Is Rebuilding Worth It?

Every year the least competitive MLB teams decide whether they will commit to “going for it” the next season, or take a step back and wait for some of their cost-controlled young players to develop into big league contributors, then invest money in the team at that time a year or two down the road.  If the situation is dire, the media and baseball executives alike will start kicking the tires on an organization needing an all-out rebuild.  In this case, teams trade away every expensive, though often productive, veteran for young prospects that can hopefully help form a more competitive and sustainable team in a few years in part due to a higher production to salary ratio.  A judgment is made that investing money into the major league portion of the organization will not yield worthwhile results in the upcoming seasons, leading to declining attendance and television ratings.  That money would be better spent on the draft and developing the players acquired through trades of the more expensive players on the team.  These often publicly announced plans usually have estimated times to completion ranging from 3-5 years, often coinciding with a new baseball executive’s contract length within a year or two.  I set out to measure the results of this strategy as it applies to total revenue, as well as how it works out in terms of return on investment.

I began to question whether rebuilding a team was a good strategy after thinking about a number of teams that struggle to come out of the “rebuilding stage” and then pondering the new changes to the playoffs.  With the current format giving two more teams the chance to make it into the playoffs and further boost their revenue, there is an even closer gap between the “contenders” and “non-contenders”.  Dave Cameron explained some of the potential problems of an all-in rebuild in an article on Fangraphs:

In baseball, the effects of a high draft pick aren’t seen for years, and the likelihood of landing a franchise player with a high pick is much lower than it is in other sports. The value of simply picking near the top of the draft in baseball simply isn’t as large as it in basketball or football, and it’s not a good enough reason to blow up a mediocre baseball team….

There’s too much variation in baseball for teams to simply accept their most recent record as evidence of their short term future. There [are] too many things that simply can’t be projected — and too much uncertainty around the things we do know — for more than one or two teams per year to simply punt the entire season and lose on purpose. Trading from the present to improve the future is one thing; trading from the present simply because we see no future is another thing entirely, and requires a level of certainty in forecasting that we simply don’t have

The real currency of baseball is not wins or runs, but dollars just like every other business.  Scoring more runs leads to more wins, which indeed tends to lead to more dollars going into the organization’s bank account.  However, there are many more things that go into the decision of a fan to pay for a ticket or watch a team on television than how many wins a team has.  It takes many years to build a sizable fan base, though it can take only a few fans enduring a bad on-field product and bad publicity—due to perceived lack of confidence in the team from ownership’s decision not to invest money in the major league team—to rip down even the most enviable support.  Since it is ownership that must sign off on the effort to trade current valuable assets as an investment in the future, the argument must be put in terms that an owner can best understand and support.  Owners are most interested in how it affects their bottom line, and rightly so for the amount of money involved.  At its most basic level, the argument to tear the major league product down and rebuild is a question of how to maximize revenue.  What follows is a look into how teams with a losing record fared across the league following different decisions regarding their payroll.

To measure the efficacy of this decision, I assembled pertinent data in the form of each team’s total revenue estimates, MLB staff and player payroll, and winning percentage for each year from 1990 to 2011.[2]  I then made a series of calculations based on these numbers, starting with finding the change in salary from the previous season for each team-year.  I then compared the revenue estimates to each of the next five team-years to calculate the percent change, as well as added up the years in full to get the total revenue over the same period.  I found the league average changes across time periods from 1-5 years to use as a baseline for comparison by which each team could be judged, both to take care of inflation as well as league-wide revenue changes.  With these numbers I was also able to calculate the expected revenue a team should generate based on league average percentages from year to year.

There is certainly some bias present with this method, since there is no clear independent variable between revenue, winning percentage and team salary.  Increased revenue growth outside of the baseball operations department’s control (such as new TV deals or stadiums) could lead to the decision to increase spending before the market size increase occurs, making it appear the payroll hike was the cause of the higher revenue since it happened first.  However, with so many socioeconomic and market complications unique to each team, I decided these variables were sufficient to illustrate the ramifications of the rebuilding decision on average.  As with any investment, owners should be able to see obvious results in these simple numbers to support further pursuit of such a drastic move.

