Free Agency Spending Trends in the Tanking Era

The MLBPA is standing firm in addressing the current perceived problems with free-agency. As an MLBPA player representative, Andrew Miller has said in a story by ESPN’s Jesse Rogers, “Competition from all 30 teams will fix a lot of what’s going on in the markets.” Max Scherzer, another rep., re-iterated this in a press release from NBC Sports Washington following signing a contract with the Mets before the lockout. Former player, Trevor Plouffe, has also echoed the “competitive imbalance” narrative on the “Talkin’ Baseball” podcast. Agents and writers seem to confirm this, as well. So, there is a palpable perceived shift in free-agency signings because of “tanking” or “re-building” clubs. If you’re having flashbacks to recent free-agent off-season gridlock, you’re not alone.

There’s a mega-ton to un-pack, here. Free-agency qualms can boil down to a few issues: (1) teams have stopped paying players for past performance and (2) are avoiding handing huge contracts to players past their prime (~ age-30+ seasons). This is obviously problematic because (3) most players are typically not eligible to test the open market until roughly this time or after. Thus, players want to maximize their value or extend their careers, which may (4) prove difficult when not all/any teams are willing or able to pay free-agents, regardless of the player’s age, quality, or sought length of contract.

Together, these issues may be coming to a head now because of the universal rise in analytics. Explicitly, more effective player projections and evaluations were not relied upon across all facets of baseball operations by all teams. Also, the concept and practice of “tanking” has come to the fore.

The frame of reference then becomes key to analyze free-agency within the context of tanking, since tanking is tied to the explosion of analytical processes of evaluating and predicting player performance of all players in clubs’ effort to be more efficient. Therefore, I’ll consider free-agency in the last ten years: Jeff Luhnow and Theo Epstein, architects of what most consider “successful tanks” were both hired in the 2011 off-season by the Houston Astros and Chicago Cubs, respectively.

ESPN.com, Spotrac.com, and FanGraphs.com differ in the contract data they make freely available. Interestingly, per ESPN, teams have signed free-agents to a relatively consistent number of major league contracts over the last decade (Figure 1A).

Figure 1. Gross trends of Major League Baseball player free-agent contracts from each off-season, from 2011-2021, as per ESPN. Off-seasons comprise the time immediately following the regular season and include the pre-season of the subsequent regular season: the 2011 off-season encompasses October 2011-Opening Day 2012. A). The total number of major league deals from 2011-2021. B). The total dollars spent on major league deals from 2011-2021.

One caveat here is ESPN can underreport some MLB free-agent contract details. An example of this is the omission of Raisel Iglesias’ contract total dollar value in 2021. A second caveat is: contract numbers are consistent when overlooking last year’s pandemic-aftershock off-season, and the current off-season (incomplete and shortened by lockout). Otherwise, clubs have shelled out a somewhat consistent number of contracts and amount of money to free-agents in the last decade (Figure 1B): no clear, global downward trend, even after the last Collective Bargaining Agreement (CBA) was signed in 2016.

The mean and median contract length, AAV, and total contract value also do not show clear downward trends, and all are relatively stable (Figure 2A-C). However, there is a clear trend to sign younger players (Figure 2D).

Figure 2. Means and medians of MLB free-agent contracts each off-season, from 2011-2021, as per ESPN. Off- seasons comprise the time immediately following the regular season and include the pre-season of the subsequent regular season: the 2011 off-season encompasses October 2011-Opening Day 2012. A) Mean and median total contract length (years). B) Mean and median average annual value (AAV, in M $US). C) Mean and median total value (in M $US). D) Mean and median free-agent signee age (years).

Interestingly, there is also an increase in minor league deals being signed (Figure 3A), and these minor league deals are also trending towards younger players (Figure 3B). Together, this supports that teams are universally opting to sign relatively younger players, whether on lower-risk, minor-league deals, or perhaps following injury.

Figure 3. Free-agent minor league contracts signed, from 2011-2021, as per ESPN. Off-seasons comprise the time immediately following the regular season and include the pre-season of the subsequent regular season: the 2011 off-season encompasses October 2011-Opening Day 2012. A) Fraction of minor league contracts signed out of all free-agent contracts signed, expressed as percent. B) Mean and median age of free-agents that signed minor league contracts.

Outside of the age trend, per Forbes, the mean revenues of MLB franchises from 2011-2021 have shown trends similar to AAV (Figure 2B), suggesting that teams do not lack money to spend. Moreover, it has been well-known since at least the last CBA in 2016 that all players have been continuing to receive less of teams’ revenues. Similarly, revenue sharing and the Competitive Balance Tax (CBT) have minimized the large-market teams’ return on investment to add to their payroll. Therefore, teams are dis-incentivized to spend on free-agents. Neglecting last year’s highly unusual off-season (following the owners’ loss of hundreds of millions of dollars) and the current off-season, free-agent contract values have largely been stable since 2011. Contracts have trended younger, and trended roughly shorter, where risk is minimal. However, this says nothing about the distribution of the contracts.

