Over at Athletics Nation I had been doing the Possible Trade Partners series leading up to the A's trade of Matt Holliday and I originally posted this piece there, though I thought it would be of interest over here as well. Reflecting on both Holliday trade, I believe they said several interesting things about market forces in baseball. After the Holliday trade this fall I remarked that it was possible that Beane was merely trading quickly depreciating assets for for a slower depreciating asset serving as a baseball version of George Soros engaging in international currency arbitrage. If that is the paradigm that Beane took it is also a good framework for evaluating the success of Beane's moves.
Two markets and four assets need to be considered when evaluating the first Holliday trade. The two markets are for closers and corner outfielders. While there were four players dealt at four positions (corner OF, closer, CF, starter), there are only two relevant markets to consider. Especially in Gonzalez's case, this is not due to a lack of value but rather the lack of a market for high grade position prospects other than in an acquisition of proven talent. With the exception of the Young/Garza deal, in recent years top prospects have not been swapped to merely redistribute positional talent between systems. Even swaps like the Golson/Mayberry are generally of failed or failing prospects that need protection from the Rule 5 draft. If you are not in contention you acquire your position prospects through the draft, international free agent signings, or by moving established assets.
The same could be said about young team controlled starting pitching. Thought there certainly are trades like the Jason Hammels deal earlier this year where back of the rotation candidates run out of options and are sent away at the end of spring training despite their team having control, and the Edwin Jackson trade when the prospect of arbitration make back of the rotation candidate in a full organization more pricey than they are worth. There are more examples of this kind of trade in the last several years (Bannister for Burgos comes to mind though that was arguably an acquisition of established talent for the Mets) but it is hardly a robust market.
The lack of both markets leaves little recourse for a team who does not have valuable established commodities. We all know there is a market for trading promising prospects for established talent, and back of the rotation candidates such as Smith or Mortenson make excellent final pieces. The other way that Beane could use these prospects is in a three way deal, but those deals are hard to consummate and pursuing three way deals is a great way to undermine the value of your prospects (asking anyone if they want them undermines future leverage). Therefore, if Beane felt that Smith and Gonzalez would be commodities that would under-perform their market value in 2009 and beyond, he had little option but to use them for acquiring proven talent. With reports of pressure from ownership to contend this year, it makes sense to trade for established assets.
So the first question when it comes to Gonzalez or Smith is would their commodities under-perform their market value in 2009. The second question is would that under-performance be less than the exchange cost of those prospects. Prospects to veterans to prospects is not an exactly fungible conversion and there are some costs to acquiring the player you want. I think it is very reasonable to have expected Gonzalez and Smith to under-perform.
Smith is the easy one. He had fragments removed from his elbow in October of 2008 and had had issues with shoulder soreness in 2007. A huge amount of the value of a 5th starter is to eat enough innings that prevents the bullpen from being taxed. Furthermore, the ability of your 5th starter to stay healthy is important to minimize the number of innings that replacement level pitching has to absorb. Smith walked over 4 batters per 9, had a .050 lower than league average BAPIP, and struck out 1.5 batters less per 9 than the league average. He was slated for a major regression to the mean as all the major projection systems (ZIPs, James, Chone) all projected him with a ERA of greater than 4.88 in 2009. Smith was going to under-perform and Beane needed to sell high on this asset.
Carlos Gonzalez is much more of a mixed bag. Everyone agrees that he has the tools to make the decision to give him up look stupid, and yet he hasn't maximized on them. Gonzalez' hadn't had a stand out offensive campaign since his 20 year old season in High A Lancaster. Even that may have been overvalued. Lancaster is an absolute launching pad in the California League and using the park neutralizer on Minorleaguesplits.com, Gonzalez's OPS drops 70 points. Furthermore, his MLE have staid constant at what his MLB production level, around a .650 OPS. Additionally he was not even an above average hitter in AAA in 2008 creating -2.8 batting runs bellow average at that level. Even his power went backwards after leaving High A, as each additional level accompanied falling ISOp Along with the lack of development between his 2007 and 2008, there were rumors that Gonzalez was not interested in accepting advice from the coaching staff, particularly about his patience at the plate. Gonzalez will never become an excellent player if he does not learn to walk more, and while his walk rate increased with the A's in AAA it plummeted to unacceptable levels while in the majors. While Gonzalez was clearly rushed, his previous walk numbers do not suggest that he will be able to sustain an acceptable walk rate unless his batting approach changes, something that seemed unlikely given his rumored hostility to the idea.
