The "Baseball Economist" Answers Your Questions
We recently solicited your questions for “baseball economist” J.C. Bradbury, author of the new book Hot Stove Economics. His responses show great range. The most fascinating answer, in response to a question about the agent Scott Boras‘s dominating performance: “I have a theory that Boras sells his own insurance to players by promising players a minimum salary in return for waiting for free agency. This way, players get insurance against injury, more income if they reach free agency in good health, and Boras gets a bigger cut.” Thanks to J.C. and all of you for playing along.
The highest-paid managers get paid multiples less than the highest-paid players. J.C., do you come up with estimates on the value of managers? More specifically, if you were the agent for the best manager in MLB (whomever you think that is), could you argue that manager pay should be doubled? – Barbara
If I had added another chapter to the book, it would have been on the value of managers. In fact, my most recent project involved quantifying the impact that managers have on players, looking at how player performance changed under different managers. Though I found a few managers who had “statistically significant” effects on their players (positive and negative), no manager helped both hitters and pitchers. I estimated the impact of several factors on performance, and when I compared models that included managers versus those that did not, there was very little impact. In summary, managers appear to have very little effect on baseball player performance, and this really shouldn’t be surprising given that most of baseball involves one-on-one contests between hitters and pitchers.
This then leads to the question: why do some teams pay managers big salaries if they are not important? I have a theory that teams use managers to signal to fans that they are improving. If your team is struggling, you can’t quickly replace a roster of 25 guys and improve overnight. Managers can be quickly replaced; thus, if fans perceive that managers do impact performance – even if they don’t – then teams can falsely signal improvement to fans. Fans who would otherwise stay at home now go to the ballpark. I tested this hypothesis for the past three decades by examining how attendance changed after replacing a manager. In the 2000s, I found that managerial replacements led to an increased attendance of about 1,000 fans per game. I did not observe similar effects in the 1990s and 1980s, though. It’s unclear why the effect is observed more recently, but not in the past. Maybe the growth in fandom in the present has produced more sensitive fans, but it’s hard to know. You can read my working paper here.
How should a college professor’s talent/value be calculated and quantified? Which metrics would you use? Do you bring these things up at your annual review? – Lenny
What a delicious question! I believe this was something that the Texas A&M system has been trying to do, and Daniel Hamermesh was critical of the measurements that the system was using to value professors. If you pick the wrong performance metrics, then you’re going to produce incentives that cause professors to pursue sub-optimal activities. For baseball teams, it’s fairly simple to assume that team owners are profit maximizers, so identifying qualities that lead to winning, which increases revenue, are easy to identify. In academia, what is it that universities are trying to maximize? Prestige, donations, spreading knowledge, etc.? And in many cases, it’s hard to quantify exactly what professors are bringing to a university quickly. What if a professor spends years researching innovative methods that revolutionize his/her discipline. He/she might receive some grants and prizes along the way, but down the road that professor may produce many wealthy alumni who give back generously many years after graduation, possibly after the professor has already died. Capturing this value is difficult.
Annual reviews do include metrics such as publications (quantity and quality), teaching evaluations, direct service contributions, etc. Some schools have very detailed criteria (for example, a publication in American Economic Review is worth a 3 percent raise), while some schools have subjective evaluations by department chairs and deans who decide how much to allocate out of a pool of funds based on performance. Even though the criteria are not always explicit, department chairs have an incentive to keep good faculty in place, and therefore reward faculty who do things that cause them to get hired away. Good teaching may be important, but it’s hard to signal to outside institutions. So, a professor who is publishing in top journals will likely get better raises. And ultimately, everyone knows that the best way to get a raise in academia is to shop yourself on the market and get an outside offer. The downside of allowing the collective wisdom of the outside market to fully govern compensation is that faculty devote resources to applying for jobs at the expense of devoting more resources internally.
I’m having more and more trouble nowadays discussing baseball with my cousin, as he’s the type who would hear the words “valuation model” and immediate attempt to dismiss any of the following points. I’m sure you encounter this too. How do you react to this? – Katie
Well, I understand this reaction, and though I think about baseball a little more intensely than most of my baseball-fan friends, I find most people are just as interested in the same issues I am. The difference is the language that stat-savvy (or sabermetric) fans often try to force on others. While I use terms like marginal revenue product and OPS in my own work, I also try and explain these concepts in a familiar lexicon. For example, let’s say your grandfather complains about Adam Dunn‘s .250 batting average, high strikeouts and horrible defense. If you respond, well he’s got a .900 OPS, which more than makes up for his strikeouts and bad defense, and his MRP is $12 million, he’ll spit tobacco in your eye (or at least mine would). Instead, you could say, “Well, while he doesn’t hit for average, he gets on base a lot with walks, and when he does get on he normally gets further than first base. When you look at all the things he does and how that translates into winning, and the revenue that teams get from winning, his worth is about $12 million to the bottom line.” You just explained OPS and MRP without the terms. There is nothing in sabermetrics than cannot be explained in everyday baseball language.
