Perception is a funny thing. After Buster Olney reported that the Nationals were willing to trade Josh Willingham again, I had a conversation with a couple of people about how well he would fit in with the Mariners. They’re not particularly new-school types, but I respect their opinions and wanted to know what they thought of him. Both thought he was okay, a decent role player who could inject some power but shouldn’t be thought of as an everyday player.

One comment in particular stuck out to me, though – “If you think he’s a substitute for Jason Bay, think again. He’s not at that level.” Curiosity piqued, I decided to look and see just how large the gap is between Willingham and Bay. And, to my surprise, I learned that the answer is “not much.”

They were born five months apart, Bay at the end of ’78, Willingham at the beginning of ’79. Bay made the majors as a 24-year-old, Willingham as a 25-year-old. They both share the same skillset: power hitting outfielders who draw walks and produce enough runs at the plate to compensate for below average defense.

But even beyond the generalizations, they’re similar. Look at their respective wOBAs plotted on the graph below.

In 2004 and 2005, Willingham racked up less than 30 plate appearances, so even though he’s got a dot on the chart, the performances aren’t meaningful. He wasn’t really a major league player until 2006, so focus on the years since. Since 2006, they have been very similar hitters. A weighted average of Bay’s wOBA since then is .377, while Willingham’s is .367. 10 points of wOBA over a full season is approximately 5 runs of offense.

This is not a small sample. We’re talking about a couple thousand plate appearances, and the difference in offensive production between the two can only be categorized as minor. Yet, there is the perception of a huge gap between the two. Bay is a middle of the order slugger, while Willingham is a nifty role player who can mash lefties. That’s the narrative, and four years of facts hasn’t been able to change it.

In reality, the difference between them is measured in fractions of a win. Bay is the better player – he’s outhit Willingham in three of the last four years, and his most recent performance came in the AL East, while Willingham has collected all of his performances in the NL. There are enough gaps to distinguish between them and say Bay > Willingham. But it’s a very small gap.

And here’s the funny thing – you guys don’t even really disagree with this very much. The Fan Projections here on the site have Bay as a +3.1 win player for 2010, while Willingham is projected as a +2.9 win player. Even if we won’t admit it, we think that they’re basically the same player going forward, just giving a slight edge to Bay.

So, with all due respect to my old school friend, I have to disagree. Josh Willingham is a substitute for Jason Bay. They’re practically the same player.

One of the main uses of value analysis is to try and find market inefficiencies. Ten years ago, on base percentage was undervalued, so teams loaded up on high walk guys that scouts didn’t care for. More recently, defense has been the undervalued asset, so teams have gone after guys who can turn balls in play into outs.

Everything is cyclical, though. As more teams pursue what is currently undervalued, it becomes more fairly valued, and the competitive advantage goes away. At some point soon, defense will probably become fairly valued again, and the teams who are loading up on good defenders will be looking for some other way to spend their money.

What will the next big inefficiency be? It’s impossible to predict, of course, but I have a guess – old players.

We’re currently in the midst of an age where a lot of teams are operating on reduced budgets, and have shifted towards trying to keep costs down by going with more inexperienced talent whose salaries are deflated by their lack of service time. Teams like Tampa Bay and Oakland are continually attempting to replenish their farm systems to ensure a never ending pipeline of cheap, effective major league players that they can pull from.

As more teams have turned to this model, young talent has become increasingly expensive to acquire. The relative value of experienced veterans has taken a hit as teams have turned towards cheaper labor, even accepting downgrades in on field production in order to keep their payrolls in check.

This has led to yet another winter where guys over 35 are having a hard time finding jobs. It’s not just Johnny Damon, though he is a good example of this effect. Over the last few years, we’ve seen numerous productive-yet-old players pushed into retirement against their will, ranging from the likes of Kenny Lofton, Ray Durham, Frank Thomas, and Jim Edmonds.

Edmonds, of course, is now attempting to get back into baseball, and seems like he may be able to convince some team to give him a job. But he had to publicly ask for a minor league contract at the Cardinals FanFest event in order to begin the discussion – no one was beating down his door.

