Every so often, we are fortunate to have one of our readers provide some great content for the site. Dave Laidig is an attorney in Minnesota, currently supporting a multi-national corporation as contracts counsel. Dave coordinates sales, marketing, finance and technical departments for an information content provider in the pursuit of business opportunities, and serves as the authorized negotiator for contracts with government entities. He earned a Master’s degree in Psychology with an emphasis on research methods and statistics; and undergraduate degrees in Psychology and Sociology. He has presented seminars and published articles in the field of government contracts, but prefers to discuss the inner workings of the business of sports to those who will listen. Thanks to Dave for his great insights….we are listening!
“You cannot say you are happy when you didn’t win.” Arsene Wenger
On pitches across America, and across the globe, the implicit assumption is that the purpose of taking the field is to win. Otherwise, why compete? And the universal desire to win creates a demand for the best players, raising their salaries. With an additional assumption of an efficient labor market, team salaries should reflect the relative talent levels on each team. Consequently, teams with larger payrolls should have more wins than those with lesser payrolls.
However, over the last four years, MLS team salaries have not been related to regular-season wins (or points for that matter). Specifically, the correlation between team salary and wins is a tiny .09 and is not statistically significant. Viewed another way, team salaries explain only 1% of the variation in wins, leaving the other 99% to other unidentified factors. Using the same data to create a regression equation, we learn that a team gets an additional win for every $10 Million in team salary. Interestingly, although still not significant in a statistical sense, the number of salaried players has a greater influence on wins than the teams’ salaries (correlation of .14). Regardless of method, increased spending by MLS teams on salaries has not lead to a corresponding increase in league wins.
Hopefully, one recent trend worth noting is that the correlation between salaries and wins has been increasing (-.199, -.283, .000, .560 over the last four years). Because of the small sample sizes, one should not place too much emphasis on the 2010 data as it easily could be a fluke year. It is possible that teams may be making better choices, but verifying this would require a few more years with similar results. But analyzed in isolation, the 2010 salaries were significantly related to wins and accounted for about 30% of the variance in wins. For 2010, an additional win cost $2.5 Million in salary, an increase in efficiency over the historical trend.
Because the overall weight of the data does not show a relationship between team salaries and increased wins, perhaps a more useful tool for management is a much simpler concept than a regression equation. Over the last four years, MLS teams average about $334,000 per win, or about $87,000 per point. For those teams spending more per win, one should be able to justify the added expense through increased revenue.
The big question is why MLS salaries were not related to victories over the last several years. One could simply assert it’s an artifact of a relatively new and growing league with resources directed to capital investments and marketing. One can identify the salary cap, even with designated player exemptions, as inducing teams to adopt similar labor acquisition strategies due to equally limited resources. Alternately, as an ironic twist, perhaps the high-priced players are called away to international duty so frequently that they do not make the impact they should. Or perhaps, at the salary range that MLS teams look for, the talent is relatively uniform. But because the increase in salaries has not necessarily lead to a competitive advantage, rational team managers must have other justifications for exceeding league averages for salary; higher ticket sales, attracting more business partners, or as a last resort – a novel approach to player valuation. Failure to obtain a sufficient return may start to affect front-office employment rates.
 Correlations are reported on a scale of -1 to 1, with zero meaning no relationship and 1 and -1 meaning a perfect relationship. A positive number means that as the predictor variable increases, the result increases (i.e., the variable move in the same direction). A negative number means that as the predictor variable increases, the result decreases and vice versa. Thus, a correlation of .6 and -.6 are equally strong, but show different types of connections between variables.
 And what’s a designated player or two between friends?
 About two-thirds of the teams have total salaries that fall within +/- 15% of the league’s median, not a whole lot of variation.