ESPN.com: Draft Kit | [Print without images] |
Adam Dunn may have hit just .219 last year, but he did have 34 homers and 86 RBIs. |
Fact 1: Major League Baseball players, as a collective whole, batted .253 in 2013. That was the lowest batting average for the group in 41 years, and it also represented the seventh consecutive season in which the league's mark declined.
Fact 2: Major League Baseball hitters, as a collective whole, struck out in 19.9 percent of their trips to the plate, the highest single-season mark in the history of the game. That number has now risen in eight consecutive seasons.
These facts are hardly coincidental, and they reveal a growing swing-freely, trade-average-for-power trend permeating the game.
Fantasy owners need to react accordingly, as when it's no longer a black mark to strike out 200 times nor to bat sub-.250, in the real game, it's no longer such a problem to absorb said player on your fantasy team (so long as that player contributes something in another meaningful category, naturally). Judging by early ADP, many have begun to recognize this, but it's possible that others haven't realized the extent of this steady categorical decline.
For instance, using Player Rater data to demonstrate an "average" batting average, here are the batting title-eligible players, from a few select seasons, who registered the scores closest to zero in the given year:
Notice the more-than-25-point differential between Drew's 2006 and McLouth's 2013 -- not to mention the near-25-point differential between the majors' collective averages -- a perfect illustration of the league's shifting trend. Granted, those are two select seasons in which the league's batting average resided at an extreme -- 2006's .285 was the highest of any season from 2001 to 2013 by more than eight percentage points -- but for those who have played fantasy baseball for a decade-plus, it's nevertheless a critical reminder that a .300 batting average carries a lot more weight today than in the past (and conversely, that a .250 batting average is no longer so damning).
Or, to put it into Player Rater terms, a player who once made a negative contribution to your fantasy team with a .270 batting average in 500 at-bats in 2006, actually made a positive contribution to it in 2013.
This necessitates an adjustment to your strategy, and because some fantasy owners won't be quick to keep up with such a changing trend, it presents a potential advantage. In certain leagues, in fact, low-average hitters might represent the bargain opportunity of your draft, often to the extent that you could even consider adopting an extreme approach:
You could punt batting average.
By "punt," we mean deliberately sacrifice any hope of contention in a chosen category, by entering your draft or auction with the intent to score a "1" (the lowest categorical score in Rotisserie). This means -- at least if you're employing the strategy correctly -- avoiding hitters particularly strong in batting average and placing emphasis upon hitters especially weak in the category. You'd go out of your way not to draft Joe Mauer, while instead targeting Mark Trumbo, even if at a dollar/round greater price. The goal, naturally, is to win every other category by focusing all your draft/auction assets on them instead.
This might sound like a foolhardy strategy on the surface, but in a season like 2013, it resulted in greater success than you'd think. Two of the noted experts-league titles were won by teams that punted batting average: Larry Schechter's team in the League of Alternative Baseball Reality (LABR) American League-only, and my own in Tout Wars' National League-only.
Schechter's LABR-AL team claimed the crown by a 14 1/2-point victory margin, despite accruing only two Rotisserie points in batting average (his team hit .2499, or only .0044 points better than the last-place team). At the auction table, he purchased noted low-average hitters Adam Dunn ($15), Mark Reynolds ($13) and Carlos Pena ($9), and he didn't roster a hitter projected to bat .280 or higher; his strategy became apparent to me approximately one-quarter of the way through.
Speaking to Schechter about his strategy post-auction, his take made a lot of sense: He prepared a set of adjusted dollar values extracting batting average from the mix, in order to determine ceilings for these players who had more appeal to him. It's an experiment worth pursuing for those considering punting: Collect dollar values for a traditional Rotisserie 5x5 system, then run them again using only the other four (home runs, RBIs, stolen bases and runs scored) and compare to identify the players most and least attractive in your quest. For those handy with a spreadsheet, this is easily done using our Custom Dollar Value Generator.
