I’ve moved!

Hello all! I’m happy to announce that the website has switched to a dedicated domain. You can find all of the baseball analysis you’re used to seeing here at tenthinningbaseball.com

The new domain will feature Newswire links, game reports, and occasional contributions from other writers.

Some of my Orioles-related content has also been and will continue to be featured at orioles-nation.com. Please check out Jordan’s site and stop by the forums.

Thanks endlessly for the support.

-Dan Sanchez

MLB, MiLB, and Misaligned Incentives

Major League and Minor League Baseball have agreed to extend their Professional Baseball Agreement through 2020, per a press release from MLB today.

According to sources cited by Baseball America, “…the only significant change in the agreement is the tax rate on tickets that minor league teams pay major league clubs.” The tax rate adjustment is intended to reflect the status of the economy and save MiLB an estimated 2 million dollars over the next few seasons.

The adjustment is a good thing on its own, as MiLB could clearly use a boost in revenue. But, as Rob Neyer points out, so long as MLB and MiLB continue to operate as financially separate entities, yet purport to function with the same goal in mind, MiLB will be conducting business with misaligned incentives – the type of misaligned incentives that fail to provide the bottom tier of minor league players with acceptable living wages.

The collective bargaining agreements between the MLBPA and MLB are set up to protect (surprise) Major League players – in particular, the top tier of major league players. MiLB continues to function with the primary goal of player development, but operates as a more or less independent business, affording its players very little protection.

Either MiLB and MLB must function with their own independent, autonomous, players’ unions, or the two must merge with the agreement to treat all of their employees equally and fairly. Until MiLB adopts incentives consistent with its operating procedures, the overlooked masses who make the PBA’s agreement to “field at least 160 minor league affiliates” possible will be offered insulting compensation and very little security.

Featured Newswire Link 3/7/11: BtBS’ Darowski on Peak-Weighted WAR

The newly created Tenth Inning Newswire will provide a collection of links to outstanding articles, blog posts, and other write-ups that are (at least loosely) relevant to baseball and/or statistical analysis. Certain links will be featured on the main blog page. Additionally, a collection of the week’s links will be posted every Sunday and stored in a database on the site.

Our first featured link comes from Beyond the Box Score’s ADarowski. Darowski has been a proponent of weighted WAR systems for some time. On January 19th, he defined a statistic he called “weighted WAR,” or wWAR. This is a metric that attempts to quantify a player’s career value by giving ‘extra credit’ to peak seasons. It essentially combines the player’s career WAR with additional wins for MVP and all-star caliber seasons during peak years.

Not surprisingly, the metric can be applied logically to Hall of Fame debates. Since the criteria which HoF voters are asked to use give consideration to both peak performance and longevity of career, wWAR is a quick and dirty way to discuss a player’s HoF merits. The metric is simple and subjective, and the details are certainly open for debate, but I thought it a nice preliminary way to get the ball rolling on peak vs. longevity discussions. Darowski recently began attempting to compile a list of HoF-worthy players by position using wWAR as a guide. He’s begun with the catchers, but will presumably complete his list soon.

Check it out, and join in the discussion:

http://www.beyondtheboxscore.com/2011/3/7/2022771/hall-of-wwar

Why the Baltimore Orioles Should Treat Adam Jones Like a Stock Option

In Finance, options are used to mitigate risk and secure the possibility to capitalize on future inefficiencies in the market. The basic “call” option, in a nutshell, works in the following way: the buyer pays a relatively small premium to the owner of the stock for a clearly defined period of time in order to purchase not the stock itself, but the option to buy the stock further down the line for a set reference price that is agreed to by both parties.

Options create more opportunity for gains because, since current market valuations are likely to be more accurate and ubiquitous than hypothetical futures, options give the buyer the chance to cash in should market shifts cause the underlying asset to be valued higher than the reference (or strike) price. When stocks are believed to be undervalued, buying an option on the stock, although pricier if exercised than the actual current value of the stock, protects the buyer against the possibility that the stock never becomes valued higher than the strike price. Generally, the premium is low enough for the buyer to easily absorb should the asset not rise sufficiently. This methodology is particularly lucrative if the stock in question is subject to volatility, such as (1) stocks that have dropped temporarily due to market swings or PR  scandals or (2) stocks that are either start-ups or valued prior to the cusp of some boom in the market.

