As Draft Day approaches, many owners must make last-minute keeper decisions, which are often more complex than the binary, keep-him-or-dump-him variety of decision-making. In many leagues, owners can’t keep players for free; instead, a keeper’s cost depends on his Draft Day acquisition price or his 2011 performance. For these decisions, it’s helpful to identify rebound players—those who PECOTA projects will increase in value in 2012—although owners in redraft leagues will also find it helpful to identify Draft Day bargains. When your competitors are short-sighted, as is often the case, they will undervalue these rebound candidates, and owners in snake drafts can see who might slip to later rounds as a result.
Continuing last week’s theme, I’m looking at those players whose value is projected to rebound the most in 2012, leaving out players who lost most of the season due to injury or who should return less than $5 in 2012. Because PECOTA tends to project players conservatively, it’s notable when it expects top-shelf players to increase in value for 2012. I was surprised to find some of the names below, and I expect that BP Fantasy readers will be too.
Ryan Zimmerman | 3B | Washington Nationals
Mixed: $18 | NL-only: $22 | PECOTA 3B Rank: 4 | 3B ADP: 4
|
644 |
.286 |
24 |
3 |
87 |
85 |
I covered Zim in the first January Keeper Reaper, noting that his 2011 performance undoubtedly suffered because of his torn abdominal muscle and that “all signs point to a return to top-shelf status for Zim,” tempering this with reminders of his recent injuries. PECOTA agrees that a rebound is likely, with an increase of more than $17 over his measly $0.51 return in 2011 mixed leagues, while his NL-only price should more than double from $10.53. Still, if Zim hits PECOTA’s projected.286/.350/.473 (.293 TAv), that’s a level that’s more like the 2006-2007 Ryan Zimmerman model than the more recent variety, at least on the surface.
Looking underneath the statistical hood shows that PECOTA expects Zim to continue controlling strikeouts while giving up a bit of patience. His projected 16.0 percent K% is in line with his improvements over the past few years; on the other hand, his 8.9 percent BB% would be below his career average but a rise from last season’s 8.0 percent. These point to a good batting average and a very nice OBP—one of his statistical strengths.
PECOTA ranks him sixth in OBP among third base qualifiers—the same spot his home runs also rank. All in all, Zimmerman should return to top-flight status at the hot corner for 2012 and beyond.
Alex Rodriguez | 3B | New York Yankees
Mixed: $18 | AL-only: $20 | PECOTA 3B Rank: 6 | 3B ADP: 6
|
572 |
.275 |
30 |
9 |
84 |
80 |
It’s hard to think of A-Rod as either undervalued or overlooked, but his 2011 campaign was a definite disappointment, partly due to the injured thumb that sidelined him for most of July and August. His productivity has also slipped over the past several seasons, though, while his DL time has increased. A-Rod hasn’t played 162 games since 2005 and hasn’t exceeded 150 games since 2007, averaging just 124 games (and 538 plate appearances) since then. His SLG has declined each season since 2007, and he hasn’t hit more than 29 home runs since 2008—the last season he hit over .300.
PECOTA sees him returning to form in some areas, but he’s still unlikely to hit .300 or crank more than 30 dingers again, and Seattle’s 40-40 man in 1998 shouldn’t even reach double-digits in steals. A-Rod’s declining batting average comes in part from some slippage in his strikeout rates, and PECOTA’s projected 19.6 percent K% sees that trend continuing, as that level would be his worst since 2008. His projected 11.7 percent BB%, on the other hand, will keep his OBP strong; like Zimmerman, that’s one of the reasons he’ll bring value to a fantasy squad, as he slots just ahead of Zim in that category among third basemen.
As for A-Rod’s increasing fragility, he should get some time at designated hitter this season, and his career reverse-platoon splits (.955 OPS against fellow righties, .945 OPS against lefties, against whom he also slugs 11 points lower) mean he could spell either Andruw Jones or new Yankee Raul Ibanez. Giving him time to rest his aches and pains should help bring profitability to his owners again—last year’s $5.15 mixed-league and $13.37 AL-only returns for A-Rod were well below any likely Draft Day bids. While you’re unlikely to sneak the Yankee Everyone Loves to Hate under anyone’s nose, you can still leverage the uncertainty of his recent struggles into extra Draft Day cash this season, especially in OBP leagues.
