A.k.a. This good move stuff takes some getting used.
A.k.a. Heilman and Olson are racking up the frequent flier miles.

Today the Mariners potentially solved their glut of Major League back-rotation starting pitchers by shipping off Aaron Heilman, acquired from the Mets earlier in the winter in the J.J. Putz blockbuster. Heilman is now on his way to the Cubs of Chicago in exchange for infielder Ronny Cedeno and Garrett Olson, who himself was just recently acquired from Baltimore for outfielder Felix Pie.

From the Mariners perspective, this could not have worked out much better for them. Heilman was going to be hard pressed to make any impact on the big league team given his position of wanting to start, but not being good enough to do so. In exchange for him, the Mariners get Ronny Cedeno whom the Cubs have soured on, but has a phenomenal minor league track record as both a hitter and a defender and Garrett Olson who boils down to being a lefty, much younger, version of Heilman, albeit with less Major League success, but one with options left that allows the Mariners to give him some more seasoning down in Triple-A while they do their best to move Carlos Silva or Jarrod Washburn out of the way.

For the Cubs, well, they managed to clear some room on their 40-man roster I suppose. Heilman is probably a better option for 2009 and can serve as a swingman on the staff, moving in and out of the rotation as need be, but that is the best that I can come up with for them. There’s no doubt that this trade is a win for the Mariners, and the best that I can give the Cubs is a pass. They have turned Felix Pie and Ronny Cedeno, two huge prospects as little as a year ago, into Aaron Heilman.

And no, that is not going to help them get Jake Peavy. Any rebuilding team would be far more interested in Cedeno and Olson than Heilman.

More signings! More more more! We’re not even going to have a free agent market in the future because every player will get locked up through their age 30 season or beyond and with the rising intelligence of GMs across the league, the bottom will fall out of the market completely!

Okay, maybe not, but it’s not the most improbable thing that I will ever utter here. Anyways, on to today’s inking, Ubaldo Jimenez and the Colorado Rockies came to an agreement on at minimum a four-year deal with a pair of club options. If those are turned down, the deal covers Jimenez’s final two club-control years plus his first two arbitration years. The club options cover his final arbitration and first free agent year.

It is a little rare, though becoming less rare, to see long term contracts covering pre-arbitration years and we do not have a good framework like we have for the arbitration rewards, so for the time being, I am just going to assign those seasons a half million value. Therefore, the minimum of this contract, two club control and the first two arbitration years, adds up to one market year plus an additional one million dollars. If the club options are exercised then we add the final arbitration year (0.8) and a full market year, bringing the total market years to 2.8, still plus the additional million bucks.

In an effort to better visually display the projections, I put them all in a chart this time around, reproduced below. As I have mentioned, I have three main methods for coming up with pitcher projections: I take CHONE and Marcel straight from here at FanGraphs, and I complement those by looking at the pitcher’s tRA* from the past year from StatCorner to see if that indicates anything divergent to the other two.

In Ubaldo’s case, Marcel and CHONE form a bracket on projection between 2.3 and 3.1 wins for 2009. On the deal without the club options that means that a fair value deal, under the assumption of $4.5 million per win, would come in between $10.3 and $13.5 million, while the Rockies are actually on the hook for just $10 million. With the club options exercised, fair market would dictate $27 to $36 million while Colorado would be forced to pay just $22.75 million.

Add another data point that teams are continuing to get great deals in locking up their young talent and avoiding arbitration and getting a few market years in well.

Alternate title: Exalt to the heavens! The Royals got one right.

The news broke early today that Zack Greinke had signed to a four-year extension, buying out a pair of arbitration and free agency years each. Unsure of when the financial details would be leaked, I went ahead and started out sketching out the projection for Greinke’s value.

The first, and hardest step, in pitcher evaluation is settling on a projection for him. As per my usual course of action, I sampled a couple different systems, notably CHONE and Marcel and then threw in some looks on his past xFIPs and tRA* values. Greinke is one of the more rather difficult cases to project because his playing time has been all over the map, but not completely due to physical injuries. How do we project his innings pitched going forward? Do we apply the same sort of emotional/mental risk as we normally do with physical injuries for other pitchers?

Luckily, CHONE and Marcel provide, based on the couple other areas that I surveyed as well, perfect bookends to what I consider the reasonable projections for Greinke going forward; Marcel being the most optimistic and CHONE the most pessimistic. The league average FIP was right near 4.40 in 2008, so assuming that the projection systems maintain that the league will repeat itself, I plugged the FIP and innings pitched estimates from CHONE (3.97 over 137 IP) and Marcel (3.67 over 160 IP) into the winning percentage formula that I noted in my review of Brian Fuentes.