B.J. Upton could come at a bargain this year after hitting a paltry .184 last season. |
Doing my own research for Tout Wars, I identified two bargain candidates using pitch-tracking data -- Carlos Gomez and Paul Goldschmidt -- that I planned to pursue relentlessly in the auction, but first recognized a thread between them common to Schechter's strategy: Neither was projected to hit for a particularly high batting average. (Bear in mind that, in 2012, Goldschmidt had batted .286 behind a .340 BABIP and 22.1 percent strikeout rate, so at the time his odds of declining in the category appeared greater than 50/50. I genuinely believed he had a 40-homer, 15-steal cap, but at the expense of a .270 average.)
In fact, upon closer research, several of my other bargain-bet candidates fit the description: Pedro Alvarez, Ryan Howard, Jay Bruce, Jason Heyward, B.J. Upton and Dan Uggla. At that point I made the decision: I'd punt batting average in order to guarantee I'd get my top two guys and, with some luck, at least a few of my other targets. I wound up acquiring Bruce ($30), Goldschmidt ($28), Upton ($27), Howard ($22), Gomez ($20) and Uggla ($16) -- bearing in mind that Goldschmidt and Gomez were required buys per my strategy -- but Heyward ($32) and Alvarez ($17) wound up too expensive for me at the times of their nominations. I didn't roster a single player projected to bat higher than .282 and, using our projections, I was forecasted to finish dead last in the category, nearly five batting average points behind the next-best team.
This team wound up winning Tout Wars with a 10 1/2-point victory margin, despite finishing -- as projected -- dead last in batting average (.2475, .0013 points out of 11th place).
Now, here's where the punt-batting-average strategy takes the interesting, in-season wrinkle: Thanks to hot starts by both Gomez and Goldschmidt -- they were batting .321 and .330 through May 31, and .295 and .313 entering the All-Star break -- my team performed better than expected in the category initially. At the approximate quarter-pole of the season (scoring period ending May 19), it was in eighth place (five Rotisserie points); and at the mathematical midpoint it was in 10th (three). This meant that, had I chosen to alter the strategy midstream via trades, or by having acquired via FAAB players who stood to increase my team's batting average, I could've done so.
It turns out that such an in-season opportunity never presented itself, but it doesn't mean that it couldn't to your team. And that's the takeaway of the punt-batting-average strategy, relative to punting many of the others: Thanks to the volatility inherent to the category, it is one of the very few that you could correct if you needed to do so in-season. Trying to punt batting average doesn't necessarily mean successfully punting, if only because the right mix of players could conceivably overachieve.
As an aside, the wrong mix of players could comparably underachieve, though that's further evidence that putting too much emphasis on batting average as a draft-day target is a poor idea.
Utilizing a punt-batting-average strategy isn't one that must be limited to AL- or NL-only leagues. Even in our 10-team standard mixed, it's conceivable that such a strategy, executed brilliantly, can result in a championship; this runs counter to many owners' belief that the shallower the league, the more likely you need to tally as many points as you can in every Rotisserie category.
Thanks to the tremendous research of Mike Polikoff, who oversees our league manager product, we know that last season, one ESPN league champion was crowned despite a team batting average as low as .251.
That's not to say that punting batting average will work across the board, nor that it's the strategy to use, in ESPN standard leagues. After all, we also know that less than 1 percent of ESPN league champions (0.83 percent, to be exact), managed a team batting average beneath .258 ... the very "net-zero" batting average produced by McLouth. And, per Polikoff, the average team batting average by standings position last season shows that a typical champion finished with a .2754 mark, whereas the typical last-place team finished with .2671, that quite a bit higher than .251, or even .258. To punt means to decrease your margin for error in every other category, and precious few who punted -- or at least took the strategy to its extreme -- actually wound up winning.
But back to that point from a few paragraphs back: It's the volatility inherent to batting average that makes this category an overrated one, in terms of weighting it in draft-day/auction-dollars stock, and one more easily manipulated to gain a categorical advantage via trade/pickups in-season.