Joey Votto | 1B | Cincinnati Reds
Mixed: $36 | NL-only: $33 | PECOTA 1B Rank: 3 | 1B ADP: 3
|
686 |
.303 |
32 |
10 |
103 |
101 |
The 2010 NL MVP slid back a bit in 2011, going from incredible to just really good. Mixed-league owners expecting another $38 return had to be disappointed when he brought them “just” $28.63, but that’s not too shabby. PECOTA’s .303/.392/.530 (.319 TAv) projection for Votto in 2012 only represents a slight step backwards from his .309/.416/.531 performance in 2011 and—since this is a weighted-mean projection and PECOTA gives him a 63 percent Improvement Rate—Votto has a decent chance of beating it.
Votto’s K% and BB% have gotten better each of the last three seasons, although PECOTA expects him to produce around his career averages of 18.4 percent K% and 12.9 percent BB%. Those still excellent numbers regardless, particularly from a player with a .237 career ISO. His .227 projected ISO next season ranks him 17th among all players and ninth among first basemen. And, like Zimmerman and A-Rod, Votto’s OBP is another rare and valuable skill—only Albert Pujols is projected for a higher OBP among first basemen, and only Pujols and Joe Mauer are expected to beat him among all players.
Aside from bouts of anxiety and depression following his father’s death in 2008, Votto has been extremely healthy, both physically and mentally. In the three seasons since battling those issues, he’s missed just ten days of action, all of them in 2010. With his health and consistency, it’s easy to see why he remains an excellent value pick, though NL-only owners should note that he’s only projected to retain his $33 value from last season. If he played in a bigger market, his fantasy price tag would be much higher, so exploit the small-market minds of your fellow owners to snag a stable, healthy, and young Votto.
Eric Hosmer | 1B | Kansas City Royals
Mixed: $18 | AL-only: $20 | PECOTA 1B Rank: 10 | 1B ADP: 8
|
682 |
.284 |
19 |
8 |
85 |
85 |
Like Zimmerman, Hosmer also got some Keeper Reaper ink a few weeks back, when I determined that he looked undervalued. His ADP has risen since then, however, exceeding PECOTA’s expectations—an overvaluation owners should beware. As I explained in that January column, “The one deficit in Hosmer’s performance was his slugging, weak only by the standards of first basemen and by Hosmer himself, whose power was ranked by many as a perfect 80 on scouting scales.” PECOTA is also pessimistic about Hosmer’s power; his .444 projected SLG is 30th among first basemen, while his 19 homers rank him just 21st.
A further weakness to Hosmer’s projected 2012 line is in his .335 OBP, also 30th among first basemen, bleeding him of all three offensive stats expected from his position. With little slippage from his core skills and no dropoff in luck projected, it’s PECOTA’s tendency to undervalue strong seasons from youngsters that’s pulling his numbers down. Hosmer’s projected 16.1 percent K% and 7.3 percent BB% are similar to his 2011 numbers of 14.6 and 6.0 percent, respectively, and his .316 BABIP is right in line with his .314 mark from last season.
Though many Royals fans may stamp their feet and holler at PECOTA’s lack of recognition of one of their young talents, one need only look at Justin Upton to see a young, highly touted player who took several years to achieve consistency… or look at his older brother BJ to see a highly touted young player who has yet to do so. Hosmer, given just a scant 53 percent Improvement Rate, could beat PECOTA, but as I also noted in my column, his chance at a sophomore slump are good too. A full season of production could boost his value more than $4, but all signs point to Hosmer’s perceived value exceeding this, so bid cautiously on Draft Day; only keeper-league owners should be really hot on this guy for now.
Mark Reynolds | 3B | Baltimore Orioles
Mixed: $20 | AL-only: $19 | PECOTA 3B Rank: 3 | 3B ADP: 10
|
656 |
.234 |
35 |
11 |
91 |
83 |
Beginning the ranks of the truly undervalued—a theme I will continue next week—Reynolds is a poster child for overreaction and public perception. With his infamous quote about not caring about strikeouts, the power-hitting third baseman allowed the perception of his whiffs to cloud the other two True Outcomes he brings to the table. Despite leading the league in strikeouts four times (making baseball history in three of them), Reynolds has placed in the top six of the league in homers (and the top four in AB/HR) for each of the last three seasons, while placing sixth in the league walks in 2010. His OBP has exceeded his batting average by more than 100 points each of the past two seasons and by 93 points for his career.