Many cite Gonzalez's excellent defense while he was with the Athletics in 2008 as a reason that the A's should have kept the budding center fielder. However, all fielding metrics require a large sample size when evaluating performance and are subject to much statistical noise. For example, Carlos Gonzalez in about 75 full games put up essentially the same UZR score that Ryan Sweeney has put up despite Sweeney looking terrible with his UZR in center field. That coupled with numerous scouting reports that questioned Gonzalez's range when he filled out. Therefore, while it was certainly clear that Gonzalez has the potential to become a plus defensive center fielder, the likelihood is not enough to outweigh questions about his bat.
Alternatively, the market for closers is well established. Huston Street was a known commodity having been shopped without much interest the previous deadline. The market for closers has been discussed for years as being inflated beyond its value, a fact that has been on display since the 1999 trade of Billy Taylor. Unfortunately for Street's value, he had a penchant for blowing saves, compiling only a 76% save rate in 2006-2008. While that number belies Street's value, as we saw at the trade deadline in 2008, it prevents him from being over valued in the same way that Billy Taylor, or even Francisco Rodriguez was.
Street has also had a history of injury. He missed more than a third of the 2007 season with an irritated ulner nerve a condition that can reoccur and often requires surgery. He had spent time on the DL for a groin injury in 2006 before pitching through a groin issue in 2008. Additionally, Street had missed more than 10 games both with hamstring and with chest problems. This is all to say that Street was not the picture of health, and the biggest indicator of health problems is previous health issues.
Street definitely had value. His FIP has never been less than .9 below league average. He has never Ked less than 8 batters per 9, nor, until 2008, had he walked 3 or more batters per 9. His ERA has never been hire than 2008's 3.73 and he was under team control for two more seasons at the time of the trade.
So while Street had trade value, he wasn't exactly a dramatically overvalued asset, and his stock looked likely to decline with increasing cost from arbitration, losing his closer role, and declining peripherals during the 2008 season. Furthermore, holding on to Street if was to decline further would be disastrous. Usually the going price for closers is about two B prospects (from looking at trades like the Sherril deal earlier this year), but the price even for good or better than average relievers is much much lower. If one believed that Street would continue to deteriorate, trading him while the market perceived him as a closer would be wise.
Usually during the offseason there are numerous teams searching for established closers and a dearth of reliable options on the market. That was not true in 2008. The offseason saw two premier closers, Francisco Rodriguez and Brian Fuentez, as well as two secondary closers, Kerry Wood and Trevor Hoffman, be signed in free agency as well as three closers, Street, Gregg, and Putz, get traded. Nearly a fourth of the teams in baseball got one of these available closers. Therefore, by completing the Holliday trade before the glut of closers hit the market was a very astute move by Beane, similar to allowing Haren to set the market for Santana rather than waiting too long to consumate a deal.
There were reasons to believe that Gonzalez, Street, and Smith would underperform but that didn't mean they didn't have value. At Beyond the Boxscore, Jabberwocky did an excellent job of breaking down the value of the first Holliday trade using VIctor Wang's research on prospect evaluations and Sky Kalkman's Trade Value Calculator. Jabberwocky used a low estimate for Gonzalez's ability to discount for bust potential, and I think accurately pegged Street and Smith's value. All told the package the A's sent to Colorado had an estimated expected value 43.2 million.
