clock menu more-arrow no yes

Filed under:

The free agent reliever market is wonky

This winter’s free agent reliever contracts haven’t made a lot of sense.

Kansas City Royals v New York Yankees Photo by Elsa/Getty Images

You’re selling your house. It’s pretty similar to the other houses on your street, and you look at comps to determine a listing price. The house with the in-ground pool listed for $200,000. The one that badly needs a new roof and windows sold for $300,000. Then there’s your neighbor with the state-of-the-art kitchen that sold for $220,000. Finally, the house with dead lawn fetched $350,000. What price should you list?

This isn’t a math problem. There isn’t really a good answer because the comps don’t really make sense. Such is the current free agent relief market.

On the MLB Trade Rumors Top 50 Free Agents list, there are ten relievers who have agreed to contracts. The biggest domino, Craig Kimbrel, has yet to fall. We can be pretty certain that his contract will be in a class by itself in terms of length and value, so he’s not really comparable to the others anyway. Therefore, this is a good time to evaluate the state of free agent reliever contracts.

MLB Trade Rumors makes predictions for each free agent contract. Let’s take a look at how those predictions held up to the actual contracts by comparing average annual value (AAV).

As it turns out, they did a pretty good job. Those lines match each other reasonably well, though there are a few exceptions. Zach Britton and Andrew Miller both signed for significantly more than predicted. Britton’s three year, $39 million deal to return to the Yankees (which is risky) beat his $33 million projection by $2 million per year.

Miller’s contract was more AND less than expected simultaneously. He received a two year, $25 million deal from the Cardinals. MLB Trade Rumors projected him to get a three year, $27 million deal. The AAV is $2.5 million higher, but it’s less by one year and $2 million overall.

The biggest disappointment from the projections was Joakim Soria. MLB Trade Rumors thought he was worth a two year, $18 million contract, but the A’s signed him for two years $15 million. That’s an AAV drop of $1.5 million.

The MLB Trade Rumors predictions are just that: predictions. Their methods for developing them are their own. Let’s see how well the actual contracts correlate with on-field performance by WAR.

Contract Value-WAR Correlations

2018 WAR AAV Correlations Total Value Correlations
2018 WAR AAV Correlations Total Value Correlations
bWAR -0.6 -0.2
fWAR -0.3 0.0
WARP -0.6 -0.7

Huh?!? The correlations are all negative! At least one of the following four statements must be true for this tier of reliever free agents:

  1. The worse you pitch, the more money you will earn.
  2. MLB teams are bad at evaluating relievers.
  3. The three WARs are bad at evaluating relievers.
  4. The metrics used by MLB teams to appraise relievers are largely not incorporated into WAR.

#1 seems implausible, if not downright stupid. #2 and #3 are also unlikely for different reasons, yet one of them is probably true. #4 is the best bet here. It’s impossible to know how teams evaluate relievers, or any player really. Here are a few more crude guesses that could factor into contract offers:

More Correlations

Metric AAV Correlations Total Value Correlations
Metric AAV Correlations Total Value Correlations
2019 age -0.2 -0.3
2018 Velocity 0.0 0.4
2017-18 Velocity Change -0.4 -0.3
2016-18 IP -0.7 -0.7

Most of these aren’t very strong correlations, but there are still a lot of negative numbers. Age apparently isn’t much of a factor in AAV, and probably not total value either, but if so, it appears to be negatively correlated.

Maybe MLB teams prefer older relievers? Not likely. With a sample of just ten pitchers, it’s probably just that the older ones happened to also be better. For example, David Robertson and Adam Ottavino will enter their age-34 and age-33 seasons.

The velocity correlations actually make a little sense. There is no correlation at all between velo and AAV, but there is with total value. Generally, the relievers who throw harder appear to earn more total money by way of longer contracts.

However, we’re back to negative correlations when we look at velocity change from 2017 to 2018. Zach Britton signed the biggest contract of them all (until Kimbrel signs), yet he also lost the most velocity. His 94.9 miles per hour sinker (which he throws far more than any other pitch) dropped from 96.1 in 2017.

The strongest negative correlation of all was with innings pitched. This is skewed by Jesse Chavez, who’s two year, $8 million deal with the Rangers was the smallest of the group. Chavez was second in MLB in 2018 with 95 13 relief innings, trailing only Ryan Yarbrough’s 118 23 (though Yarbrough was more like a starter pitching after the opener in Tampa Bay). In 2017, Chavez started 21 games, so of course he threw way more innings than any of the others.

If we remove Chavez from the group, innings pitched since 2016 has a -0.3 correlation with AAV and a staggering -0.8 correlation to total value. In other words, this still fails to make any sense. The less a player pitched, the more likely he is to sign a longer, larger contract.

Take all of these numbers with a grain of salt. First of all, we’re only dealing with a sample of ten pitchers. That’s not really enough to make too many conclusions. Secondly, there are other factors at work here. For example, Britton and Miller are the only two lefties, and they signed the two best contracts by AAV. Handedness probably had a lot to do with the market.

If you really want to make conclusions from this data (don’t), Craig Kimbrel should be expected to sign a minor league contract. That obviously won’t happen, but it sure has been a strange winter for free agent relievers.


Daniel R. Epstein is an elementary special education teacher and president of the Somerset County Education Association. In addition to BtBS, he writes at www.OffTheBenchBaseball.com. Tweets @depstein1983