Because his walks push his at-bats down, his batting average is more subject to the whims of BABIP, and Reynolds has seen two seasons of remarkably low hit rates on balls in play. His .266 and .257 marks from the past two seasons are well below his .301 career average and his .338 and .323 marks from 2009 and 2008, respectively. While the truth lies somewhere between these extremes (PECOTA sees his BABIP at an even .300 next season), Reynolds’ average could even be higher than the miserable .234 of his weighted means projection. Whatever his batting average, that’s not where Reynolds helps you, but most owners can’t look past that terrible line, the way you might be drawn to a lonely pimple on the otherwise flawless face of a supermodel. Most owners fail to recognize that his power production and his ability to swipe a few bags more than compensate for his batting average drag.
So, to me, it’s not surprising to see Reynolds ranked third among third basemen in value despite a boasting an average usually found among utility infielders. Reynolds is not only ranked third among third basemen, he’s projected to deliver a dollar less than David Wright and Jose Bautista.
While this is tantamount to heresy among Toronto fans, to say nothing of the fantasy community at large, it does show how little respect Reynolds gets. My recommendation to you is to talk up his strikeouts on Draft Day, like pointing at the pimple on Adriana Lima’s face just before her Victoria’s Secret fashion show. Remind your owners that he doesn’t care about his whiffs as a way to remind them that they should. Let them focus on the whiffs while you collect on all the fantasy value they’re missing.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
There doesn't seem to be much agreement about his BA though across the projection systems.
NL: http://pfm.baseballprospectus.com/index.php?cid=6435
AL: http://pfm.baseballprospectus.com/index.php?cid=6434
Mixed: http://pfm.baseballprospectus.com/index.php?cid=6430
If you want your projections look a certain way, though, and you know exactly how you want them to look ahead of time, what's the point of looking at the projections at all? Just go with what you think then. If you want them to match ADPs, why not just use ADPs?
Projection systems can take a lot of things into account and process them in a more precise manner than the human brain is capable of. That's not to say that they aren't wrong on occasion--by the mere nature of the sample size of a baseball season some portion MUST be wrong to a degree, even if a system was able to isolate every player's true talent level perfectly--but it's been shown that adjusting individual projections by hand that just "look" off actually decreases the overall accuracy of the system.
As Derek says, the projections take the "feel" out of the process, though the percentiles on the PECOTA cards (which are in mid number-crunch now) can also give you as sense of the outer edges of a player's potential performance. If you feel that a player is going to exceed or fall below his PECOTA projection (something that we BP Fantasy writers will sometimes express), that doesn't mean the system is wrong. It just means the guy is under/overperforming statistical expectations--and it's what allow us at BPF to provide insight and tweaks to the PECOTA projections.
One other thing I like looking for in the projections is the number of plate appearances. Some projection models may predict a player to miss the equivalent of 30 games, such as the Alex Rodriguez PECOTA projection. ARod now has 2 ways to outearn his dollar value: he can simply perform better than what was expected in the same number of plate apps, or he can stay healthier than was forecasted and accrue more counting stats due to the higher number of PAs. I'm not saying it is safe to project 155 games for ARod because it isn't; but I think that there is sometimes value in grabbing a player like this, especially in leagues that allow regular transactions for injuries. You may end up with something like 130 games of ARod + 30 games of a bench player at 3B = a pretty good stat line for that one roster spot.
And you're point about A-Rod is spot on. I've written about this before too and tend to get asked that sort of question in my chats pretty frequently (Braun is a guy this really applies to this year), and in mixed leagues, this is a really important concept.
I will re-ask the questions. Is Bautista's $22 value compared to Reynolds $20 realistic projections of value & what will cause Bautista's value to drop from $35 to $22. If not, you should be inspecting your system.
Yes, I am fustrated at what I am seeing in PECOTA projections for 2012. Bottom line, what I want as a user is accurate forecasts. At this point, my confidence is not very high in the changes to PECOTA.