Striking first in unloading Street to set the market, Beane also had to wade into the market for outfielders prematurely, before its collapse. The problem with striking early to set the market is the risk that accompanies a large number of variables. The economic down turn's affects on baseball's landscape were not as well known when Holliday was acquired as they were when prime free agents were signing deals at values severely discounted from past seasons. Holliday's salary which was considered very reasonable at $13.5m before the economic collapse became much much more exorbitant, as even big spending teams like the Yankees were unable to add payroll at the deadline this year. The changing landscape hindered Holliday's value as a trade chip but also undermined his immediate value to the A's as players like Adam Dunn, Bobby Abreu, Pat Burrell, and Juan Rivera all signed significantly below value deals, where Holliday represented an opportunity cost to their acquisition.
Beane placed the assets that he felt were depreciating in a deal for Matt Holliday under the premise that if the team didn't compete, he could trade Holliday at the deadline, and if the A's did compete and Holliday was kept, he would reap two type A free agents. Unfortunately, as a free agent to be Holliday's level of depreciation was known to be steep. Usually a market for elite players at the trade deadline exists, so if Beane was to acquire talent it make sense for him to go big or go home if you will. However, there was certainly a risk that Holliday moving from a weaker league and a notorious hitters park would perform below expectations an significantly hurt his trade value. That being said, the flipside is also true that Holliday could hit very well in Oakland and cement his status as a truly elite player. Theoretically, it shouldn't be possible for Holliday to hit better in Oakland since his talent level is constant and his environment became much more detrimental to putting up nominally good numbers. Therefore, Beane placed a bet on a free agent to be who should be experiencing deflating market value based on the perception of nominal production, which I view as a significant mistake and almost the unraveling of Beane's plan.
With the perceived struggles Matt Holliday faced in Oakland, his market (though not necessarily his real value) plummeted. The economic down turn that was only beginning to become apparent when Beane traded for Holliday, helped further the market shift toward valuing prospects and players under control, further deteriorating Holliday's value, even though he was arguably the best bat available on the trade market. Therefore picking Holliday as the vehicle for holdings was not a wise choice. There was the potential to acquire talent that was under control for longer and would have elped the A's beyond what was always suspect contention this year. Also there were other players whose perception wouldn't suffer moving to the Coliseum as much, such as Javier Vasquez (the solid #2 that Nico wanted), Nick Swisher from the pale hose or Josh Willingham and Scott Olsen could have been had for much less, or Adrian Beltre would probably have been available for the right package, all of which before the season I would think of as safer bets to hold their value for an arbitrage type deal.
Therefore, evaluation of the Holliday deal was a mixed bag, which is essentially what I said at the time:
Cargon proved himself as an excellent defensive center fielder. However his lack of pate discipline was concerning. I never saw the high average that Taj was right to contend was essential to having an OBP that was high enough to support his poor plate discipline. My view of the trade is simply that Holliday’s value is going to depreciate slower that a flukey fringe starter like smith, an injury and blown save prone closer with a increasing salary, and a centerfielder with a high ceiling and high bust potential.
I didn't like the Holliday vehicle but I thought the cost was reasonable and was impressed at the time that we only gave up what we did, which minimized what I thought was a poor decision to try to contend.
So while not being happy with the process initially, lets look at the results. Predictably, Holliday struggled upon moving to the cavernous Coliseum which hurt his perceived value. He did go on a rampage toward the end of his tenure. During the season the further deterioration of the economy coupled with less fans in the seats and higher debt loads for the two of the highest payroll teams, continued to push the market toward overvaluing younger cost controlled assets compared to past years and even when Beane mad the first Holliday trade. There weren't many suitors as teams that looking at the map last year like the Mets, Atlanta, and San Francisco who should have been involved were out of the equation early or never even called on Holliday. Teams like theReds and Royals didn't stay hot long enough to delude themselves that a one player was enough. The Tigers wanted in but their farm system is about as healthy as the ones outside of Chernobyl.