I first looked at all teams who were under .500 and put them into two main groups:  those whose payrolls increased or stayed at the same level in the following season and those who decreased payroll.  I then looked at only teams with winning percentages under .450 and .400 to see if there was any higher or lower benefit to rebuilding a team further from or closer to contention.  Here are the results:

A lot of information here, but it is best to see the numbers in comparison with each other.  The columns labeled “Ch-Rev X yr%” show the change in annual revenue for each year as a percentage of year 0 minus the league average.  “Tot Rev X yr” refers to the total amount of revenue since year 0 ended as compared to the revenue expected having followed league average changes.  I also included the sample sizes for some perspective on these results.  Let’s go year by year for both groups.  In the first year (after year 0), those who decreased payroll all come out low in revenue, around $3M lower than if they changed at a league average rate.  Those who increased payroll did a bit better, with only those teams below .400 failing to increase revenue in year 1.  Not a big difference, but it is interesting that even after having one of the 38 worst seasons out of 303 losing teams, increasing payroll seems like a better platform to increase revenue.  However, a season below .400 (64-98 over 162 games) is pretty difficult to bounce back from so quickly.  Skipping ahead in the decreased payroll group, the first time any substantial positive revenue starts to show up is in year 3 for the <.450 winning percentage group.  At this point, however, the total revenue figure is already -$4.4M.  By year 5, all three groups are doing better than expected in yearly revenue, though over the course of those five years, their total revenues are below what they should be based on league average.

Moving through the payroll increase group, not only is the first year better off, but also is the entire picture.  For the <.500 group and the smaller <.450 teams, the yearly change in revenue beats league average in every year, leading to an average of $20.6M and $26.1M more than expected respectively.  For the <.400 teams, there are no above average years until years 4 and 5.  Even still, there is a much larger increase in revenue for them than in the decreased payroll group, leading to better than average total revenue over the five-year period, though only $0.9M better.  When compared to the decreased payroll teams under .400, that’s a difference of $7.5M, which is substantial enough to have any owner’s attention.  The difference is even more profound in the larger <.450 and <.500 groups, with a difference of $23.6M and $36.0M respectively in favor of increasing payroll between years 0 and 1.

So far, it appears that investing in the big league team is the better option from a total revenue standpoint.  Next, let’s see if the amount of the payroll increase or decrease has any relationship with the magnitude of change in future revenue.  To explore the answer to this question, only the year 5 numbers are really necessary to get a good feel for the relationship between the two.  I also included a simplified calculation of the average five-year return on investment to give us another measurement for the salary increase group—the total revenue increase over expected minus the salary increase divided by the salary increase.  Below are the results of the entire sample of teams with a lower than .500 winning percentage, split into groups based on percent increase/decrease in payroll from year 1 to year 2:

 

These seven groups illustrate that the further a team gets away from standing still regarding team payroll, the better revenue streams will be in year 5.  However, the best groups by far are still those who increase their payroll, with the higher increases correlating to the higher revenue totals over the next five years as well as higher yearly revenues in year 5.  It would appear that not only does the decision to invest money in a losing MLB team seem more viable than slashing payroll, but the percent of payroll added has a positive correlation with revenue over the next five years.  The group with the largest payroll decrease does show a positive total revenue and yearly revenue compared to league average.  However, we are talking about only nine teams out of the entire 20-year sample, which makes sense because not many organizations have the opportunity to slash more than 40% of their payroll at one time.

So far the data limits a rebuilding decision to be worth it only if a team is under .500 and can shed more than 40% of its payroll, though with less impressive revenue gains down the line. Let’s try to further define a rebuilding effort with stricter conditions to see if there are other subcategories of teams that could benefit from a complete overhaul.  This time we will limit the win percentages to three cutoffs:  <.450 (72 wins or less), <.430 (69 or less), and <.400 (64 or less); and the payroll changes to more than 20%, either increase or decrease.  Here are those results:

We see the same pattern even in these more specific payroll reduction groups, with all three subgroups making less money over the five years than expected.  All three also have better annual revenues than expected, though the loss of income over five years mitigates these successes.  There is one interesting finding in the payroll increase subgroups.  Though any owner would jump at the opportunity to triple or quadruple an investment in five years, the <.400 winning percentage group actually registers a negative return on investment after this period.  This would seem to lend some credence to a very bad team—64 wins or worse in this case—choosing to slash payroll and rebuild.  However, remember that this payroll increase is a bit on the extreme side at over 20%, and it does only involve a nine-team sample again.  For one more set of data, let’s see if those <.400 teams can show a better return with simply increasing payroll by any amount:

This is an even less desirable result.  Even though the teams had higher yearly and total revenues than expected, the amount of money to make it happen is more than makes sense for an owner to invest in this case.  Since this is such a small group of teams, perhaps more information can be gleaned by looking at the best two and worst two teams in this group.  If there are confounding elements at work it could provide a better picture for the type of scenario that makes the most sense.  Here are the figures for these four teams:

The 1996 Tigers produced a then-team record 109-loss season, followed by a string of losing seasons the next five years.  However, the huge majority of the revenue jump came from moving into Comerica Park in 2000, not due to any new television deals and certainly not from increased on-field success; it is hard to pick up anything extra useful from that.

The 2003 Padres were also buoyed in year 1 by a new stadium revenue increase that coincided with better on-field play by the major league team as soon as 2004.  This improvement was facilitated by players already with the team such as Phil Nevin, Mark Loretta, Jake Peavy, Khalil Greene and Ryan Klesko, though the additions of Ramon Hernandez (trade), Terrence Long (trade), David Wells (FA) and Brian Giles (’03 trade) made the team a force in the big leagues.

As for the least fortunate teams in this group, the 1998 Devil Rays enjoyed a nice first year as a franchise in the revenue department.  However, the incessant losing combined with the attempts to build with well past-prime players like Jose Canseco (1999), Bobby Witt (1999), Vinny Castilla (2000) and Greg Vaughn (2000) squashed any chance of maintaining fan interest and revenue.  By the 2002 season they lowered payroll and began to build the team with younger players.

The 2004 Seattle Mariners suffered a horrendous season after a few years of contention and attempted to fill various holes with free agents Adrian Beltre, Richie Sexson, and Yuniesky Betancourt among others.  The additions made were all either unsuccessful or simply not enough to replace the talent lost from 2003.  Though a team like the 2003 Padres was able to successfully build on an underachieving talent base and take advantage of a new stadium to boost revenues sufficiently, it would seem that a team with this poor of a platform year in terms of MLB success will not benefit from investing money in the big league product.  Exceptions could include moving into a new stadium or other foreseeable revenue increases, as well as a team of underachieving but talented players already in place as in the case of the Padres.

How about some examples from the payroll decrease side with the same poor winning percentage (<.400)?  Here are the best and worst:

 

The 1999 Minnesota Twins decreased payroll and had been doing so for the previous 2-3 years.  They are a good example of building from within, with homegrown stars AJ Pierzynski, Torii Hunter, Corey Koskie, Doug Mientkiewicz and Johan Santana producing at the same time by 2001.  In addition to Brad Radke, the early 2000s Twins put together a strong run based almost exclusively on players that were drafted or signed as amateurs.  The only issue keeping this from being a successful 5-year rebuild is the fact that all of these players were drafted or signed between 1993 and 1995.

The early 1990s Cleveland Indians were a successful rebuild by most standards, though the opening of Jacobs Field in 1994 also enhanced this revenue increase.  Here again, the most important players of this successful run were in place before the payroll change occurred.  Albert Belle (drafted 1987), Jim Thome (1989), and Charles Nagy (1988) came via the draft, and Carlos Baerga and Sandy Alomar Jr. were acquired in the trade of Joe Carter in 1989.  With these players in place, Manny Ramirez (drafted 1991) and Kenny Lofton (trade) were the only two truly big moves made close to the time of the payroll decrease.  From these two examples, it would seem to hold that the best timing for an all-out rebuild is just before a wave of prospects are about to make an impact at the major league level.  A true rebuilding effort without players already in place appears to take more like 7-8 years to come to fruition.