We see options for the first type of volatile assets utilized successfully in MLB contracts quite frequently. This is the type of contract doled out to free agents coming off of “down” or injury-plagued seasons. Usually, a team will agree to a short term, low base contract with the player while including a second year “option.” The option, if exercised, generally costs more than the short-term contract. But, ideally, the price would still be lower than the market value of the player should he rebound in health or productivity. If the player breaks out under team control, the team reserves the right to exercise the option at a relative discount to the market. If the player does not recover, the team may decline the option (usually for a small buyout). This is an excellent low-risk/high reward strategy.

Options for the second type of volatile players are much more rare. Occasionally, when a player extends with a team or signs a long-term free agent contract, the team will slap an option year or two at the end of the deal. This is all well and good, but it doesn’t really protect the organization against the risk of attrition, bust, or even stagnation, because there are still quite a few years of guaranteed money built into the contract.

The perfect candidates for pre-boom options are players in their initial cost-controlled years who possess high ceilings but have not yet broken out. These players, like all players still dreaming of the big pay-day once their indentured servitude in the closed MLB market expires, seek stability and lucrative pay. Teams hope to lock their future stars down before they get expensive. The only roadblock to pleasing both parties is risk. The organization asks, “what if we commit millions to young Danny Baseball and he never turns into a star?” Danny Baseball asks, “what if I sign too early, for too little, and become a star?”

I believe that stock options, call options in the truest sense, provide the solution to this quandary.

Let’s consider the situation of Adam Jones and the Baltimore Orioles. Adam Jones is a 5-tool talent, a former top prospect acquired from the Seattle Mariners in the epic five-for-one Erik Bedard trade in 2007. Jones has turned himself into a productive starting center fielder, although, to this point, poor plate discipline, inconsistency and some defensive questions have prevented him from taking the next step into star territory. Jones is young, athletic, and charismatic, all the things you look for when committing to a franchise face. But the uncertainty over whether he will make that next leap has prevented the sides from seriously discussing a long-term deal at this point. Baltimore, it seems, would like to wait until they consider Jones more of a sure thing.

I consider this an inefficient strategy.

With every ticking season, Jones creeps closer to free agency, and his incentive to sign an extension rather than wait to test the market fades. Should Jones have a breakout season in 2011, his contract demands will rightfully sky-rocket. Understandably, Baltimore’s front-office is averse to the risk of guaranteeing Jones tens of millions of dollars only to see him stagnate, regress, or get bogged down with injuries.

The solution? Stock options, of course!

Let’s engage in a hypothetical. Imagine that the Orioles’ camp has assessed Jones’ value-at-risk and decided that they are comfortable offering Jones a 5 year deal, buying out Jones’ three remaining arbitration years and two free agent seasons, structured in the following way:

2011: $3 MM

2012: $6 MM

2013: $9 MM

2014 (FA): $10 MM

2015 (FA): $10 MM

Remember, the first three seasons are arbitrated and cost-controlled, so their value is suppressed. Almost all extensions are structured in this way. This deal would amount to a total of 5 years, $38 MM. If Jones stagnated as a ~2 WAR player, those two FA years will have been purchased right at market value.

Jones, however, doesn’t have much incentive to take this deal. A solid or better season in 2011 would cause a boom in his expectations. Let’s say, for the purposes of this hypothetical, that Jones’ camp countered with the following demands:

2011: $3 MM

2012: $6 MM

2013: $9 MM

2014: $12 MM

2015: $13 MM

2016: $15 MM

For a total value of 6 years, $58 MM. Jones’ version gives him an added year of security and 20 million dollars more in guaranteed money. Assuming that this discrepancy is too large to bridge, I propose this counter offer from Baltimore’s camp:

2011: $4 MM

2012: $7 MM

2013: $10 MM

With an option to buy at the following reference price:

2014: $11 MM

2015: $12 MM

2016: $14 MM

Take a look at what that deal does for both sides. If Baltimore exercises the option at the strike price, Jones gets his six year, $58 MM deal to the dollar. It’s simply structured a bit differently. Should the Orioles decline to purchase the option, Jones will have received $3 MM (plus a small buy-out reward) more during his arbitration years than he likely would have otherwise, and he gets to hit free agency like he wanted to all along. Should Jones break out, Baltimore would likely exercise the option and get three more years of production at below-market rates. Even if Jones fails to develop, the Orioles will have simply wasted one million dollars per season for three seasons. Yes, one million dollars is a lot of money, but in the realm of MLB it is hardly cost-prohibitive or hampering in any real way. It’s the price of acquiring Cesar Izturis to sit on the bench.

