keyboard_arrow_uptop

When you're the Padres, relaxed in an offseason hibernation that dates back several color schemes, your first splash of the offseason isn't supposed to look like the Matt Kemp trade.

You're not supposed to be the team getting the big piece. You're supposed to give him up and get back the pieces who are lottery tickets, the ones who may never turn into anything. You're supposed to get the guys in their 20s for the 30-something, not give them up.

Yet, in a way, this trade made total sense for the Padres and the Dodgers, whom Craig Goldstein pointed out were taking their first dump, or something like that, and not just as it concerned their specific rosters.

From a pure standpoint of risk, which is how these trades often sort themselves out, this made perfect sense. The Dodgers were getting the 30-something position player's risk profile in Yasmani Grandal, while the Padres were getting the prospect.

Before getting much into the specifics of this deal and how it fits with other moves that these teams made, the trade immediately took me back to one of my favorite questions that Ben Lindbergh and Sam Miller ever fielded on the Effectively Wild podcast. Coleman from Southampton, United Kingdom wrote for a topic that would be addressed around the 3:00 mark of Episode 386:

If a pitcher was projected to post a 3.50 ERA with 200 innings, he could probably command a similar number to Burnett on a one-year deal. If, however, all 30 teams were given a crystal ball-based guarantee that he would achieve those numbers, how much will that amount change? … While the risk of injury and poor performance is eliminated, so is the possibility of a truly great season.

To give a brief summary of the responses, Sam thought his salary went down. Ben thought his salary would go up. Sam thought Ben was wrong.

Sam's argument, paraphrased: General managers all think the players they’re signing are going to hit their upside. The team that signs a player is the team that values him the most highly. If you take out any variation, teams may not value that all the same, but you eliminate the winner's curse aspect that inflates salary.

Ben's argument, paraphrased: Teams place a (negative) value on the injury risk, and that risk is gone. There's more uncertainty about pitchers staying healthy, and so this guarantee would be worth more for pitchers.

In the stock market or most other valuation type situations, there would be a clear answer to this question one way or the other. In the stock market, for instance, if two stocks or portfolios have the same expected return, the one with the lower variance is more desirable. In more practical terms, you require a higher return if you're going to take on a higher risk. The options market goes the other way, putting higher value on higher volatility assets.

On major league rosters, the answer isn't so simple. To take the question from the podcast to an only slightly more realistic place with no crystal balls at our service, would you rather have (and thus put higher monetary and trade value on):

Player A, who has a 50 percent chance to be a 4-win player and a 50 percent chance to be a 3-win player

OR

Player B, who has a 50 percent chance to be a 6-win player and a 50 percent chance to be a 1-win player

The stock market would have a consensus on Player A being valued higher, but the 30 baseball teams probably shouldn't have a consensus at all. Specifically, the Padres should want Player B, the Dodgers should want Player A, and that difference in valuation of risk is—as much as the LA outfield glut or San Diego's desire for a brand name—why I thought the Kemp trade made sense.

If your acquisition makes you a 93-win true talent team whose expected win totals are concentrated somewhere around 90-96, there's a lot less to be gained from that player being volatile and hitting his upper deciles than there is to be lost if he's high volatility and winds up on the low side.

On the other side, if your acquisition makes you an 81-win team, which might even be optimistic for the Padres, you need all the volatility you can get. You need the team's luck combined with that player going beyond his projection to reach anything like a playoff number. And if he bombs, what's the difference if you win 78?

To the Dodgers, a catcher like Grandal who is a 2-win player and has a great chance to be a 2-3 win player this year is gold. They're the team to beat. They need as little to go wrong as possible.

To the Padres, Kemp, who is injured-ish and whose last four seasons of WARP numbered 9.5, 4.0, 0.8, and last year 3.8 with an All-Star-type second half, the possibility, however slim, of him hitting the high end is worth the risk. If he's bad, they were bad anyway, but if you're paying for the chance for it to go right, he's a great guy to maybe go right on. So is Wil Myers, who might be his 2014 awful or who might be Rookie of the Year.

If the Padres bring in a couple of these outfielders and it costs them playing time for Seth Smith, maybe they got no better in expectation. But even if they didn't—if they were an expected 81-win team with the safe players and an expected 81-win team with the volatile players—their chances of exceeding that number went up.

As a bit of an epilogue, none of this explains how the Dodgers were the team that took on the huge question marks in Brandon McCarthy and Brett Anderson on this stacked roster. But it might just be a different calculation if you can buy your way out of the low end of volatility in spring training or the trade deadline. Then you're talking about the options market, where the answer is clear.

You need to be logged in to comment. Login or Subscribe
fawcettb
12/18
Smart article. Thanks.
therealn0d
12/18
" relaxed in an offseason hibernation that dates back several color schemes" I LOL'd.
ivynora
12/18
Myers is entering his third season; confused about the ROY comment. Perhaps you meant he has potential to recapture his ROY performance?
therealn0d
12/18
Yeah, I interpreted it as "might be his Rookie of the Year self."
tbwhite
12/18
Thank you for this article. Too often I see articles where people say player A is expected to produce 4 WAR and player B 3 WAR, therefore any team that prefers player B is dumb and/or if teams prefer player B there is a market inefficiency, while completely ignoring the risk/variance of players A and B. If a team prefers player B they could be dumb or player A could be much riskier and they simply don't have an appetite for risk. There is no market inefficiency unless you can prove that greater returns are possible without assuming greater risk. People always forget the last part, they think if they show greater returns available somewhere they are done.
TeamPineTar
12/18
Excellent, Zach. Is Craig aware that anybody on the staff thinks this well and this clearly? :) Happy Holidays!
MartnAR
12/18
Good article! Helps you understand risk assessment within baseball.
misterjohnny
12/18
Another view is the Dodgers can grab two or three Type B players and count on one of them to be a six win player. The other two get DFA'd.
PeterCollery
12/19
Tangential note on finance: Options on volatile assets are valuable not because their buyers prefer volatility, but rather because such options can limit losses arising from that volatility. The risk aversion that you (accurately) described as lowering the price of volatile stocks serves to increase the value of options on such stocks.
stephenwalters
12/20
Yes indeed -- very insightful, Zach. Good to see how teams and analysts are applying finance and economics (as well as statistics) to personnel decisions. Keep up the good work!