Teams have become cautious with the contracts they give to aging players, not wanting to get burned paying too much to a guy who may end up not having anything left in the tank, but I feel like we’re passing the point of caution and shifting towards a market failure. If a guy is a good player at 35, you should not expect him to be useless at 36. Yes, you regress his projection for aging, but players who go from good-to-terrible in a single season are the exception, not the rule.

Given the contracts that quality older players have been settling for over the last few years, I think we may see teams in the market for value increasingly going for the graybeards.

Last year, the Yankees decided not to retain an aging corner outfielder in an attempt to upgrade their defense and save money, as Bobby Abreu was seeking a significant paycheck for his durability and consistent offensive abilities. He found the market for his services nearly non-existent, however, and eventually settled on a one year, low money deal with a club in the AL West that was in need of an offensive boost.

Fast forward to today, and history is apparently trying very hard to repeat itself. This year’s aging corner outfielder that the Yankees are not retaining is Johnny Damon, and just like Abreu, he’s finding out the hard way that teams do not value him like they used to. This afternoon, Buster Olney reported that Damon’s best offer may come for $4 or $5 million on a one year deal from the Oakland A’s – an AL West club that could use an offensive boost.

2008 Abreu and 2009 Damon had very similar offensive seasons.

Abreu: 684 PA, .296/.371/.471, 10.7% BB%, 17.9% K%, .176 ISO, .368 wOBA
Damon: 626 PA, .282/.365/.489, 11.3% BB%, 17.8% K%, .207 ISO, .376 wOBA

Basically the same kind of production, once you adjust for the change in Yankee Stadiums. They’re similar hitters at similar spots in their career. Both have declined from their prime, but not fallen all that far.

There is one big difference between 2008 Abreu and 2009 Damon, however, and that’s on defense. Abreu was victim of the backlash against bad defensive corner outfielders. From 2004 to 2008, his UZR/150s as an LF were -10, -5, -14, -3, and -25. At the very best, you could surmise that he was below average, and realistically, he was more likely to be terrible. He was a borderline DH, the kind of defender who shouldn’t play much outfield except in case of emergency.

Damon, on the other hand, throws like a girl but is otherwise a decent defender. It’s true that his 2009 UZR is not good, but his total UZR over the last three years at the position is a +5, and as we try to remind people frequently, larger sample sizes are needed when evaluating defense.

Damon is clearly a better defender now than Abreu. He’s not great, and he may even be reaching the point in his career where he’s below average, but he’s simply not a guy that costs your team a ton of runs in the field.

Abreu was a bargain on a one year, $5 million deal with the Angels, even as he proved that he didn’t really belong in the outfield anymore. Damon, though, is basically the same hitter, just with better defensive skills, and he might have to settle for less than what Abreu got? This is a market correction gone way too far.

Even with the reduced costs of wins, Damon is easily worth $8 to $10 million for 2010. Just like with Abreu last year, teams will be kicking themselves in a few months if they let him sign for peanuts. There are enough clubs out there that could use a +2 to +3 win left fielder that this level of disinterest is simply a market failure.

A few years ago, the NFL saw a shift in how running backs were used. Rather than sticking with the single starter model, many teams shifted towards job shares, preferring to let two guys split the playing time in order to keep both more fresh and healthy than either could be by themselves. It became more efficient, in some cases, to have multiple players at the position than a single player getting all of the playing time.

We may be seeing something similar in baseball with the designated hitter. If there’s been one theme to the last two free agent periods we’ve seen, it has been a lack of interest in guys who cannot play the field, or who play it poorly. Teams have significantly pulled back from aging offense-only types, finding value in other types of players at the expense of some legitimately good hitters.

But it doesn’t appear to be just a renewed focus on the value of defense. Several teams are all but abandoning the idea of having a full-time designated hitter to begin with. The White Sox yesterday admitted that they wouldn’t be bringing Jim Thome back, because they just didn’t have the at-bats for him. Instead, their DH position will be filled with a rotation of guys that cannot hit as well as Thome. Yet, they see value in having the flexibility to use the spot for various players.

