Retrospective Player Valuation: NL Hitters

Welcome to my annual look at retrospective player valuation here at Baseball Prospectus. Over the next few weeks, I will be writing a series of posts examining how players performed from a fantasy perspective in 2016. This is the third post in a series of six. The first two posts in the series focused on AL-only leagues, the next two shift their focus to NL-only, and the final two posts will examine mixed leagues.

Before I dig in, here is a brief description of the charts below.

The $ value column is based on my Rotisserie-style, 5×5 formulas. It doesn’t exactly match anything in Baseball Prospectus’ PFM, but is derived using a SGP valuation model (something the PFM does offer). There are two important things to know about the values:

1) They are derived using the 168 best perceived hitters and the 108 best perceived pitchers (read most expensive or highest draft position) on Opening Day 2016, not the best 168 hitters and 108 pitchers at the end of the season.

2) The values of the 276 most expensive players add up to $3,120. This is important, and comes into play in the next two columns.

Sal is the average salary of the players. This column is derived from the prices in CBS, LABR, and Tout Wars, the three expert leagues that convene before the regular season starts and have three complete NL and AL-only auctions with no frozen players from which to derive data. While I would love to use more leagues to derive each player’s average salary, most Rotisserie-style leagues are keeper oriented and average salary data is useless for the purposes of this exercise thanks to auction inflation.

The +/- column subtracts each player’s salary from his earnings and shows whether he gained or lost his fantasy owners play money. Decimals aren’t displayed in the $ and Sal columns, which is how Jonathan Villar can cost $38, earn $11, and turned a $28 profit.

MG is yours truly, your heroic pricer and proud prognosticator since 2013. Another good reason to look back is to see if the fantasy expert you are following is good at what he or she does. It is easy to make predictions in March and never revisit those predictions or (worse) cherry pick the ones you got right and take a hollow victory lap. But how good are we at what we do? The prices below are from my fourth and final installment of Rotisserie style bids from late March 2016. I have always taken others to task for their predictions; now it’s time for me to face the music.

PK are Peter Kreutzer’s bids for AL and NL-only leagues, published at Patton and Company’s web site. It is useful to look back and see how you did, but even more useful to compare your predictions to someone else’s.

2015 shows what the player earned in 2015.

Last year saw one of the most “predictable” outcomes among the best NL hitters in a long time.

Table 1: Top 10 NL Hitters, 2016

Six of the 10 best hitters in 2016 were repeat performers from 2015, with Bryant narrowly missing the Top 10 (he was 12^th). These seven batters got paid, with an average of $33 in both salary and earnings. However, predictability doesn’t equal top-tier performance. For only the second time since 2010, no NL hitter earned $40 or more. Villar was better than 2014’s best hitter, Dee Gordon ($34 that year), but it was still a relatively “poor” season for the best NL hitters. It would have been an even poorer year without some of the inflated stolen base totals at the top.

The top four hitters on Table 1 combined for 144 stolen bases. This doesn’t take away from everything else these hitters contributed to their fantasy teams. You can’t appear on this table with stolen bases alone (Billy Hamilton earned 74 percent of his NL-only fantasy earnings with steals, and was the 19^th best NL hitter in 2016). However, in a steals-hungry environment, a one-category, elite performance like Villar or Marte’s helps. The biggest league difference between the AL and the NL is in stolen bases. Each steal in the AL was worth 39 cents. In the NL, each steal was worth 32 cents. It doesn’t sound like a lot but it adds up. Villar’s season would have been worth $42 in AL-only.

I already lamented my Tout Wars performance in an earlier column and won’t repeat that whining. It is worth noting that my bids were more competitive than I expected. I had the highest bid on Braun and Segura and tied the expert market on both Blackmon and Murphy. This was a noted improvement over 2015, when Blackmon was the only player I would have “purchased”. LABR was even more conservative than I was, and on a player-by-player basis, Kreutzer hung back, only “buying” a share of Arenado.

When the most expensive hitters are (relatively) predictable, a good deal of overlap between Table 1 and the next chart is to be expected.

Table 2: 11 Most Expensive NL Hitters 2016

Five of the 10 hitters on Table 1 find their way to Table 2. It isn’t these five hitters who are the headline, but rather the other six (there are 11 because Gordon and Votto tie in average salary). They all lose significantly, dragging the best hitters down to a $12 per hitter loss on average. If you were wondering back in March why I was so reluctant to push the top hitters in Tout up another tick or two, Table 2 is your best evidence. I didn’t have the highest bid on any of these players.

The NL experts spent more on the perceived best hitters than they have in years, even though there was no existing evidence that an earnings spike was coming.

While my results were good, was my process bad?

Table 3: 10 Most Expensive National League Hitters: 2012-2016

2016’s 10 best hitters were the most successful they have ever been since 2012, and by quite a large margin over 2013-2015. The expert market did what you would expect an expert marketplace to do. It chased the best hitters relative to what they earned the previous season. However, I would still contest that I did the right thing and that the experts were too aggressive, particularly in Tout Wars.