The resulting win totals were 3.5 wins out of Marcel and 2.5 out of CHONE. My own projection system for pitchers pegged Greinke at a tad over 3.0 wins itself so I feel pretty comfortable about the 2.5-3.5 range as a projection. Granted, I think that’s a projection with more possible room for upside than downside so the dollar figures are going to be a touch on the conservative side given that most systems are going to be skeptical about Greinke’s ability to log high-quality innings based on his years prior to 2008.

Salaries have stagnated this offseason and so I have been maintaining the use of $4.5 million per win as the general cost to teams on the open market. It might turn out to be a bit higher or lower once the market wraps up, but I don’t think it’s hitting the $5 million market that many thought at the onset of the offseason. With those parameters, plus the known 40/60/80 breakdown for arbitration values, I arrived at a fair market payout of $34 million to $48 million.

That is a pretty big range, but remarkably the Royals came in close to the bottom rung of that, with the AP reporting that the contract is worth $38 million guaranteed. The Royals are paying Greinke like a sub-3 WAR pitcher and for someone of Greinke’s capability, that’s a steal. The trend continues of extensions through arbitration years being a boon for teams and free gent signings being a mostly bust.

Continuing from the past few days, I this time separated Felix Hernandez’s starts based on the percentage of fastballs he threw within the first 20 pitches. Five times during the season Felix throw 65% or fewer fastballs during this opening stretch. For reference, those five games were: April 1st against Texas, April 6th at Baltimore, April 11 against Anaheim, July 28th at Texas and August 29th against Cleveland.

Counting up the batters that came to the plate within that 20 pitch barrier (including those that finished their plate appearance past the 20-pitch threshold), here were Felix’s relevant totals for those five games:

27 batters faced, 15 ground balls (56%), 1 fly ball (4%), 1 line drive (4%), 8 strikeouts (30%), 2 walks (7%).

That is a pretty dominant stretch and it came against some really good hitting teams. How did Felix do in the other 26 games?

147 batters faced, 55 ground balls (37%), 27 fly balls (18%), 20 line drives (14%), 31 strikeouts (21%), 14 walks + HBP (10%).

It doesn’t take someone well versed in FIP or tRA or any advanced metric to see which of those two lines is better. Interestingly, of the five games in the first sample, three of them were Felix’s first three starts of the year. I wonder why he quickly departed from it when even by traditional measures, he was successful.

Five games certainly isn’t a big enough sample size to draw definitive conclusions from, so I dearly hope Felix provides us with some more such games in the future. Then I would like to look not only at his performance during the first 20 pitches, but from pitch 20 onward as well to see if the improvement sustained itself throughout the game or not.

Will mixing in his off speed stuff earlier in the game cure Felix of all his problems? Of course not. He still has pretty lousy command of his fastball. What it will do though is force batters to not just sit fastball and the evidence leans heavily toward showing us that in those situations, Felix becomes a much tougher pitcher to face. The type of pitcher we all thought he would be by now after his debut in 2005.

Today, moving on from yesterday where we looked how Felix’s overall fastball percentage moved as the game wore on. This time, I drop the overall view and instead treat each pitch number separately to get a sense of when (besides early in the game) that Felix dials up the heater. Included in this graph is a red bar marking his overall average for all pitches and a rolling 5-pitch average to give you a more smoothed sense of the trend.

We see that Felix reaches his overall average at about the 20-pitch mark into a game and then hovers around that mark for the remainder of the start, with some obvious variation. It is still a clear downward trend over time.

I decided to zoom in on those first 20 pitches, where Felix throws a fastball just under 80% of the time. Take a breath and soak that in first; on average, 16 of the first 20 pitches that Hernandez offers at the beginning of a game are a fastball. That’s predictability to an extreme and if I can spot it, you better be sure that opposing advanced scouts can spot it.

Here I made a time line of Felix’s 2008 season and in blue you will see the percentage of fastballs thrown in the first 20 pitches for each start. Again the horizontal red line marks Felix’s overall fastball ratio and I have included a horizontal green line to mark Felix’s average fastball ratio amongst this sample.

Next time, I will take a look at the games where Felix does exhibit some differences in pitch selection at the onset.

(I know I already have another Part 1 out there without resolution, I’ll get to that eventually, but for the moment this has captured my attention)

Near the end of June in 2007, Dave Cameron wrote an open letter to then-Mariners Pitching Coach Rafael Chaves pleading with him to modify the game plan for Felix Hernandez when it came to starting out games. Namely, Dave pointed out the predictability of Felix throwing a high percentage of fastballs at the start of games. A year and a half later and has anything changed? I decided to take a look at Felix’s 2008 through the perspective of Dave’s letter and his intent.