The Tigers, Mariners, and even the Yankees appear to pursuing similar strategies. Detroit has declined to bring in an additional power hitter, preferring to use their DH to give Carlos Guillen, Magglio Ordonez, and Miguel Cabrera time away from their gloves while keeping their bats in the line-up.

The Mariners are going into the season with LF and DH being a job share between various players, depending on where Milton Bradley is healthy enough to play on any given day. And while the Yankees were willing to spend $6 million to bring in Nick Johnson, he’s the kind of guy who simply can’t play every day, which will allow the Yankees to use the DH spot to rest Jorge Posada and any other veteran who needs it.

Teams are choosing to increase their flexibility, even if it comes at the expense of some production. Increasingly, teams want the option to use the DH spot as a pseudo off day for their regulars, or as a fall back plan if their banged-up position player is unable to acceptably field his position. With the move towards 12 man pitching staffs, limited bench sizes put a premium on roster flexibility, and teams are reacting by devaluing players who can’t play the field.

Given that there are only 14 designated hitter jobs in baseball to begin with, this is bad news for aging players. If even half of those teams move towards a rotating-DH plan, you’d be left with only a half dozen or so full time, offense only players. To get one of those jobs, you’d have to be a monster of a hitter, a David Ortiz in his prime kind of guy. And once you decline even a little bit, your chances of getting another job go out the window.

It will be interesting to see how teams react to this emerging DH usage.

Back in September, I was lucky enough to grab lunch with Jonah Keri and Patrick Sullivan at The Friendly Toast in Cambridge, where we mostly marveled at how big the kitchen in that place must be to support a roughly 142 page menu. Once we got our food figured out, the conversation shifted to baseball. Curiously enough, a decent part of the discussion between an Expos fan, a Red Sox fan, and a Mariner fan focused on the potential moves the Baltimore Orioles might make.

The Orioles were interesting to all three of us. Through some savvy transactions (okay, mostly the Erik Bedard trade) and good drafts, the O’s have amassed an impressive group of young talents. And after years of overspending on bad veterans, the team had been financially sound, giving themselves a significant amount of financial flexibility heading into the winter.

Flush with cash and some obvious holes, it seemed like an off-season of good spending could setup the Orioles to be an interesting team in 2010, even pushing into fringe contender status if their young talent all matured quicker than expected. Especially as another strong buyer’s market emerged, this appeared to be a perfect opportunity for the Orioles to add some key pieces that could grow with the core already in place.

Instead, they ended up spending $25 million in 2010 (and $6M in 2011) to acquire Kevin Millwood, Garrett Atkins, Miguel Tejada, and Mike Gonzalez. They filled four holes, but the return on investment seems… underwhelming.

Let’s start with the two infielders. The signing of Tejada and Atkins fills the two corner spots, with Tejada expected to play third while Atkins shifts to first base. By pretty much any projection you use, both are expected to be roughly league average hitters – CHONE expects a .335 wOBA from Tejada and a .324 wOBA from Atkins. Given that both are adjusting to new defensive positions and on the wrong side of thirty, forecasting their defensive value is a bit tricky, though I wouldn’t expect either to be particularly good at their new spot.

Tejada is probably a league average player while Atkins is a bit below that. Overall, the production that the duo will offer the Orioles should be around +3 wins. Given that the market is paying about $3.5M per win on one year deals, Baltimore paid almost exactly what you’d expect. They didn’t overpay, but these weren’t bargains either.

On the pitching side of things, they inherited $9 million of Kevin Millwood’s contract, then gave Mike Gonzalez $6 million per year for both 2010 and 2011 and surrendered their second round pick for the right to do so. Millwood, like Tejada, is a decent bet to be a durable, league average-ish player for the next year, while Gonzalez will offer value out of the bullpen, though my position on relievers being less valuable than the market believes is well known. Again, you’d expect about +3 wins of value from the pair this year, but given the higher acquisition cost, Baltimore overpaid on both of these guys.