Can you walk us through what you think would happen to the competitive balance of MLB if an NFL-style salary structure (cap) was introduced? My belief is that the best teams today are the smartest, and that this would continue with or without a cap. Whattya say? – Milt
I agree. In fact, when you look at competitive balance in baseball over time, the competitive balance measures that MLB has implemented in the past 15 years to improve competitive balance don’t seem to have had much effect. The graph here shows the Noll-Scully ratio of competitive imbalance, which measures the standard deviation of winning percentages to an ideal standard deviation (1 = perfect competitive balance).
The LOWESS-fit trend shows that competitive imbalance has been decreasing over time, which is a good thing – but that improvement leveled off in the late 1980s. Even though teams are now sharing more revenue and facing a luxury tax (an almost-salary-cap), competitive balance hasn’t gotten any better. In general, research by sports economists doesn’t support the notion that salary caps in other leagues have improved competitive balance.
What would happen if MLB adopted an NFL-like salary cap? Instead of spending big on free agents, the big market teams would spend their resources on non-salary factors that would help them win. The advantage won’t disappear, it will shift. We might see a bigger scouting department, new foreign development academies and more spending on foreign free agents. If winning is valuable, teams always have an incentive to spend to win more. With a salary cap, the losers will be players, who will have that wealth transferred from them. However, the 1994-1995 strike made it clear that the MLBPA will never accept a salary cap.
The Mets’ poor performance each of the past few years can’t be blamed on a low payroll, as they are almost always at the top of the list in terms of team payroll. They have to be at the top of your list of the most poorly run franchises in Chapter 6, right? I haven’t read the book yet but had to ask. – Mitch
Right at the top of the list of the worst-managed teams of the 2000s. Not only did the Mets spend a lot, they didn’t win much either. The Mets problem is easy to identify: The Mets have a habit of signing high-dollar stars rather than focusing on building a better internal scouting structure. The Mets have developed some good prospects, but they have a penchant for buying players when they are expensive. Some examples of bad contracts include signing Oliver Perez to a three-year, $36 million contract in 2009, signing Francisco Rodriguez to a three-year, $37 million contract in 2009,?and signing Luis Castillo to a four-year, $25 million contract in 2008. In 2010, the team was not ready to contend yet signed?Jason Bay to a four-year, $66 million deal. Both Bay and Rodriguez were also huge disappointments in 2010.
I think it’s fun to compare the strategies of the Mets and the Twins, whom I find to be the second best-managed team in the 2000s (the best was the Oakland A’s, but I think that team has been covered enough). During the last decade, the Twins averaged $32 million more in playing value than they doled out in player salaries, while the Mets paid out an average of $25 million more than they received in playing value.
Before the 2008 season, the Mets acquired via trade and then signed former Twins star Johan Santanta when his performance was at its peak. Though he has been quite good for the Mets, the Twins had no problem winning and avoided paying the ace starter almost $23 million/year for six years. In the three seasons since, the Twins have made the playoffs twice, and the Mets have missed the post-season every year. How did the Twins do it?
The key to success lies in acquiring young talent when the collective bargaining rules allow teams to pay players far less than their market value. For their first six years of big-league service, the salaries that players receive are restricted by MLB’s Collective Bargaining Agreement. The Twins exploited these rules by developing talent within their organization, while the Mets concentrated on bringing already-developed talent in. Comparing their 2010 rosters, the Twins drafted 21 of their players, which is equal to the number of players the Mets had signed as free agents. By building a strong farm system, the Twins have been able to survive with young and cheap talent.
The Mets’ market size ought to give them an advantage that allows them to sign better players, but the spending strategy the club has adopted clearly hasn’t worked. If the Mets had adopted the Twins’ method of operation – spending far less for players than they receive in playing value – they might be the most valuable franchise in sports.
If your son was a baseball prospect, would you recommend or demand that he hire Scott Boras as an agent?
