World Series time! Enjoy Premium-level access to most features through the end of the Series!
February 15, 2013
Mike Trout and Regression Obsession
Like many fantasy players, I spend little if any time during the season worrying about what a player will earn the following year. Even in keeper formats, I don’t invest a significant amount of time trying to figure out future earnings.
While I didn’t have an exact dollar value assigned to Mike Trout for 2013 back in October, I assumed that I’d have him ranked first or second in AL-only formats and first, second, or third in mixed formats. Besides Ryan Braun and Miguel Cabrera, there were few players who seemed capable of putting up big enough fantasy numbers to come close to Trout.
I assumed my thinking would represent conventional wisdom, and that nearly every expert would agree that Trout would be at worst a top-10 player in mixed formats. However, this has not been the case. Although they don’t represent the majority, a number of experts have given Trout a thumbs-down review for 2013. These recommendations have advised anything from moderate to extreme caution with Trout. One expert went so far as to say that Mike Trout would only be the eighth-best outfielder in fantasy this year.
Instead of simply dismissing the anti-Trout crowd as off-base, I thought I’d take a closer look at Trout and determine whether or not the regression argument has any merit.
Mike Trout’s Sophomore Season: Establishing a Baseline
Most of the negativity surrounding Trout has been very vague. Few if any specific examples of comparable players have been provided, and those examples that have been offered have typically been players who don’t fit Trout’s profile in terms of age and/or skillset.
One reason this is the case is that it is extremely difficult to find any players who come close to matching what Mike Trout did during his age-20 season.
Table 1: 11 Best Age-20 Seasons by Adjusted OPS (All-Time)
With the notable exception of Hoblitzell, this table is a veritable who’s who of all-time greats. While there are several potential career outcomes for Trout, his strong rookie campaign at such a young age is an indicator of a superstar career trajectory, as opposed to that of a good or average one. While Trout might regress, league average rookies like Butch Wynegar or Rick Manning aren’t instructive when attempting to establish a baseline for future performance.
Yet, this is what I suspect some experts are doing. Instead of attempting to isolate Trout based on his skillset and what he has done to date, they are comparing him across the board to all rookies. In one instance, I saw an expert comparing Trout to all players across all age brackets. From an analytical standpoint, comparing Trout to a 28-year-old Danny Tartabull, a 34-year-old Jim Edmonds, or a 30-year-old Joe Torre offers little to no value.
Comparing Trout to the best 20-year-olds ever isn’t ideal, either. Every player in Table 1 is an outlier. As a result, even though Trout sits with this group of players in terms of his performance at this age, that doesn’t mean that Trout is necessarily going to have a Hall of Fame career. However, looking at comparable 20-year-olds is a better way to examine what Trout might do going forward.
If we can agree that it is more sensible to compare Trout to the best 20-year-olds to play the game, then we can dispense with the idea of comparing Trout every single player in the history of baseball and instead look at how he stacks up in his sophomore season against similar 20-year-olds. Did the best all-time 20-year-olds regress to the mean in their age-21 seasons?
Table 2: Performance of Elite 20-Year-Old Hitters at Age 21
The first three columns in Table 2 represent the player, his age-20 season, and his adjusted OPS in that season. The four columns on the right list what the player did in his age-21 season.
Taken in the aggregate, to some degree the data support the idea that even elite hitters are not immune to regression. The hitters here dropped an average of 10 points in adjusted OPS across the board. Additionally, their batting averages, K:BBs, and BABIPs all dropped on average to varying degrees.
However, the slippage for Trout’s age-21 OPS+ comps in batting average, K/BB, and BABIP is slight. Furthermore, while the aggregate data suggest that Trout will slip, there are also individual examples of players who did better in their sophomore seasons.
For the sake of argument, though, assume that Trout will slip from an adjusted OPS of 171 in 2012 to this group’s average adjusted OPS of 147 in 2013. A 147 adjusted OPS would have been good for eighth-best in the major leagues last year.
This is all well and good, but I suspect most of you don’t play adjusted-OPS fantasy baseball. How does this translate to fantasy earnings?
A Poor Man’s Mike Trout Is Still a Rich Man
In 2012, Mike Trout earned $47. This put Trout first overall, three dollars ahead of runner-up Ryan Braun and seven dollars ahead of Miguel Cabrera.