And now for my caveat: I am in no way privy to front office discussions or the asking and offering prices of either camp. These numbers are entirely made up, though they seem fair by my valuation. In any case, the details don’t matter so much. What is important here is the methodology. Players earn so little in their cost-controlled seasons that even a premium that is relatively insignificant in terms of an ML organization’s payroll can represent a 25% or more increase in pay. If, as some have theorized, Scott Boras and other agents might be successful at convincing their clients to hold out for free agency by offering them monetized insurance, this is the organization’s rebuttal. I imagine it would be very difficult to turn down a contract that offered both additional money up front and the opportunity for a lucrative long-term deal as well. From an organization’s standpoint, the contract might have some small incentive effect to push the player to stay productive before the option expires.

There are plenty of other ways to structure option-based extensions, like paying a higher yearly premium for the right to go year to year with increasing options. However they be utilized, contract options provide an avenue for small- and mid-market teams to exploit inefficiencies in the market in order to retain their talent. The more limited the resources, the more creativity is necessary in the negotiations. Smaller market teams just can’t afford to have big money contracts go bad. It’s not that these teams should be averse to the risk of locking up talent, they should just be finding creative ways to mitigate that risk.

On Ability Grouping and Educational Success

The United States, we are told, is a meritocracy. A man’s successes or failures are his deserts, and his deserts alone. The idea that our most successful compatriots are those who possess the most finely tuned inherent skills is an essential pillar of the American Dream. When someone succeeds in this country, it’s because he or she deserves it.

This assumption, of course, is far from the truth, but often for reasons more subtle than we might imagine. That is precisely the focus of Malcolm Gladwell’s best-seller, Outliers: The Story of Success. The book, naturally, is marketed as a sort of thinking man’s guide to success, as was his wildly popular, The Tipping Point. Both books, however, have a far more interesting, and far more sincere, purpose at heart. In Outliers, Gladwell makes a well-honed and compelling attack on the way we conceive of success and the ‘self-made man.’

One of the most intriguing ideas of the book is introduced in its first chapter. Gladwell recounts the story of Roger Barnsley, a Canadian psychologist and avid hockey fan, who noticed one day that the vast majority of the members of a particular hockey roster were born in the first three months of the year. Puzzled, he decided to look at another roster, and found almost identical results. Now consumed with what he thought must have been a strange anomaly, Barnsley began to scour the rosters of every hockey team he could find, from the most amateur to the highest profile professional leagues. Almost invariably, Barnsley found the result the same – over 50% of any given team’s roster was likely to have been born in the first three months of the year, with the results often skewing more strongly. Indeed, many teams had nearly a complete lack of representation from the population born in the last quarter of any birth year.

Of course, as Gladwell explains, this has nothing to do with the inherent qualities of being born under one zodiac sign versus another. The explanation was quite simple. From the earliest pee wee leagues, Canadian hockey leagues use an arbitrary cutoff date – usually January 1. A player born on December 31st would have to play up an age bracket, while a player born on January 2nd was able to play down a bracket. It was entirely conceivable, and it happened frequently, that one player born in early January could play against competition born almost 12 months later. At such a young age, nearly a full year of development is a huge deal. When selecting players, coaches usually preferred the biggest, strongest, most developed players – the players who hit the hardest, skated the fastest and seemed to stand out among their peers. And these were exactly the type of players who were selected for all-star and travel teams, the type of players who received the best instruction and encouragement, and the type of players who were pushed to make the leap to professional hockey.

Some coaches were confusing ability with physical maturity, and others didn’t care much; they just wanted to win that season, and to win they needed to pick the most advanced players. A simple, and arbitrary, selection process at a young age had very real consequences for these young players. The oldest ones of their class progressively accrued more and more advantages each year, while the youngest were shut out from opportunity from the start.

As Barnsley explains:

“These kinds of skewed age distributions exist whenever three things happen: selection, streaming, and differentiated experience. If you make a decision about who is good and who is not good at an early age; if you separate the “talented” from the “untalented”; and if you provide the “talented” with superior experience, then you’re going to end up giving a huge advantage to that small group of people born closest to the cutoff date.”

This three-pronged approach to ability grouping happens in a much more critical area of life as well, namely the public education system. Consider the parallels: our education system provides arbitrary cutoff dates that separate students into age classes, though two students in the same age class could still have birthdays nearly 12 months apart. In education, we use standardized testing and other measures to “select” the most talented and intelligent students. We then use these selection results to sort the students into different streams (i.e. “gifted and talented,” standard, and a learning disabled tracks). Finally, we allocate our resources such that those who have been selected for the most prestigious streams are provided the best learning environments, raw materials, and instruction. When a student is selected for the gifted track, he or she subsequently receives better instruction the following year, increasing the gap between the student and his or her ‘contemporaries.’ The next year, the same thing happens, and so on. What we see is not just one initial advantage, but an accruing of advantage that continues throughout life.