But, beyond just the dollars per win stuff, there’s the real head-scratching part of all this. The Orioles added about +5 to +6 wins to their roster for 2010, but that’s not a big enough improvement to make waves in a tough AL East, and only one of the players – the relief pitcher, no less – is under contract beyond this season.

They improved their team’s chances of being decent this year, but have done nothing to improve the team’s future. There was no multi-year deal for an undervalued position player (say, Adrian Beltre), who would provide value both in 2010 and beyond. The deals that were made were handed to a couple of 35-year-olds, a 31-year-old reliever, and a 30-year-old.

There’s just very little upside in any of these deals. Even in a best case scenario, Tejada and Millwood continue to be above average players, make the team good enough to fight the Rays for third place in the AL East, and then file for free agency again next year. Or, I guess, you could hope to trade them at the deadline for a prospect if they’re playing well, but then that raises the question of why the team didn’t just take the $15 million they gave those two and plow it into the draft and international free agents in the first place.

This just isn’t the off-season that I thought the Orioles had in them. This is a safe, low risk, low upside winter that doesn’t really change the fate of the franchise much at all. They got a little bit better for next year, but lost an opportunity to really help the 2011/2012 teams that could actually be quite good.

Over the weekend, in a thread over at Tango’s blog, the idea of a “Sabermetric Library” was raised. As noted over there, one of the positives of the academic journal process is to catalog the work that has been done, making it easily searchable for future readers who are not following the discussion in real time. The statistical analysis crowd doesn’t have that kind of formal structure, which makes it difficult for those who come later to catch up on what has already been done.

Rather than employing a full time “librarian” to keep up with the most recent work, I thought perhaps we could just attempt to crowd-source this idea. So, that’s what we’ll attempt to do in this thread.

In the comments below, I’d like to encourage you to think back to influential articles that you’ve read about the game, and if you can, link to them. If they were written a book, link to it at a particular bookseller of your choice that carries it. If you can quickly summarize the conclusion, even better.

It doesn’t have to be an epic research piece that changed the face of analysis (such as Voros’ piece on DIPS), though those obviously fit in here, too. But if there is a blog post somewhere that explained something in a way that allowed you to understand it for the first time, link to that. If there was an interesting discussion on a popular topic (Blyleven for the HOF, maybe), then link to that.

The goal would be to populate the comments with enough resources to allow someone to go through and read a Best Of The Sabermetric Community collection of writings. There are a lot of good writers out there doing good work, but given the size of the internet, some of it can get lost in the shuffle. Let’s preserve the pieces that deserve to be kept alive, and at the same time, create a resource for those who come along in the future to find out about the work that has already been done.

In order to keep the layout easy to read, I would ask that you refrain from commentary about this post. Please limit comments to the format of linking to important pieces, with necessary comment about that piece as an abstract of sorts. If this takes off as I hope it does, we’ll do a discussion thread on another day about potentially culling the list, giving space for people to argue for or against any of the linked pieces below.

Dayton Moore’s run of transactions since taking over as Royals GM has been so bad that the natural reaction to any Kansas City move now is scorn and derision. So, not surprisingly, when it came out last night that they had signed Rick Ankiel to a one-year contract, there were critics who immediately pointed to the career .311 on base percentage and laughed. I think this is a good deal for the Royals, though, and Moore should get credit for making his team better.

Ankiel’s story is well known, so we’ll skip the whole pitcher-turning-into-a-hitter aspect and just focus on the last few years. As an outfielder, he’s displayed a couple of strong skills – serious power, above average range, and of course, a strong throwing arm. He’s rough around the edges (most obviously with pitch recognition), which is to be expected from a guy who spent most of his life training to pitch instead of hit, but the strengths outweighed the weaknesses.

In 2007, he posted a .364 wOBA, then followed it up with a .360 wOBA in 2008. The power dwarfed the relative lack of walks, and he was a well above average hitter. Last year was a disaster, certainly – the power dried up and his wOBA fell to .288, acceptable only for a middle infielder or a catcher. But injured players don’t often play well, and Ankiel was clearly at less than 100 percent for most of the season.

