The Mega-Challenge Trade

A version of this article was previously published at ESPN.com.

In a recent ESPN piece, I wrote about how I was dealing with my under-performing pitching staff. One idea I offered was trying to trade my bad eggs for someone else’s equally bad eggs. Maybe a “change of scenery” could work for fantasy players too. And in fact, I did pull off a deal later that week in which I swapped Dallas Keuchel for Matt Kemp. You might call that a “challenge trade.”

But then it got me to thinking… what if you tried to orchestrate a massive deal in which a bunch of the worst pitchers were dealt for a bunch of the worst hitters? Who would win? I’m not talking about swapping a bunch of Alfredo Simons for a bunch of Ryan Hanigans. These would have to be star players who are putting up horrible numbers this year.

So, a week ago Thursday, I created a poll here and asked the following question:

MEGA-CHALLENGE TRADE: Owner A deals Carlos Gomez, Justin Upton and Prince Fielder to Owner B for Matt Harvey, Chris Archer and Dallas Keuchel. Who wins?

Of course, timing is everything; I’ll get to that in a minute. But at the time, this was a perfectly reasonable exercise.

When I created the poll, I had an immediate response in mind. But the more I thought about it, the more I realized that there is a lot of nuance to this question. Here were the results, through this past Sunday:

20%    The owner getting the hitters wins.
59%    The owner getting the pitchers wins.
12%    The players continue to struggle and nobody wins.
8%    Eventually, the players turn it around and everybody wins.

These results surprised me. Any of the responses were defensible, but the spread between hitters and pitchers seems extreme. I can easily argue for the pitchers, but not by a nearly three-to-one margin. Let’s look at this sextet in a little more depth.

Coming into the 2016 season, this is how we viewed these players:

HITTERS             ESPN      NFBC      TW$
==============      ====      ====      ===
Carlos Gomez          38       50       $26
Justin Upton          55       45       $30
Prince Fielder        70       82       $24
       AVERAGE        54       59       $27

PITCHERS            ESPN      NFBC      TW$
==============      ====      ====      ===
Matt Harvey           30       29       $29
Chris Archer          50       47       $20
Dallas Keuchel        43       43       $21
       AVERAGE        41       40       $23

ESPN = Pre-season ADP
NFBC = Pre-season ADP
TW$ = Tout Wars mixed league dollar values

The pitchers went a little higher in draft leagues, but we spent more on the hitters in auction leagues. Fielder drives down the hitter values; Harvey drives up the pitcher values. But in the end, both groups are pretty close. How about now (through Monday’s games)?

HITTERS              HR    SB    Avg     R$
=======              ==    ==    ===    ===
Carlos Gomez          0     5   .182     $2
Justin Upton          2     1   .215     $4
Prince Fielder        3     0   .167     $4
       TOTAL          5     6   .200    $10

PITCHERS              W     K    ERA    WHIP     R$
========             ==    ==   ====    ====    ===
Matt Harvey           4    50   5.37    1.54    -$4
Chris Archer          3    72   4.62    1.51     $2
Dallas Keuchel        3    58   5.58    1.47    -$4
       TOTAL         10   180   5.21    1.50    -$6

This is where it gets curious. The poll results were overwhelmingly in favor of the pitchers, yet this trio has dug quite a hole, under-earning the batters by more than $15 so far. Did the respondents really think they have a greater chance for a rebound?

That’s the inherent challenge. When a starting pitcher does well, he can help your team in four categories. But two of those are ratio categories, so a poor performance can do great damage. Compare this to batters, who can help you in five categories, but can inflict damage on only one – batting average.

We can use a bit of game theory to demonstrate the decision-making process. Let’s use the Minimax Theorem to decide which side of the trade makes the most sense:

 

	Fare well		Fare poorly
	POTENTIAL GAINS		POTENTIAL DAMAGE
HITTERS	HR, R, RBI, SB, AVG	1	AVG	3
	5 categories		1 category
	POTENTIAL GAINS		POTENTIAL DAMAGE
PITCHERS	W, K, ERA, WHIP	2	ERA, WHIP	4
	4 categories		2 categories

In this grid, there are four possible outcomes. You can choose the hitters, who may fare well or poorly, or you choose the pitchers, who also may fare well or poorly.

I’ve rated each outcome based on how desirable it is to me. “Hitters faring well” is the most desirable since it potentially impacts five categories; I’ve rated it “1.” The outcome of “pitchers faring poorly” is the least desirable since it can potentially do damage to two categories; I’ve rated it “4.” According to the Minimax Theorem, in decisions like these, the best course of action is always to choose the option with the “least bad rating.” Since choosing the hitters has a downside rating of “3,” that would be our choice.

This is clearly a simplified example. We could also estimate the actual odds of each outcome occurring, which might affect the ratings. For instance, if our advanced analysis indicated that the three pitchers had all been snake-bitten by bad luck – high BABIP, low strand rate, high home run-to-flyball rate – that might shift our opinion. As it turns out, all three have just been bad. Ditto for the hitters.

However, the environment in which you’d have to make this decision is fluid. In the short week since I posted the poll, here is what has happened:

HITTERS              HR    SB    Avg
=======              ==    ==    ===
Carlos Gomez          0     0   .000  Disabled
Justin Upton          0     0   .133
Prince Fielder        1     0   .167
       TOTAL          1     0   .154

PITCHERS              W     K    ERA   WHIP
========             ==    ==   ====   ====
Matt Harvey           1     6   0.00    0.43
Chris Archer          0     7   1.13    0.88
Dallas Keuchel        1     5   2.57    0.43
       TOTAL          2    18   1.23    0.59

Yes, timing is everything. Maybe the poll results were not a reflection of the best percentage play but more about what has happened over the past week. That would mean the respondents believe these pitchers have been “fixed” after one solid outing.

Maybe they have been, and maybe they deserve the benefit of the doubt due to their respective track records. But one start is a very small sample.

In a vacuum, I might still vote for the hitters.