This process of ability grouping and accruing advantage would be problematic enough if we were actually selecting our most gifted and talented students appropriately. But, because of our use of age cutoffs, the selection process is largely arbitrary. Gladwell discusses the research economists of Kelly Bedard and Elizabeth Dhuey, who found that the difference in scoring on a standardized math test between two otherwise equivalent fourth graders, one born toward the beginning of a calendar year and one born toward the end, could be as much as 12 percentile points. If both students were around the cutoff for above average scores, this variance would be more than enough to qualify one student but exclude the other from selection for a gifted track. Our educators, like Canadian hockey coaches, are confusing ability with maturity. Just as is the case with physical development, 12 months can mean a world of difference in mental development for children. To exclude a student from accruing advantages throughout life simply because he or she was born later in the calendar year is arbitrary and unethical.

There is a range of potential solutions to the problem – we could, for instance, refuse to participate in ability grouping until the teenage years  as they do in Denmark, or create quarterly age cutoffs rather than yearly age cutoffs in public schooling – but more important than fixing this individual issue is addressing the way we look at success and failure in all walks of life. In education and elsewhere, we need to understand that success is not just a function of merit; that unfair selection processes expose us to or exclude us from accruing advantage quite frequently; and, perhaps most difficult of all to fathom, a person cannot be wholly credited for his successes or debited for his failures unless we adequately neutralize for context and isolate for the things he or she can truly control.

Indeed, making an effort to provide accruing advantages to those whose intelligence is initially undervalued, rather than overvalued, might be more crucial to our growth as a nation.

A Few Words on Christina-Taylor Green

“Forces beyond your control can take away everything you possess except one thing, your freedom to choose how you will respond to the situation.”
— Viktor Emil Frankl

 

On Sunday, January 9, one day after an apparently politically-motivated shooting in Tucson, Arizona, left 19 wounded and at least 6 dead, the spotlight of stories in the media began to shift. In the initial attempt to make sense of the catastrophe, most of the focus was on the shooting of Representative Gabrielle Giffords (Az.) and District Court Chief Judge John M. Roll. The synopses often included a small side note, usually reading, ‘among the victims was a 9 year-old girl.’

Yesterday, the world demanded to know who that little girl was. As the remarkable details of Christina-Taylor Green’s life were revealed, the event became more difficult to fathom. Ms. Green – a title too formal to describe the average 9-year-old but somehow appropriate when discussing one with such lofty ambitions – was born on September 11, 2001. She was one of 50 babies Featured in “Faces of Hope,” a collection of photos and words dedicated to children who were born in the midst of the 9/11 attacks.

Perhaps because of, or perhaps in spite of, the context of her birth, Ms. Green developed into a child beyond her years. Even at her young age, she was very interested in politics. She had made vocal her intentions to attend Penn State, pursue a career in politics, and use her means to help those less fortunate.

It was only natural that, when her representative held a meet-and-greet outside a Safeway in Tucson on Saturday, January 8, Christina-Taylor chose to attend. The events that transpired have been well documented.

Now, the world will never see Representative Green, the woman born on September 11th, 2001, who graduated from Penn State and became a civil servant. Or Ms. Green, the Olympic Swimmer. Or Ms. Green, the ballet dancer. Or whatever iteration of her life she may have chosen to pursue over the next century. Christina-Taylor Green is dead, and we must come to terms with that fact.

*****

As I read about the birth, life, and death of Ms. Green yesterday morning, alone in a room of a house not my own, I began to cry. My girlfriend walked in to find me in a pathetic state and asked me why I did this to myself. Why do people insist on reading about all of the bad things that happen in the world? What benefit does it provide?

I didn’t have a good answer at the time. I’m still not sure if I do.

Like her, I am disgusted by the media’s attempts to sensationalize tragedy. I cringe at the exploitation inherent in painstakingly interviewing families of victims to squeeze out a good story. I often find the whole process trite at best, objectifying at worst. I could spend more than a few days writing about the sickness of catastrophe-porn.

But, at my core, I believe it is important to be aware of and to understand the events that occur around us – good and bad. I believe that the narrative of human life at large provides us with the means for creating meaning in our own lives.

Moreover, I believe it is the role of the media to provide, contextualize and analyze that narrative. At times that means extrapolating symbolic importance from every day events. At times that means finding meaning and hope in the midst of events which, prima facie, may seem to provide only meaninglessness and hopelessness.