A key point of emphasis by many is that elite free-agents account for more of all the free-agent spending. Teams continue to shatter free-agent contract signing records (Mike Trout and Bryce Harper in 2019, Gerrit Cole in 2020, and Max Scherzer in 2021) not only in terms of AAV (Trout, Scherzer), but total contract value (Trout, Cole) and length (Harper, Cole). Thus, teams are willing to pay the brightest free-agent stars for their “peak” years before the back-end of their aging curves; however, clubs are not only limiting “over-paying” for most other free-agent veterans, but may also be replacing them with cheap minor-league talent to offset the continued rise mega-contract AAVs. We can see this in a few ways: the accelerated dip in median AAV relative to the mean AAV from 2015 onward (Figure 2B, gold relative to blue); in the constant median total contract values (Figure 2C, red), despite increases in the corresponding mean total contract values (Figure 2C, purple). Another way to visualize this is the density plot by year of free-agent AAVs in Figure 4A.

Figure 4. Distributions of MLB free-agent contracts from 2011-2019, as per ESPN. Off-seasons comprise the time immediately following the regular season and include the pre-season of the subsequent regular season: the 2011 off-season encompasses October 2011-Opening Day 2012. A) Density distributions of all ML contracts signed in terms of Average Annual Value (AAV in millions $US). B) Boxplots of each MLB team’s free-agent contracts from 2011-2019.

A density plot normalizes for number of observations. Each season’s distribution of free-agent AAV is right-tail skewed, although the maxima seem to incrementally shorten while the distributions widen over time. This suggests that more free-agent contracts are being signed with somewhat higher AAV over time, but the bulk of free-agent contracts are consistent over time. The right-tail skew is also evident upon examining the small-market clubs’ distributions of AAV in major league free-agent contracts (Figure 4B). This is further supported by the MLBPA’s grievance filed against the Rays, Marlins, Pirates, and A’s for lack of spending in 2018, and confirms the notion that free-agent mega-contracts drag the mean free-agent AAV and total values up, while the vast majority of free-agent contracts (represented by the median) continue to stay much lower. A growing mean AAV with a constant median AAV suggests there are fewer contracts or smaller AAV for those contracts. In other words, if teams have limits on payroll and roster spots, but will spend more on certain players, less payroll or fewer spots remain for the rest.

So, it appears on the surface that the bulk of free-agent major league deals are becoming shorter and given to younger players. Intuitively, the mean AAV and total contract value are both consistently larger than median AAV and total values. However, the growth in these mean values is actually less than the rough $100M increase in average team revenue in the first half of that time-frame. Furthermore, the mean AAV and total contract values for free-agents spike in 2015. These trends seem to coincide with additional revenue streams of local media deals.

Alternatively, these trends coincide with the 2016 CBA, perhaps as bargaining chips by owners, or as pre-emptive moves in anticipation of unfavorable CBA changes the following year. Primary changes in the 2016 CBA included the increased CBT threshold (the proportion of its penalty also rose) and elimination of 1st-round draft pick cost attached to the qualifying offer (QO). The free-agency spikes in 2015 are interesting because the QO and its 1st-rd draft pick compensation were implemented in the previous CBA (2012-2016)— one would expect these global spikes to correlate after the removal of that penalty. Perhaps teams were not yet as universally analytically savvy and did not value 1st-rd draft picks as much as in more recent years. Craig Kimbrel and Dallas Keuchel are well-known, perhaps now forgotten, reminders of the ruthless global shift in front office strategies toward free-agency and toward all baseball operations. Recall that both pitchers came off reliable 2018 campaigns, declined the qualifying offer, and sat out until after the 2019 June amateur draft, so that their signings no longer cost draft picks. Remember also that this was after the removal of the 1st-rd pick attachment to the QO (their signings only cost later-round picks). This clearly demonstrates that even competitive/large-market teams now value higher-round draft picks (not just 1st-rd picks) over most free-agents.

Together, all of this implies: (1) a general shift away from free-agent spending, (2) smarter evaluations of player performance and projections to enable smarter spending (not considered in this piece), (3) teams believe in their development processes or are (4) recognizing the value of top prospects, and (5) revenue sharing and the CBT are limiting the growth of the majority of free-agent earnings as represented by stagnation of median free-agent contracts in the latter half of the “tanking era.” When contrasted with the mean free-agent contracts, this last point can also support the notion that the “middle-class” free-agents are getting “squeezed out.”

Some solutions to these emerging trends relate to competitive imbalance and include: raising the luxury tax threshold, setting a payroll floor, devising a more complex draft order than a lottery or reverse order, and penalizing teams that do not directly invest revenue sharing into payroll. Ultimately, while a macroscopic view of free-agent spending can’t support competitive imbalance claims alone (if at all), the players have legitimate gripes about the status quo over the last two CBAs, during a definitive “tanking era.” Their gripes are particularly true in light of the fact that payroll positively correlates with winning. If more teams punt entirely on free-agent spending, then this logically squeezes the numbers of marginal free-agents and their contract values. This is compounded by limiting the clubs that can spend to spend. Therefore, we could and should expect these trends to continue unless the CBA is substantially restructured.