His PECOTA value, I would imagine, is going to have been lowered because of the increased weighting on past seasons. As Mike alluded to, this is a place where qualitative analysis can supplement the projections, and that's why we write articles. The projections aren't the end-all-be-all, but they are indeed going to be accurate on the whole. That said, all they take into account--all they *can* take into account--is the numbers. If there are other reasons why Bautista is not the same player that he was a few years ago--and I believe, in his case, there are--then we can make adjustments based on that. But qualitative information is difficult and/or impossible to systematically take into account for a projection system (unless you have a database of detailed scouting reports and such like MLB teams might), and that's true of all of them, not just PECOTA.
I don't think it's fair, however, to judge a system's accuracy before the season begins. If we knew for certain how a player would perform, then what would we need a projection system for in the first place? Maybe Bautista will outperform his PECOTA this year, but the testing that Colin, Rob, and crew ran showed that, on the whole, extra backweighting increases the system's accuracy. Judging a system based on the anecdotal evidence of a single projection (or a handful) that doesn't pass your eye test doesn't necessarily make it inaccurate.
The reason there are screams of horror over PECOTA outlier projections is that over the last couple of years, some of those projections have been obviously wrong. The Kila/Bowker problem of last year was severe and there were articles citing Kila as good based on PECOTA. This was despite the repeated and persistent observations by readers that this was wrong. I specifically outlined why the system was doing this, and why it was wrong, and I did so before Kila or Bowker or any of the other 26+-aged minor leaguers who were being misprojected got a 2011 plate appearance.
I would therefore be cautious in saying it's reader error to dismiss projections that look wrong by eyeball.
PECOTA now seems to have very serious problems with the very young. The comp lists are blatantly fouled. I believe the book PECOTA's are even worse for them. I'm ready to bet the over on Mike Trout, Freddie Freeman, and Eric Hosmer, among others. (There are other problems, like the Chipper problem.)
Maybe I'm wrong about this year; that's always possible. But, again, before you discount your readers' concerns about PECOTA, you should remember just how right your readers have been in the past.
--JRM
I guess for my purposes, I would rather have "accurate" than "conservative", and it strikes me as a long time BP reader that while PECOTA changes (it feels like it has changed so many times that the current version should be called "PECOTA 5.0" something), a constant is that each year, as I scan the numbers, I always have the sensation that I would make FAR more "over" bets than "under" bets on player performance.
Injuries are the wild card.
One approach I have gravitated toward in recent years is to calculate each player's 162 game career average production (I play in a draft points league, where every active player's points accumulate over the course of the season, with far more categories than traditional 5x5 rotiss formats). From this, I have a feel for who is the most productive per game or per plate app.
I then make several adjustments - for playing time based on injury and likely opportunity (example - Josh Hamilton is a top 10 player in terms of points per plate app, but he is a 130-140 game player given his health track record), and for where a player is in his career -- growing, prime, or past prime. I also try to factor in some type of mean reversion, for players coming off seasons that were clearly well above or below established levels of performance.
I think BP is great--but the constant adjustments in PECOTA, and its "conservative" bias sometimes make me wonder how usable the model really is for the typical reader. Just got the book yesterday and of course stayed up way too late reading it --- and again was struck at how many projections seem markedly below what I'd expect to be a reasonable over-under line.
one other comment --- it seems that in years past, the "improve" and "breakout" percentages gave one a better sense of magnitude. This yar, it seems like very few hitters are projected to improve, and many are projected to regreess. That strikes me as pretty counter-intuitive.
Thanks for any comments and keep up the great work! TB
I agree with the conservative nature of PECOTA. There was 40+ players with .300+ TAv in 2011 (A down year for offensive production. PECOTA projects only 20 players will have a TAv greater than .300 in 2012.
PFM, in a 12-team, 2-catcher league with 5x5 and PosAdj set to OFF, has 27 players with 3B eligibility worth positive value, and only 13 catchers with positive value. However, only 18 or so 3B are needed on rosters compared to 24 catchers.
Catcher is definitely scarcer than 3B.
That's the point of putting Positional Adjustment ON--to reflect the weight of a player vs. a potential positional replacement. Otherwise PFM disregards positional scarcity.
With it on, the average value of a catcher (top 25) is $10.13; the average value of 3B (top 25) is $9.57. On the other hand, the lowest-ranking 3B is Chipper Jones at $1, while the worst C is Carlos Ruiz at $0.72. And 17 players at either position will earn $5 or more.
Bottom line: they're both very scarce position, with 3B getting a slight edge.