Is this narrative sometimes a fabrication, if not an outright lie? Probably. But it is precisely this fabrication that allows us to create meaning. The choice to do so is no more a fabrication than the choice to wallow in nihilism or to ignore the power of the human narrative.

Christina-Taylor Green is one of those rarest of outliers – one who, despite the tragic brevity of her time on this earth, was charged with the task of significantly altering the course of our narrative not once, but twice. Hers is the first death among the 50 children who were profiled nine years ago. In time, every last face of hope will be buried. In still more time, all of those who knew them will be buried as well.

Nine years ago, Ms. Green’s birth played a small role in helping a nation make sense of an unfathomable event. Making sense of her death will be a far more complex and arduous task. But there must be meaning there. We must find it.

Projecting Mark Reynolds

Image Courtesy of AP

On Monday, December 6th, the Baltimore Orioles completed a trade that sent right-handed relievers David Hernandez and Kam Mickolio to the Arizona Diamondbacks in exchange for slugging third baseman Mark Reynolds.

Reynolds, who will be 27 at the start of the 2011 season, brings some much-needed right-handed power to the lefty-heavy Orioles lineup. Despite setting single-season records in strikeouts in each of the last three years, Reynolds has averaged 35 home runs a season over the same span.

Much ink has been spilled discussing the merits of the trade, which seems to be a win-win considering the needs and philosophies of both organizations, but I’d like to focus this analysis on the somewhat foggy task of projecting Reynolds’ 2011-2013 value.

Reynolds is under contract for two years and an option with the following breakdown:

2011: 5 MM guaranteed

2012: 7.5 MM guaranteed

2013: 11 MM club option (500 K buyout)

In essence, Reynolds is guaranteed 13 MM over the next two years (buyout included) with a club option that could bring the deal to 3/24. Even with the option exercised, this is a good value contract for an average or better starting positional player in his athletic peak.

But, just how good does Reynolds project to be?

*****

Players of Reynolds’ profile tend to be quite polarizing because they tend to do two very obvious things (strikeout and hit home runs) and one not-so-obvious thing (walk) quite often. These hitters are often maligned for their tendency to whiff, worshipped for their ability to hit the long-ball, and, well, somewhat ignored for their ability to walk. With strikeouts, walks and homers comprising 51.5% of his plate appearances over the last three years, Reynolds is the quintessential Three True Outcomes hitter.

We call these outcomes “true” because they have almost no dependence on external factors (e.g. defense, luck, etc.). When the batter hits a ball that leaves the ballpark in fair territory, it will almost always result in (at least) a run. When the batter receives four balls before either receiving three strikes or putting the ball into play in fair territory, it almost always results in a runner safely reaching first base. When a batter sees or swings at strike three before seeing ball four or putting the ball in play, it almost always results in an out.

Any ball that is put into fair play, however, is not a true outcome. These balls can be caught, be fielded on the ground and thrown to first to retire the runner, or fall safely for a single, double, triple or inside the park home run. But the end result of these outcomes is variant and dependent on multiple external factors. (Yes, it’s true that a ball hit 400 feet to left center may leave certain ballparks and remain in others, but for this reason it’s correct to say that a fly ball is not a true outcome, while a home run is.)

Not all TTO hitters are created equal. The ratio of each outcome to the others will significantly alter the value of the hitter’s production. When the spread of outcomes tips toward strikeouts, a hitter will see his batting average (and subsequently his OBP and SLG) dip. When the hitter increases his BB/K ratio, he will start to reach base enough to salvage a respectable OBP despite a paltry batting average. When a hitter increases his HR rate, he might be able to hit for enough power to mitigate the costs of sub-par on base skills.

The best TTO hitters (see: Bonds, Barry) were wildly successful because they achieved the right balance of outcomes. While Bonds and Reynolds have both been just about as likely to walk, strikeout or hit a homer as to put a ball fairly into play, Bonds’ and Reynolds’ K/BB ratio are almost reciprocals of each other.

Still, these types of hitters don’t have to be Bondsian freaks of nature to be valuable. In fact, TTO hitters are, generally, more likely to create above average run production for their teams. Over Reynolds’ brief four-year career, he has been able to do just that.

In 2007, 2008, and 2010, Reynolds was able tip the scale to just above average. On average, Reynolds produced about 4.5 runs (i.e. about half a win) more than the average third baseman with his bat. Yes, he struck out close to 200 times a season over the same span, but he produced enough of the two good outcomes to outweigh the negative effects of the bad.

The one season I skipped over, Reynolds’ 2009 campaign, was a career year, a year in which he saw that perfect alchemical combination of outcomes that transformed him into a borderline elite player – he posted a .260/.349/.543, good for 25 runs (about 2.5 wins) more than the average third baseman.