Projecting guys coming off of injuries is always harder. You don’t know to what extent their performance suffered due to the injury, so you have to give that season less weight than you usually would, even though it is the most recent data point. For Ankiel, I think realistically, we can expect him to post a .350ish wOBA if he’s healthy, even though the projection systems will spit out a lower number than that.

For $3 milllion, that’s a good piece. And unlike last year’s splurge on a veteran power hitter, Ankiel actually fills a hole for Kansas City. They simply didn’t have a right fielder before this move, so Ankiel is not blocking off any other talented players. He simply makes their team better, and does so at a low cost.

Dayton Moore has made a lot of bad moves, but this is not one of them. He made his team better without spending significant resources to do so. Kudos to him for a nice signing.

Since Eric Byrnes was designated for assignment last week, a variety of writers have noticed that his offensive downturn the last couple of years is entirely driven by tremendously low BABIPs – .226 in 2008 and .229 last year. His performance in the numbers that are usually more indicative of talent level (BB%, K%, ISO) are basically in line with his career marks. And while BABIP is certainly more in the control of a hitter than the pitcher, it still can vary significantly from year to year. So, if the core skills are still in place and BABIP can be driven by luck, it’s feasible that Byrnes could bounce back and become a solid offensive contributor again.

Maybe. But he is an example of why you can’t just look at a hitters BABIP and regress to a league average mean, because Byrnes has one particular skill that destroys his ability to get hits on balls in play; he is the master of the infield fly.

Since 2002, Byrnes has hit 244 infield flies, more than anyone else in baseball. Vernon Wells is second on the list with 222, but he has 1,300 more plate appearances than Byrnes over that time frame. Steve Finley is third on the list with 161, almost 100 pop-ups behind Byrnes in only 200 fewer trips to the plate.

7.67% of all of Byrnes’ plate appearances since 2002 have ended with an infield fly. Tony Batista is the only other player in that span with greater than seven percent of his PAs resulting in a pop-up. Not surprisingly, his BABIP since 2002 was just .244.

In fact, these numbers actually understate how pop-up crazy Byrnes has been of late. In 2002 and 2003, just 15% of his total fly balls stayed on the infield. His final full season in Oakland in 2004 saw the rate jump to 20 percent, finishing with the 3rd highest rate in baseball that year (behind Batista and Juan Uribe). But after leaving the A’s, he took it to another level.

2005 – 23.9% infield fly percentage – #3 in baseball.
2006 – 24.8% infield fly percentage – #1 in baseball.
2007 – 26.7% infield fly percentage – #1 in baseball.
2008 – 22.2% infield fly percentage – #3 in baseball.
2009 – 25.3% infield fly percentage – #1 in baseball.

Over the last three years, among players with at least 500 PA (total, not per season), Byrnes is #1 in IFFB% by a mile. His 25.6% mark is followed by Jeff Mathis at 21.1%. Eric Chavez and Chris Burke are just over 20%, and then there’s Mark Ellis at 18.4%. No one else in baseball is over over 18%.

We cannot look at Byrnes’ low BABIP and conclude that he’s gotten unlucky. His BABIP is a reflection of the fact that he is constantly hitting 100 foot flies that are easily grabbed by an infielder and have no real possibility of becoming a hit. He’s not hitting lasers at people. This isn’t bad luck. This is bad hitting.

Byrnes may have another good year or two left in him if he can get healthy again, but do not project him back to anything close to a league average BABIP. Given his proclivity for the pop fly, he’d be lucky to crack .260 or .270.

The Mariners have reportedly agreed to a 5 year, $78 million contract extension for Felix Hernandez. Mariner fans (including this one) are obviously thrilled with the deal from an emotional standpoint. We’ve been attached to King Felix since he was 17, and after watching the likes of Randy Johnson, Ken Griffey Jr, and Alex Rodriguez depart, it is refreshing to see the team finally lock up a superstar to a long term deal.

But, I’ll keep my celebrating to a minimum over here. Let’s take a look at this deal from the perspective of what Hernandez surrendered by signing this deal.