Among the statistical innovations of the past two decades that have gained widespread exposure in fantasy baseball, batting average on balls in play (BABIP) helps illustrate this categorical volatility. Though I've argued many times over the years that BABIP isn't the be-all, end-all measure of "lucky outcomes," it does, still, measure a certain amount of a player's good fortune on balls put into the field of play. Players with unusually high or low BABIPs might have gotten a few more or less lucky bounces than others, and the results of these outcomes can greatly influence the BABIP category.
In short, your preseason projections for batting average are the most likely of the primary Rotisserie five to miss the mark; that's assuming, of course, that you've estimated the hitter's playing time accurately (if not, the others will surely be further off). Large swings in batting average are common, and are the reason that, in any valuation formula, you should lighten the category's weight to compensate the likely in-season variance.
To again use ESPN's data to illustrate, consider the chart below. It shows the average Rotisserie points earned by category depending upon a team's finish in the standings; so, for example, it shows that the average first-place team scored a little more than seven Rotisserie points in batting average, while the last-place team scored just more than four. And, most critically, it shows that batting average exhibited the narrowest range of average Rotisserie-point outcomes, meaning it had least influence of any of the 10 on final standings.
Moving to individuals, there is perhaps no greater example of batting-average volatility during the past decade than B.J. Upton. Everyone recalls that he batted a putrid .184 last season -- that the fifth-lowest batting average by any player with more than 400 plate appearances in the given year since World War I -- but how many remember that he also batted .300 in 2007? Oddly enough, Upton's two worst single years in terms of K rate also came in those two seasons: He had a 28.1 percent rate in 2007, and 33.9 percent in 2013.
Or, for another example, let's examine the aforementioned Uggla. His .179 mark last season matched Rob Deer's single-year worst in baseball history among players with the modern 502-plate appearance minimum (Deer's occurred in 1991), but Uggla also batted as high as .287 just four years ago; granted Uggla's second-best year in terms of K rate happened in that 2010 campaign (22.1 percent) and his career worst occurred last year (31.8 percent).
These two represent more extreme examples of such variance, but every year, a handful of players wind up somewhat overrated or underrated, merely as a result of what were outlier -- comparative to their career track records -- batting averages. The following chart identifies the 15 hitters with 400-plus PAs in 2013 whose batting averages had the greatest differential comparative to their career marks entering the year (minimum 1,000 PAs entering 2013) and, where applicable, provides a possible explanation for the boost:
Next, here is the reverse: These are the 15 hitters whose batting averages were the furthest beneath their career marks entering the 2013 season.
Now, this isn't to say that every player on the latter list is bound to revert to his previous career batting-average norm and therefore be a fantasy bargain, nor that all on the former list will regress to his career mean and wind up an overpriced bust; that's part of the reason for the "explanations." To illustrate this effect, let's next turn the calendar back one year, and run the same report using 2012 statistics (400 PA minimum) comparative to careers entering the 2012 campaign (1,000 PA minimum).
As you can see, players like McCutchen, Beltre and Molina continued to excel, which is unsurprising considering both their reputations as well as the skills improvements they have made in recent years. Those are the examples -- and using 2013 players, Freeman, Hosmer and Molina appear similar such candidates -- to scratch from a "regression candidates" list. Conversely, Ryan, Youkilis and Suzuki were players who had shown waning skills -- with Upton, Betancourt and Suzuki again representing 2013 candidates -- who shouldn't have been expected to return to prior levels.
Still, use this data (or any of your own like it) as you attempt to find undervalued batting-average targets or overvalued players to avoid. The key is to discover an explanation for your theory; using blind stats is foolish.
And make sure to do said research in advance of your draft, so you can make a proper decision on how you want to attack the batting-average category. The last thing you need is to adopt a haphazard strategy, wasting unnecessary picks/dollars at the draft table, then lack the requisite knowledge to make the adjustments that would inevitably be needed in-season.
This research above is done so that you won't needlessly waste hours coming to this conclusion about batting average yourself. Now it's your call: What strategy will you use to attack this often-misinterpreted category?