Now a legitimate 40+ HR threat, it seemed Reynolds had emerged as one of the games young stars. But, one year, 211 strikeouts, and a sub-Mendoza batting average later, Reynolds found himself shipped out from the organization that had drafted, developed and extended him in exchange for a couple of unestablished relief pitchers.

What exactly happened?

*****

Well, for one thing, Reynolds started striking out more – above even the prolific standards he had previously set for himself, actually whiffing in 42% of his plate appearances.

Additionally, the balls he did put into play were increasingly being hit into the air rather than on a line. While this will not have a damaging effect on a hitter’s home run rate, it will certainly affect his batting average. Balls hit in the air tend to fall for hits less frequently than balls hit on the ground, and much less frequently than balls hit on a line. Accordingly, his batting average on balls in play fell from the .340ish area, in which had been previously living, down to .257 in 2010. This, combined with his now astronomical strikeout rate, led to an (admittedly) embarrassing .198 batting average. It doesn’t matter if you’re Mark Reynolds, Carlos Pena, Adam Dunn or even Barry Bonds: having a batting average that low will have a significantly taxing effect on your OBP% and SLG%.

The good news is that, while Reynolds had been striking out more and more each year, he had also started to walk more and more each year. His 13.9 BB% in 2010 was a career high, indicating that the bump in K% might have had less to do with new holes in his swing being exposed and more to do with an increased willingness to take the count deep before swinging at a pitch he liked. That’s right, his increase in strikeouts were actually a product of increased, rather than decreased, discipline. This kind of bump is actually quite common, if usually less exaggerated, in young sluggers who are approaching their peak years.

Additionally, while Reynolds’ .234 isolated power was down some from his remarkable 2009 campaign, it was still higher than he had posted in any other year. So, Reynolds was walking more and hitting for just as much power as usual – the only thing dragging his performance down was his exceptionally low batting average.

Reynolds’ career low .257 BABIP was certainly fueled by his decreased LD rate and increased FB rate. However, statistical variance (i.e. luck) probably played its part as well. Given Reynolds profile, the shift in batted ball tendencies should have, indeed, caused a drop in BABIP, but only down to about .290. Adjust his production for that neutralized BABIP and you’re looking at a .231/.353/.466 slash line, which, considering the league drop in offense in 2010, would have easily been his second most valuable offensive season.

Take into consideration also something that I haven’t seen mentioned much – a freak hand injury that caused a brief cliff-dive at the end of Reynolds’ season. Toward the end of August, Reynolds injured his thumb. He rushed back from the injury only to post a .078/.213/.078 line in September. Over that span, Reynolds struck out in 31 of his 64 plate appearances and had no extra base hits. Clearly, he wasn’t right. The tailspin dropped his season OPS by nearly .050 points and likely robbed the slugger of the opportunity to reach 40 home runs for the second straight season.

I know – what if, what if, what if…

It may not be wise or fair to speculate that, if Reynolds’ line had been neutralized for luck and he hadn’t injured his thumb late in the year, he may have posted another .850+ OPS, 40+ HR season. Anecdotal and statistical observation suggests that Reynolds was getting a little too homer happy, sacrificing consistent, hard contact for extra loft in his swing. He still has holes that he needs to patch up, and there comes a point where a plate approach may be too disciplined. But, isn’t it prudent to consider all the mitigating circumstances surrounding his down season?

Setting anecdotes aside and just looking at comparables, Bill James projects a .233/.337/.490 line for Reynolds in 2011. And, just looking at the numbers, why wouldn’t one expect at least that for his mid-case projection? Sure, Reynolds is coming off a down year, but he will be just 27 next year – in his athletic peak – and, prior to 2009, had been getting better and better. Once one factors in the residual effects of his thumb injury last season, I think it’s safe to bet on a healthy Reynolds rebounding to be at least an .800 OPS bat, warts and all, with considerable upside.

Even if Reynolds stagnates around an .800-.825 OPS, he’ll still likely be a 2-3 win player. Statistically and anecdotally, Reynolds has made vast improvements defensively. He may only be a -5 defender at this point, but that’s still good enough to keep him around 2.5 wins above replacement.

Valuing wins at about 4.5 MM on the free market, you’d have to pay 30-35 MM over three years for that production via free agency. At 2/13 or even 3/24, Reynolds should provide solid value. Factor in significant upside, and Reynolds appears to be a bargain.