But this doesn’t tell the entire story. Because Trout was called up 23 games into the 2012 campaign, he didn’t play a full season. If Trout had been with the Angels for 162 games, he would have projected to put up 35 home runs, 97 RBI, 57 stolen bases, 150 runs and a .329 batting average in 651 at bats, good for $54. Hence, $54 in earnings represents a more accurate point of departure for discussions surrounding Trout’s value going forward.
The data in Table 2 do not portend significant regression for Trout. But let’s play devil’s advocate for a moment, take Trout’s adjusted 2012 line, and reduce his numbers by 20 percent.
Mike Trout 2013 Hypothetical Season: 651 at bats, 120 runs, 170 hits, 28 home runs, 78 RBI, 46 stolen bases, .261 batting average, $35
Thirty-five dollars in earnings still would have been good for fourth-best in the big leagues last year, behind Braun, Cabrera, and Andrew McCutchen.
This expectation would fit in well with what Alex Rodriguez, the most modern comparable in Tables 1 and 2, did in his age-21 season. A-Rod went from earning $45 in 1996 to $33 in 1997. A dip into the mid-$30s in earnings has historical precedent and would still put Trout in rarified air.
But what if the naysayers are 100 percent correct? What if all of the doom-and-gloom predictions for Trout are accurate?
One of the more negative Trout prognosticators had this to say about Trout’s potential performance:
“If we are right, and we are a minority of one, then a reasonable expectation for Trout might be 15 HR and .280 BA, with downside on that BA.”
Here are the logical steps I took to bring Trout down to this negative projection:
Mike Trout’s 2012 ADJUSTED 35 home runs, 97 RBI, 57 stolen bases, 150 runs, .329 batting average, $54.
Mike Trout’s 2013 PESSIMISTIC 15 home runs, 60 RBI, 45 stolen bases, 80 runs, .280 batting average, $31
In terms of pushing Trout down the ranks, now we’re finally getting somewhere. Those pessimistic numbers would have put Trout 10th.
No matter how hard you might try to push Trout down the rankings, it’s next to impossible to drop him too far. In real life, a drop in batting average and slugging percentage would cripple Trout’s value considerably. In fantasy, Trout’s steals are going to be a considerable source of value, assuming that he continues to run.
Table 3: Age 20-22 Mike Trout Stolen Base Comparable Players
Table 3 lists 19 players who are age 20-22 comps to Mike Trout based on stolen bases and shows how they did the following year. Based on age-similar stolen-base kings, Trout should more or less maintain his steals. Projecting him for 45 thefts seems fair.
We Are All Regression Candidates
The argument against Trout that intuitively makes the most sense is that the odds of him doing what he did last year are poor. I agree emphatically with this point. In a Rotisserie-style auction, I won’t be placing a $54 or $47 bet on Trout. My stopping point will be lower.
There is a selective flaw in the argument against Trout, though, and it is stunning that no one (to my knowledge at least) has mentioned it even in passing.
Table 4: Performance of Players with $30+ Salaries 2009-2012
Table 4 represents the 113 players from 2009-2012 who received an average salary of $30 or more in the three expert AL- or NL-only Rotisserie Leagues (CBS, LABR, and Tout Wars). Players who earned within $3 of their average salary in one direction or the other were put in the “draw” category.
History tells us that Trout is extremely likely to earn less than he did in 2012. But history also tells us that nearly every hitter with lofty expectations is also extremely likely to earn less. Seventy-four percent of all hitters with a $30+ price tag lost money, with a mere 11 percent improving on what they did the previous year.
It is acceptable to assume that Mike Trout will not duplicate or improve upon what he did in 2012. It is not acceptable to assume that Trout is going to fall off the map from a fantasy perspective. Even if his home runs and batting average slip considerably, Trout’s speed will provide plenty of value in the run and stolen-base columns and keep him in the neighborhood of $30. While it is fair to expect Trout to slip, it is unfair to assume that every other player in baseball isn’t subject to the same expectations based upon historical data. Weighting all of these factors, Trout should be one of the top three fantasy targets this spring, and it is difficult to see ranking him outside of the top five overall.