Summary of Free Agent Market Trends

During this series of articles that have comprised my FanGraphs Residency, I have updated my analysis of the free-agent market that I last researched over three years ago. The vast majority of my new findings have suggested that teams have gotten smarter about spending in line with true player talent, all the while spending roughly the same share of league revenue as they were spending before.

Perhaps my biggest finding is that the OPP Premium has declined. Teams used to receive significantly less WAR for signing other team’s players as they did for re-signing their own players, and this seemed largely related to private information that teams knew about their own players. As teams have become more aware of this phenomenon, the evidence suggests that they have become more careful and have driven up the price of their own players while being more reluctant to sign players on other teams.

This is especially true for pitchers, who used to have the largest OPP Premium. Hitters appear to have actually increased their OPP Premium, which is probably more related to a handful of expensive players who did not pan out rather than teams collectively getting sloppier about signing hitters.

On the other hand, positional premiums remained largely intact. Teams continue to spend more for offense-first positions than defense-first positions, and this suggests that competitiveness for available free agents among large-market and win-now teams plays a big role in driving up prices at offense-first positions. In other words, there are probably more bidders for center fielders than for shortstops, because a team with a good center fielder can still get some value on a free agent center fielder by selecting one of their two center field-capable players to send to left field or right field. A large-market team with an elite shortstop probably does not get much value by bidding on a free-agent shortstop, making the market more limited for shortstops and the price lower as a result.

Relief pitchers no longer cost as much per WAR relative to other free agents, however. Teams seem to have become smarter about the relative value of players who are on the field for longer, and have driven up their prices accordingly. Starting pitchers remain priced roughly in line with the average free-agent cost, however.

Albert Pujols’ contract hasn’t quite worked out the way the Angels hoped. (Photo: Keith Allison)

Looking at specific player types at the statistic level, we see that good baserunners have become more fairly compensated over time. On the other hand, there may still be some low-hanging fruit among slick fielding infielders, although the difficulties of measuring defense make this harder to say for sure. Teams were already pretty smart about what types of hitters were valuable when I performed this analysis several years ago, so there was not much more to discover there.

Starting pitchers with good peripherals have been paid better as well, while pitchers with high win totals have been less lucky on the free agent market. A number of high-profile, free-agent disasters among power hitters at offense-first positions and uniquely among closers (but not other relievers) seem to have led to particularly high cost per WAR in practice for those groups, although the most likely scenario is that this was not a trend so much as a few notable expensive examples.

In general, the bigger picture is that the low-hanging fruit has mostly disappeared. It has become harder to identify undervalued players using simple data, which means that teams will have to continue to innovate if they want to get good bargains on free agents. Some of this is simply better forecasting of performances, but it also requires better forecasting of free-agent markets. Being able to identify what types of players end up overpaid or underpaid due to competition is an important aspect of free-agent targeting.

Furthermore, determining which types of players a given team has a competitive advantage at signing could be an important source of value. For example, a team with solid outfield defense could still get value by outbidding other teams for a fly-ball pitcher. Similarly, a team with a pitching coach who teaches a good curveball could help create value out of pitcher with talent who is struggling to find his curveball. Simple park effect differences between teams could be a source of value as well.

Of course, there are many other things that I have not ever tested about the free-agent market that may not constitute low-hanging fruit, but are obtainable with good analysis. The teams that succeed on the free-agent market will be the teams that operate on all fronts in searching for better bargains.





Matt writes for FanGraphs and The Hardball Times, and models arbitration salaries for MLB Trade Rumors. Follow him on Twitter @Matt_Swa.

9 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
rhennessy1572
6 years ago

The idea that teams pay more for WAR by position makes me question the underlying positional adjustment within the WAR framework itself. WAR is supposed to adjust for scarcity of production by position, therefore if teams pay any kind of premium based on $$/WAR by position than they are adjusting over and above what WAR is already supposed to account for.

Charles
6 years ago
Reply to  rhennessy1572

It doesn’t bother me. The positional adjustments were based on actual data for defensive results of players shifting between positions.

Would you also argue that teams underpaying for good baserunning value or overpaying for bad baserunning value would imply that the WAR models are adjusting the overall WAR of a player by too much for what the data says the baserunning value is worth?

Teams can make poor decisions. Or, those decisions can be influenced on something other than pure WAR.

jcsbck
6 years ago
Reply to  rhennessy1572

I would assume if every team started over, you’d see that fantasy draft go more by WAR for the earlier rounds. It’s impossible for WAR to be used in situational situations, and teams that currently have rosters just that.

But once you get into the 3rd and 4th tier,teams may only want you for a specific niche. Again that defeats the purpose of WAR. I’d rather have a mediocre player in a less important position. Or give a RP a big 1 year deal to trade him at the deadline for prospects of positions of need

Farm systems influence teams spending on players too, but prospects don’t affect WAR. And since every team thinks they have the next best SS of all time, and prospect rankings do this, it effects the true value of WAR in free agency.