There’s still signficant risk, as Reynolds walks a fine line in regard to contact rate. Any more regression in his ability to make contact and he could see a free-fall. However, considering his age and the mitigating circumstances, I’d bet on above average production from Reynolds going forward.

Nate Silver Still Delivering: House Effects and Bias in Forecasting

Nate Silver, whose fantastic fivethirtyeight blog for the New York Times has been wildly successful, is still offering up plenty of analysis relevant to the sabermetric community despite his departure for politics. In his newest post, When ‘House Effects’ Become ‘Bias,’ Silver distinguishes between several statistical terms used to describe variance and error in projection systems.

The term ‘House effects,’ as Silver points out, refers to systems or portions of systems that produce outlying projections relative to their peers. In the weeks leading to the 2010 mid-term elections, for example, a Gallup poll showed a 15 point advantage for Republicans in house races, while most other polls predicted a much more modest victory.

After the conclusion of whatever event or sampling a projection system purports to predict, we can assess the system’s projections relative the actual outcomes. This can be done by looking at one or both of the following: margin of error – the total or average percentage points by which a projection system ‘missed’ the outcomes – and what is called statistical bias – the leaning of projections too heavily in one direction or another.

Election polls provide a nice, varied set of data in which bias can be seen clearly. For example, one poll-based projection might end up having a large margin of error (missing frequently on predicted outcomes), but not show any particular preference to Democratic or Republican candidates. Another poll might predict more correct outcomes, and generally miss by a smaller percentage of points, but show results that frequently overestimate the votes for Republican candidates. This is what political scientists mean by the term ‘bias.’ It is important to note that this form of statistical bias is not the same as partisan bias, or the skewing of projections based on the political views held by the conducting group, regardless of what overlap between the two there may be.

These terms are equally important when assessing the accuracy of projection systems in baseball. To give one example, Adam Jones received the following preseason OPS projections for 2010:

Bill James: .804
CHONE: .846
Marcel: .773
ZiPS: .818
Fangraphs Fans: .804

Taking a look at the consensus, we could have concluded in March that Marcel’s predictions carried somewhat notable house effects, with its OPS projection a relatively low outlier among its peers. Here we start to see that house effects are not always a bad thing, as Jones actually posted a .767 OPS in 2010. Sometimes one projection system might be keen to certain predictive factors that largely go unnoticed by others. Anyone who has studied statistical projection has undoubtedly come to the conclusion that even the most accurate MLB projection systems carry a high margin of error, and sometimes the consensus projection can miss significantly.

Much of this, however, can be explained by the idea of statistical variance (the statistician’s euphemism for luck). OPS is a combination of various inputs of outcomes, and often fails to account for how a player truly performed. If we look at the outcomes a hitter has the most control over, we can see that there wasn’t as much discrepancy between the systems as we might have thought. Each of the four simulation based systems (I’ll take the fans’ prediction out of the equation for now), for example, suggested that Jones would draw a walk in 6.4 to 7.0 percent of his plate appearances and strike out between 20.1 and 20.9 percent of the time. Most of the incongruity between systems was about batted ball data – which, all controversy aside, is much more susceptible to chance and variance than things like strikeout and walk rates. So, it seems that the main point of divergence among these projection systems was in attempting to predict chance, a task sisyphean at best.

While margin of error is one way of analyzing the utility of projection systems, it might be more helpful to look at bias. We can do this by isolating one bivalent variable and assessing the accuracy of the system in either direction. For example, if we believe that CHONE was overly generous in its assessment of Jones because it fails to account for the learning curve among young outfielders, we could assess its margin of error among outfielders under the age of 25 over the last 3 years. If patterns start to occur after running regression analyses as such, we can conclude that there is a statistical bias. One system might be consistently overrating rookie pitchers in Baltimore because it fails to consider the favorable pitching conditions in the AAA International League in general and Harbor Park in particular. Another system might be failing to accurately translate pitching performances for flyball pitchers in the AAA Pacific Coast League because the launching pads in the league make it nearly impossible to assess how well that type of pitcher would perform in a neutral context.

These are obvious pitfalls that have led to statistical bias, and they have largely been corrected in most systems, but it’s tough to say what other anomalies are being missed. What can be guaranteed is that no projection system is free from bias unless it is either perfect or unrealistically random. Digging for hidden biases is essential in attempting to find market inefficiencies in any realm.

A Look at Tim Adleman

Image courtesy of minorleaguebaseball.com

On June 8th, RHP Timothy Adleman was picked by the Baltimore Orioles in the 24th round (718th overall), making him the first Georgetown Hoya to be selected in the MLB Amateur Player Draft since Matt Bouchard in 2007. The 6’5″, 210 pound junior was Georgetown’s most relied upon starter in 2010, leading the staff in starts (14) and innings pitched (83.1).