There is a difference between looking at the average salaries and looking at the individual high bids. Adding a dollar to my bid for Arenado might have helped me in Tout but it still would have left me two dollars behind the bidders in CBS and LABR, as well as Kreutzer. I could have purchased Gordon if I hadn’t chickened out in Tout, but it still would have left me behind in LABR. Average salaries give you an idea of how the market did overall but cannot predict how each individual league will play out. This particularly true among the NL experts. CBS and Tout are liberal with their spending, while LABR is conservative.

Returning to process…

The price shift in Tout this year was primarily due to the belief that the high number of teams that were “tanking” in the NL would lead to poorer outcomes for many position players and a wider distribution of earnings between the top and the bottom. If the best hitter earned $41 in 2015, then it stood to reason that the best hitter in 2016 would earn $43 or $44. A rising tide lifts all boats. Did it work out this way?

Table 4: Best Hitters, 2015 NL versus 2016 NL

It did not. In fact, the best hitters in 2015 were better than the best hitters in 2016. The first thirty hitters are nearly identical; it is not until the 31^st hitter that there begins to be some separation. Was this due to teams “tanking” the regular season? It is certainly possible. This idea is rather difficult to explore; it is safe to say based on Table 4 that spending more on the top hitters because of tanking wasn’t the best idea.

The tanking concept makes me wonder: did the most profitable hitters come from second division teams or teams that were expected to contend?

Table 5: Top 10 Profits NL Hitters, 2016

Seven of the 10 hitters on Table 5 are from second division teams. There was a great deal of opportunity that came from buying second-division players. Villar and Segura were obvious bargains, but beyond this pair there were more than a few players who cost next to nothing and turned a nice profit. Second division teams might not produce profits overall. Second division players can and often do.

There was more risk built into these hitters in 2016 than there was in 2015. An average salary of eight dollars in 2015 displays more floor than ceiling, and while Villar and Duvall may seem “obvious” now, back in April it was entirely possible that both could have fallen flat and been replaced by other options in Milwaukee and Cincinnati.

Stolen bases do not dominate this table like you might expect. As noted above, each stolen base was worth significantly more in the AL than in the NL. Turner and Owings certainly get a boost from their steals and it helps Galvis as well but Duvall, Story, Drury, and Murphy all finished in single digits in stolen bases. 2016 and 2015 provided a similar amount of stolen base value in the bargain bin. In 2015, the 10 biggest bargains stole 205 bases. In 2016, they stole 201. It is important to buy cheap steals in the NL, but it isn’t vital like it is in the AL.

The biggest difference between 2016 and previous years is that this is truly a speculative list of hitters. Murphy, Segura, and perhaps Pagan were guaranteed regular jobs; everyone else on Table 5 was not. Even Galvis, who held on to the Phillies’ shortstop job all season long, was expected to be supplanted by J.P. Crawford. In a weakened NL, spending more at the top didn’t pay off but speculating more at the bottom did.

If you want to see where Tout’s aggressive spending did the most damage, Table 6 is the place for you.

Table 6: Top 10 Losses, NL Hitters 2016

As dangerous as the “top tier” of players can be, it is often in Tier Two where the losses mount. Half of the players on Table 6 cost between $20-28. All of them failed to earn in double-digits. The best player in this awful group of $20-28 buys earned nine dollars. Injuries played a factor, as they always do. As was the case in 2015, the injured players on Table 6 hardly fall into the “injury prone” category. Stanton is the notable exception; everyone else who got hurt was the victim of a first-time or even a freak injury.

Once again, I took a pass on all 10 players in this group. If you’re wondering why I didn’t spend more than $17 on a single hitter in Tout (or $26 in CBS), this table offers strong evidence. There isn’t anything notable about any of the hitters on this table, but the ceiling for Harper, Stanton, or McCutchen didn’t change all that much between 2015 and 2016. Chasing one or two of the best hitters is fine. Chasing every one of the best hitters because “the worst hitters are going to be worse” is a fallacy. There are very few paradigm shifts that take place over the course of a single season in fantasy baseball. 2016 was no exception to this rule.

My conclusion to this piece last year was to avoid Stars and Scrubs in NL-only in 2016. It turns out I was more correct than I could have imagined, especially in aggressive bidding environments like CBS and Tout Wars. This isn’t so much because of the need of a balanced roster but rather because a winning team in a deep league needs a full complement of players on offense to succeed. Earnings stayed flat at the top while spending increased. By standing still and letting the bargains come to me in both NL expert leagues I participated in, I made a gamble that paid off across the board. The “worst” players in NL-only aren’t going to slip much if at all, particularly if they manage to get 500 at bats. Perversely enough, the lack of quality teams at the bottom of the NL led to more at bats for marginal players, which led to higher earnings for these players than was initially expected.

The complete list of Mike’s valuations can be found here, and will be updated as this series continues.