Frankly if you followed any of Felix’s starts and were paying attention (or happened to participate in the game threads at Lookout Landing where I was constantly harping about it), you already know the answer to the posited question above. The answer is no. Now, I could leave it at that, but a 75-word post is not going to get anyone’s attention and besides which, I love making graphs.

Let us jump right into those graphs then. Here is a chart of Felix Hernandez’s fastball frequency over time; time, for the purposes of a baseball game, being measured in pitches. until it reaches his overall average frequency, about 67%.

If you find a pitch count number along the bottom (x) axis and move upward (y) until you reach the trend line, that value will give you the percentage of pitches, on an average start, that were categorized as fastballs up to that point in the game. So, after 11 pitches, roughly 84% of them had been fastballs. By the time Felix has thrown 40 pitches in a game, that ratio is down to around 74% and it continues to fall

It’s obviously not uniformly descending, but it’s really close and it paints a stark and unmistakable pattern. Felix starts out a game gung-ho about his fastball and slowly begins to work in his other pitches as the game wears on. It’s not even a gentle downward slope, but a rather dramatic curve, suggesting that the set of pitches one through about 20 and 20 through rest of game are rather disparate. Next, we’ll dive a little further into how Felix’s fastball percentage varies on a per pitch basis.

I am almost getting tired of writing about these. Yesterday, after a weekend full of rumors indicating that the two sides were close and a deal inevitable, Nick Markakis and the Baltimore Orioles officially agreed to terms on a six-year, $66.1 million contract.

Markakis was entering his first arbitration year so this contract buys out all three and three free agent years, with a mutual option on a fourth. That’s a lot of open market years and before even looking at the numbers, the value of the salary is sending me alert bells that this is going to be a team-friendly deal.

The projections want to knock some of Markakis’ offensive improvement in 2008 off and peg him closer to his 2007 level, which for a six-year projection seems fine to me. We also want to scale back his plate appearances to account for the possible, perhaps even likely given the contract’s duration, chance that he will miss time at some point. Assuming Markakis sticks in right field, we also have a decent sample size to now estimate his defense, +2 runs per 600 PA, and given his young age, it shouldn’t deteriorate much over the time period in question.

That leaves us with a calculation of +20 runs for replacement, +2 runs for fielding, -7.5 for position and about +23-24 runs for hitting. Totaling those yields ~37 runs above replacement, 3.7 wins and 16-17 million in value. 10% for long term security and the free market six year total is $90 million. Of course, given the 40/60/80 nature of arbitration, the Orioles are not buying Markakis at six years worth of value, but rather at 4.8 years. Given that, the expected value for the contract is $72 million. Actually less than I first thought. It’s still a good deal for the Orioles, but it’s not much of a steal and given that Markakis gets to wrap up a tremendous amount of financial security, you can hardly fault him for taking a slight discount. All in all, looks like a win-win.

Everyone else was signing arbitration buyout contracts (hey, Felix!), so Hamels and the Phillies wanted in as well, coming to terms on a three-year, $20.5 million contract. What is interesting about this contract is that not only does it not buy out any of Hamels’ free agent years, it does not even cover his entire arbitration eligibility. Cole Hamels qualified as a Super Two this offseason. A Super Two is a player with less than three years service time, more than two and in the top 17 percent of said group of players. Because of that, Hamels garnered an extra early year of arbitration.

The Super Two status throws a wrench into our value calculations since our standard formula for arbitration awards, 40%/60%/80%, are based on a three-year model. The easy solution is to keep those in place and just assume that the fourth year of arbitration would be around 100% of market value. My gut feeling however is that since arbitration awards are more focused on the playing time of the player rather than his stats, that this is not the most realistic solution and that something more like 30%/50%/70%/90% is more likely for Super Two’s. I will present the math under both assumptions so that we can get a better range of values.

Projecting Hamels is thankfully pretty simple. From 2006 through 2008, Hamels posted the following FIPs: 3.98, 3.83 and 3.72. The three projection systems that FanGraphs hosts all place 2009 Hamels between his 2007 and 2008 performance and inning totals so averaging the win values for those two years is going to get us a reasonable estimate for his 2009 projection. Cole accrued 3.8 wins in 2007 and 4.6 last season, making his 2009 projection 4.2 wins. Assuming $4.5 million per win for this offseason, that’s a $19 million value.

Under my proposed arbitration weighting, the Phillies are valuing Hamels at being worth $41 million on the open market for the next three years. That’s a little over nine wins total for that time period, a mark Hamels seems sure to better. Under the 40/60/80 line, the Phillies would be paying for 7.5 wins, something Hamels might exceed in just two years time. If the projection of 4.2 wins is at all correct, Hamels looks to be worth around $54 million in value over the next three seasons. Even if you think it’s a 20/40/60/80 line for Super Twos, that’s still a discount for the Phillies. There’s no ifs ands or buts about this one, the Phillies got themselves a killer deal here. Of course, Hamels made out well himself; he’ll go into his final arbitration year with a $9.5 million starting point, meaning he’s likely to command over $12 million on that final year and he still has all of free agency ahead of him.