While his statistics were middling – Adleman posted a career 5.83 ERA over 52 games (48 starts) – there were some indicators that he might transition well to pro ball. Georgetown was generally overmatched against its opponents in the Big East, which is a favorable offensive league, and, despite his mediocre ERA, Adleman posted a solid 224/83 K/BB ratio.

Over his brief rookie stint with Baltimore’s short season Aberdeen Ironbirds, Adleman has pitched to a 3.57 ERA in 63 IP across 12 starts. He has struck out 58 batters over that span while walking 24 and posting a 1.48 ground-out to air-out ratio. His early success may lend some support to the idea that college pitchers with ground-ball tendencies are generally undervalued. Since ground-balls tend to go for hits more often than fly-balls, the limited range of college defenders hurts ground-ball pitchers much more than those who typically get outs in the air. Since ground-ball hits that defenders never touch don’t count as errors, this discrepancy is not reflected in ERA. When these pitchers are surrounded by superior defenders at the pro level, they tend to see their BABIP drop significantly.

It might not be technically correct to say that collegiate ground-ball pitchers are undervalued – scouts likely have an intuitive idea that certain pitchers will excel when given adequate defense – but it is likely the case that their success is more likely to translate, or even accelerate, to the professional level than that of their altitude-friendly compatriots. Thanks to area scout Chris Gale, Baltimore may have gotten round 6th to 10th round value out of a player who signed for slot in the 24th round.

Updating Brian Matusz

The 2010 season has been a mixed bag for 23-year-old rookie Brian Matusz. The lefty out of San Diego, taken fourth overall in the 2008 amateur draft, breezed through the minors in less than a year, posting a combined 1.91 ERA over 19 starts in high-A Frederick and AA Bowie before making his Baltimore debut on August 4th. Matusz went on to make 8 big league starts, posting a 4.63 ERA, striking out nearly eight batters per nine innings and walking fewer than three. Entering the 2010 season, expectations in Baltimore were high for Matusz, who showed flashes of dominance toward the end of his first big league stint.

There were, however, subtle indicators that suggested enthusiasm be curbed. Scouts had warned that Matusz’ offspeed deliverings were so good that he might not face any real tests in the lower and middle levels of the minors. Being able to consistently fool the competition with his change-up and curve, Matusz might not learn how to set the pitches up with his fastball – a skill that would prove crucial in the big leagues.

Still, after such dominance in his minor league campaign and a brief stint in the majors that might actually have been better than it seemed (the debutant was saddled with an unfortunate .343 BABIP and racked up a 4.08 FIP, 4.23 xFIP, and 4.35 Defense Adjusted ERA), optimism was abundant despite a 50% PECOTA and ZiPS projection that saw Matusz posting an ERA in the upper 4s.

That projection is eerily similar to Matusz current production, a 4.97 ERA over 137.2 innings. Over 25 inconsistent starts, the lefty starter is striking out fewer batters (6.93 K/9) and walking more (3.40 BB/9) than he did in his short stint in 2009 (7.66 K/9 and 2.82 BB/9). What is perhaps more disconcerting is a noticeable regression in his fastball, which has gone from an average of 91.5 in 2009 to 90.0 in 2010. Matusz has also had trouble commanding the pitch.

Just as scouts had warned, when Matusz is not able to establish his fastball and use it to get ahead in the count, his secondary pitches become less effective. He’s still throwing the fastball about 60% of the time, the change-up 20% of the time, and the curve and slider about 10% of the time respectively, but he’s gone from throwing 62.2% of his pitches for strikes in 2009 to 58.4% in 2010, and hitters have gone from making contact on 78% of attempts to 83%. The lack of velocity and fastball command has clearly hurt Matusz.

Still, there are some positive signs. Matusz is getting more ground balls, fewer fly balls and fewer line drives. Despite the improvement, he has still been unlucky on batted balls, pitching to a .322 BABIP. Component ERAs paint Matusz’ season more favorably, as he has a Fielding Independent Pitching (FIP) of 4.23.

This post comes on the heels of Matusz best start of the season – eight shutout innings against a good Texas team, with six strikeouts to one walk and just five hits allowed. It’s possible that the stress of his first full Major League season has caused Matusz some fatigue. However, despite the loss in velocity and inconsistent command, the 23-year-old is still pitching like a fringy middle of the rotation starter. With a recovery of velocity and continued experience, Matusz could still take the leap into top of the rotation status as soon as next year.