is past success - or past failure - is a weak predictor of future outcomes." Because performance from season to season is not easily predicted, knowing which players to buy next season is difficult. If you add to this point the randomness associated with post-season performance, buying a championship becomes even more unlikely. At least, that is the story I think the data is telling.