As a second year arbitration eligible player, Felix was in line for a significant paycheck with or without this contract. Based on his 2009 salary of $3.8 million, his performance last season, and what comparable (yet still inferior) pitchers got in their second arbitration year, he would have been expected to come in around $9 or $10 million in salary for 2010.

So, in reality, his 5 year, $78 million deal is more like 4 years and $68 million, because he was already getting the first 10 million. Or something really close to it. How does $68 million over the next four years, or about $16.5 million per season, compare to what we think he would have gotten had he not signed this deal?

Given a realistic projection for his performance, you’re looking at something like a $15 million payout in arbitration next year if he stays healthy. Even if he gets hurt (non-labrum tear category), he’s likely looking at another $10M-ish payout, as we’ve seen teams offer arbitration to players like Brandon Webb and Erik Bedard even coming off lost seasons, because the upside is high enough to justify a one year contract.

Alternately, a healthy Felix could have opted to sign long term next year, following in the steps of Johan Santana in buying out his free agent years at the end of his fifth full season. Johan got 6 years, $137 million. Felix doesn’t have Santana’s track record, but he’s younger and has better raw stuff, so you’d have to think an extension after next season would have been in that neighborhood.

To buy his long term security a year earlier in an attempt to avoid what happened to players like Bedard, Webb, and Ben Sheets, Felix signed at a pretty significant discount. He turned ~$20 to $25 million in payouts plus free agency at age 25 (or $10 million in payouts plus a ~$20M/year AAV extension in a year) in exchange for about $53 million over his first three free agent years.

$17 to $18 million a year for three free agent years is a really large discount over what his expected payouts for those years would have been in in the future. He sold the risk of injury in 2010 for probably something between $5 and $10 million per season from 2012 to 2014.

Given the risks inherent with pitching, it’s understandable. Felix’s family is now set for life. But he left a pretty decent amount of money on the table by signing now. Not that I’m complaining.

Over the weekend, Chuck Brownson did a post over at THT Live, showing the premium that a few teams have been willing to pay for closers this winter, even as the market for nearly everyone else collapsed. Despite a pullback on spending, some clubs still value late inning relievers enough to pay a premium to acquire guys who have had success in that role.

WAR, as you probably know, doesn’t think much of relief pitchers. The very best relievers in the game are generally worth +2 to +2.5 wins over a full season, or about the same as an average everyday player. This has caused quite a few people to state that WAR doesn’t work for relievers, because the results of the metric don’t match what they believe to be true about relief pitcher value. I think it works just fine.

While the quality of their work is very high, the quantity is low, which limits their total value. It’s nearly impossible to rack up huge win values while facing less than 300 batters per season. Yes, each of those batters faced are more critical to a win than a regular batter faced, but this is accounted for in WAR.

The average Leverage Index of a closer is about 1.8, meaning that each plate appearance is about 80 percent more important than an average PA. We give the closer credit for half of that, based on the principle of chaining. Because relief pitchers are mostly fungible, and can move from one role to another if needed, replacing a closer is not the same process as replacing a starter or a position player. If a team’s closer gets hurt, they do not then call up a replacement level reliever from the minors and use him to close out games.

Instead, that replacement level reliever who gets called up from Triple-A becomes the mop-up guy, and everyone gets promoted one slot; the setup guy moves into the closer role, the middle reliever becomes the setup guy, so on and so forth. The high leverage innings are handed to the remaining best pitchers in the bullpen, not the guy replacing the lost relief ace.

Because of this, it does not make sense to give a reliever full value for the credit of his leverage index. That leverage does not belong to him, nor does it go with him when he goes away. Remember, this is Wins Above Replacement, and the guy who replaces a closer is generally a pretty good pitcher. We want to measure the marginal win of a player over what the team would lose if they did not have him, and in the case of relief pitchers, the replacement level is very high.

That a half dozen major league clubs are overvaluing proven relievers does not mean that WAR doesn’t work for them. It’s just an arbitrage opportunity for the rest of baseball.