No matter what people say about Boras and his exaggerations, when you see a crazy contract announced, more often than not he was the agent in the negotiation. Does he extract the most value for his clients, or is there another agent out there better at this than him? – Ricky
I would love to do a study of agents to see which ones do the best job shopping their clients; however, agent-player data is only now becoming available for such a study to be done. I think Scott Boras’s reputation for commanding top dollar comes from two sources. First, he gets the best clients to begin with. He gets the most because he shops the best. He’d probably tell you as much. But Boras has also been able to convince his clients not to sign long-run contracts before they hit free agency. Players, who have all their value tied up in a single fragile asset (their bodies), have to fear that an injury might prevent them from collecting a big free-agent payday after their six-year indenture is up. Teams have used this fear to sign young players to long-term deals at below-market rates. Players trade some salary in the future in return for long-run security. For example, Jay Bruce recently agreed to a six-year, $51 million deal with the Cincinnati Reds that buys out two years of would-be free agency, when he would be paid much more than the salary he’ll receive. I have long wondered why Boras is so successful at convincing his players to resist the temptation to sign discounted long-run deals as insurance. I have a theory that Boras sells his own insurance to players by promising players a minimum salary in return for waiting for free agency. This way, players get insurance against injury, more income if they reach free agency in good health, and Boras gets a bigger cut.
Is Jeff Francouer worth $13 million a year? – Anders
No. And the Braves and Mets stupidly played the supposed wunderkid at the expense of other players. I estimate that if the Braves had swapped his playing time with the part-time player Matt Diaz, the Braves would have netted an additional $2 million per year. His $2.5 million deal with Kansas City is about what he’s worth. Here is a post I wrote about Francoeur’s worth – and the difficulty in valuing him – a year ago.
Is there any real evidence that dollars motivate athletic performance, or do most athletes try their best regardless of salary? Aren’t good coaching, self esteem and other psychological motivators more effective than large piles of money? – gregg
I think most players play hard because they love the game. Up until the players union won free agency, players made far less money and played just as hard (even harder if you ask someone older than you) than today’s players. With free agency, players get a large share of producer surplus that was previously captured by owners. Once players were allowed to shop their services on the open market, teams bid up salaries, eroding the surplus that they had previously reaped. So, even though players might work for less, they command high dollars.
Yes, this is backwards-looking, but what was your estimate of Barry Zito‘s value before he signed his big (seven-year, $126 million) contract with San Francisco a few years back? – Pete
$53 milllion. Ouch!
When teams that are out of the playoff hunt trade their stars as the trading deadline approaches, they often receive a handful of prospects back. I’m assuming that you have looked at and quantified the result of this type of deal many times. On average, do the numbers show that the teams that trade their stars get enough value in return? Or is it the other way around, or is there nothing conclusive? – Elena
Given the low success rate of even the best prospects, why do GM’s continue to settle for packages of single-A and double-AA prospects when trading top players, instead of insisting on major-league ready players? – mfw13
These questions are somewhat related, so I’ll take them both at once. An analysis of all past deals would be a heck of a study; however, I have looked at a few deals, and found that the values seem to work out. For example, I estimate that the prospects that the San Diego Padres got back from the Red Sox for Adrian Gonzalez were worth an expected value of around $15 million. That is pretty close to the difference between what I expect Gonzalez to be worth after subtracting his $5.5. million salary in 2011. In my book, I examine the deal that sent formerly-famed top prospect Andy Marte to the Red Sox from the Braves for Edgar Renteria. I estimate that Marte’s expected worth, based on his minor-league performance, was actually greater than Renteria’s. In hindsight, it didn’t work out so hot as Marte didn’t blossom into a good player. However, the uncertainty of prospects is well-known, and it dampens the expected value of prospects. Had Marte become an All-Star, the Cleveland Indians (who acquired Marte from the Red Sox before he ever played a game in Boston) would have had a good player who made far less in salary than what his play generated on the field.
In the last chapter of my book, I provide a rule-of-thumb table for valuing minor-league prospects by age and level, and I find that average prospects in the High-A level are worth about $1 million, Double-A prospects are worth about $2.3 million, and Triple-A prospects are worth about $2.8 million. Because low-service-time players can be paid salaries far less than their worth, the rights to uncertain prospects are quite valuable.
Trading away stars for prospects when you’re out of the playoff hunt is a good strategy, because there are increasing returns to winning. Good teams can collect more revenue from good play because fans are more willing to see good teams, and there is a greater likelihood that the team will collect playoff revenue. Conversely, unless a team becomes abysmally bad, once a team starts losing games in the low-70s, losing more doesn’t have much effect on the bottom line. Therefore, it makes since to shed an established good player with a high salary who can’t help you win on his own, in return for a group of prospects who may help you win in the future for wages.
Thanks for your questions.