The 1995 Blue Jays were two years prior one of the best teams in the league, but had fallen on hard times during the strike years.  Despite some strong picks at the top of the draft before 1995 in Shannon Stewart and Chris Carpenter, as well as strong performances at various times by Pat Hentgen, Roger Clemens, and Carlos Delgado, the Blue Jays could not put together strong enough years to get out of 3^rd place in the American League East.  Top picks Roy Halladay and Ted Lilly would help win games in the early 2000s, but their presences were not felt until more than five years after the payroll decrease in consideration here.

The 2001 Baltimore Orioles were most certainly a rebuilding team, as they traded away Mike Bordick, Charles Johnson, Harold Baines, Will Clark and BJ Surhoff in the previous year while Brady Anderson’s release and Cal Ripken’s retirement both happened at the end of 2001.  Both the young players already in the organization and the players received in trades did not pan out in any way the Orioles would have liked, as only Melvin Mora and Brian Roberts stood out as consistent above average players.  While the Blue Jays seem to have been fairly unlucky in the timing of their star players’ peak seasons, both of these team-year examples suffered poor revenues due to the inability to identify the right kinds of young talent either through trades or the draft.  The Blue Jays in this period lacked quality depth while the Orioles lacked top-end talent.

So, while there is certainly some precedent for trading pricey veterans for prospects and slashing payroll to ensure a more fiscally streamlined path to higher revenues a few years into the future, this definitely only applies to a very small subset of teams in the MLB.  Remember the sample sizes to which the data needed to be limited to see some evidence for a rebuilding team to come out on top.  Based on this data, one can reasonably limit rebuilding to teams with close to or above 100 losses in any given season, though the success and rate of that rebuild paying off depend on the state of the farm system precisely at that time.  For teams that do tear down and rebuild, more often than not it is unreasonable to count on “fleecing” another team in a trade for their prospects, or having all of their current prospects develop into their perfect scenario projections.  Drafting and developing a team from scratch takes longer than the five years most baseball operations executives promise to fans and owners.  Building a stadium seems to help though (sarcasm).  Another sub-group that fared well was the small portion of teams that could cut more than 40% of their payroll—again, a very small number of teams.

Beyond these fairly strict conditions, rebuilding does not show up in the data to be a worthy pursuit.  There may be other reasons that organizations decide to go down this path and convince themselves it is right for them, besides simply not having the money to invest further.  Perhaps it is out of fear of risk that teams choose not to increase payroll with a losing team, both from an owner’s perspective and that of a baseball operations department.  If owners do not have full faith in their executives’ abilities to make decisions with sound judgment, it is very easy to see why they would be quicker to sign off on a plan that calls for reduced spending.  From a baseball operations standpoint, selling patience to ownership over a three to five year period while the organization’s talent base gets rebuilt buys some time and job security.  It is much safer to say that at some point in the next five years a team will be able to contend for a playoff spot, since as teams prove every year, the most talented teams on paper do not win the most games.  Even with bad decision-making and poor return on trade or free agent investments, it is quite possible for any team to have one season in five where they consider themselves “competitive”.  History shows that unless a team is among the absolute worst of the decade in terms of record and talent, rebuilding should not be considered as a viable option.

When viewing total revenue, annual revenue, and return on investment estimates from the last 20 or so years, spending more money on a team with a losing record will help speed the recovery process and leave the team in better long-term shape.  The more an owner invests in a sub-.500 team the better the revenue streams and returns down the road.  Even for those sub-.400 teams where investment returns are close to zero, the teams going forward beyond year 5 appear better positioned and less likely to alienate fans.  This is the trickiest situation, especially one in which a team has little help on the way from its minor league system with a very poor (.400 Win % or worse) big league team.  Since a true rebuild requires more along the lines of seven+ years, an owner must decide either to brave the painful growth of the franchise for up to a decade, or invest money in the big league team immediately, knowing that the return will not be seen for at least five years, though annual revenue streams and the health of the franchise during that five years in the form of total revenue will be much more favorable.  Further research may be needed covering years 6-10 to get a better understanding of the decisions facing this small group of teams.  For the rest of the league, however, teams may only need to take a closer look at recent history to find the right path.