The Cardinals were the only team that came calling with legitimate interest and legitimate prospects. The interesting thing is that the Cardinals even acknowledged that they were giving up more value than they were receiving. According to Peter Gammons:
In the Matt Holliday case, manager Tony La Russa was able to convince Cardinals ownership that, in the words of Tom Petty, "it's the wrong thing to do/but I don't care"; tomorrow may never come.
Now here comes the difficult thing. Do we reward Beane for having the foresight to know that one team would be willing to give a bounty despite knowingly accepting less value or do we acknowledge that Beane was a inch away of losing his gambit? I choose the latter.
Most pundits declared Beane's second Holliday trade a coup. Like most trades the team that wanted to reap future gains at the expense of present ones came out with more expected value. Erik Manning over at Fangraphs broke down the trade, saying that the A's traded about $8m in expected value for $25 million dollars in expected value of Wallace alone. That fails to include Mortenson, or Peterson, which according to Victor Wang's research (summarized in pretty tables at BtB by Sky Kalkman) by are worth an additional $2.2m. Therefore, in this part of the arbitrage, as expected Beane comes a way with a decisive win, despite essentially only one team being involved in the bidding.
Some have gone as far as to suggest that the players received in the second Holliday deal are more valuable than the players traded to acquire them. I think they are wrong and using Wang's research and projections of WAR evaluations its hardly a close call. The A's gave up $43.2 million in expected value for Holliday who was worth $3.4 million dollars in surplus value to the team ($12.4 million in value - $7.5 million salary - $1.5 million cash given to the Cardinals) who was exchanged for $27.2 million in value from the Cardinals. Under these projections there is a clear $15 million loss to the A's.
There are ways to improve that number from the prospective of the A's. Street's value to the A's as a relief pitcher is less than the WAR would suggest, as the A's have a significant number of above replacement value relievers. However, this does not take into account Street's trade value, which regardless of his playing value to the A's, would remain. The George Sherril trade is a good benchmark with the Oriels receiving a B grade position prospect and a B grade pitching prospect, which carries a $12.8 million dollar expected value, which exceeds the projected value that was given to Street from his performance. Gonzalez's value is likewise minimized to the A's if the A's have an internal replacement that exceeds the baseline for WAR. Ryan Sweeney presents a strong likelihood of being a at least 1.5 WAR player which would minimize the loss of Gonzalez even if Gonzalez becomes a star. The internal replacements of the players given away are far better than the internal options at third base after the Hannahan trade. The question is is that difference enough to justify a $15 million dollar (or about 3.5 marginal wins).
I would argue that it isn't. Near universal scouting reports suggest that Wallace will not be able to man the hot corner long term. As such, I don't think that he is worth the additional value. If Adrian Cardenas was rushed to play 3b in the big leagues next year, I would project that he would be able to minimize the loss versus baseline replacement to keep from justifying a $15 million dollar loss for positional talent redistribution. Lets say Wallace has a .354 wOBA next year and is -10 FRAA over 600 PA. That would make him a 2.55 WAR player and an above average starter. Now lets look at Cardenas. If one projects him as a .312 wOBA and league average defense hes a 1.36 WAR player, and over two seasons an improvement from Cardenas to Wallace only represents 2.38 marginal wins or $10.7 million.
Overall, I think Beane had the right idea trading the three players he did for proven talent. Huston Street has bounced back significantly, but expecting him to do that was far from a sure thing. Carlos Gonzalez's patience and his value are coming around strong, though I don't ever expect him to be a star. Greg Smith got hurt and has pitched poorly, which was entirely predictable. However, I disagree with his the target for the arbitrage, as there were players that would have a better chance at maintaining their perceptual value better than Holliday, such as Javier Vasquez or Adrian Beltre. Finally the reward was not nearly good enough to justify what I believe was a poor process by exhibiting good ending results. I chalk this up as a creative but ultimately failed gambit by Beane, though the end result (a loss of about $4 million dollars) was not in any way debilitating.