Today was a very exciting day for us baseball analysts. Right about the time that we finally got used to what PITCH f/x could tell us about the game, we started jonesing for some form of HIT f/x to give us the other pieces of the puzzle. A full scale system of tracking batted balls off the bat and all the way to the fielder’s gloves will allow us to make massive strides forward in fielding metrics and even assist us in refining our offensive measurements as well.

Alas, the dream seemed far away as the expense required to track batted balls everywhere on the field is immense. Most of us tucked that dream away, waiting for the day but not conceiving that it could be anytime soon. Enter Matt Sisson who, while talking with Cory Schwartz of MLBAM, snagged some juicy scoops on what is to come in the world of f/x technology in 2009. I’ll let Sisson speak for himself:

Cory tells me that these improvements are to include an extensive “real time scouting” area in game day which utilizes Pitch-f/x data. The real time scouting would use the pitch data for the pitcher to show which pitch they are likely to throw depending on the count and situation and what zones are considered the pitchers and batters hot and cold zones. Pitch-f/x will also be expanded to provide more data and graphs for participating RSN’s to use in their broadcast as well as more data and graphs for clubs to use on their in-stadium scoreboards.

Cory also explained that we can expect the roll out of Hit-f/x, a system similar to Pitch-f/x that would use the technology already in place to track the initial batted ball data. Trajectory, angle, velocity, etc. measurements would all be recorded but the technology would be limited to just the initial batted ball data. The Hit-f/x system would not be able to track the entire trajectory of batted balls but from the recorded data, researchers would have the ability to correlate the recorded data with results data (hits, outs, errors, etc) and figure out answers to a range of questions including whether a hitter should try to hit more fly balls or if a hitter is having “bad luck” on their line drive rates. There is no firm time line for the roll out of this system yet but I was told that it is definitely on the radar for the ‘09 season. New software is being built and tested so its only a matter of time before we’re able to dive into another seemingly limitless goldmine of baseball data.

Well now. The improvements to PITCH f/x would be exciting enough, but the first inklings of HIT f/x data? This is fantastic news. We get to take a major step toward eliminating the problem of classifying batted balls based on someone arbitrary decision of ground ball versus line drive and line drive versus fly ball. This only helps augment what has been a rapid improvement in pitcher and hitter evaluation metrics and also gets us started on fielders.

If you want to follow along with the analysts attempts to figure out how to make use of this, I would suggest bookmarking the always informative Book Blog, namely this post in particular.

The Red Sox continue to avoid arbitration like the plague as Theo Epstein has added Kevin Youkilis to his list of arbitration-eligible players signed to long-term contracts. Apparently the deal is still unofficial at time of writing, but the details have sufficiently leaked to give us a good idea of the clauses. It’s going to be something around a four-year, $40-million deal with a fifth year option valued at $14 million with a $1.25 million buyout. Pursuant to tradition, we will evaluate the contract on the basis of the minimum guaranteed amount, that is four years and $41.25 million.

Youkilis may have only been a 2nd year arbitration player, but he was already heading into his age 30 season so this contract covers at least his age 30-33 years. That’s still a period where we’d expect Youkilis to remain effective, especially given his slew of good skills, but it is near the edge of danger and he certainly wouldn’t be expected to get any better. Furthermore, we have to account for a 2008 season that was far and above his previously established level of performance and a home park in Fenway that can vastly skew the numbers for right-handed hitters.

Dealing with projection first, we can note that in a rare case of agreement, all three of Bill James, Marcel and CHONE peg Youkilis as posting a wOBA in the .370-.382 range in 2009, pretty much dead on with what he did in 2007 (.373). Those figures are not park-adjusted, so we need to figure out what effect Fenway has. As a primarily doubles hitter, Youkilis has ample opportunity to exploit the green monster in left field and as such, controlling for his home park necessitates reducing his wOBA by about 2%, much in the same way we had to do with Jim Rice earlier.

That means that we’d project Youkilis to post a park-neutral wOBA in the .363-73 range next season, worth about 20 runs above average. Youkilis now has a sufficient amount of time logged at first base and with consistent UZR results in all three seasons, it’s good enough to go with his average performance there, worth six runs above average. 20 runs of offense plus six runs of defense and 20 runs for replacement (he’s projected to be right around that 600 PA mark) net Youkilis with 46 runs over average. He loses 12.5 for playing first base however, leaving him with a grand total projection of 33.5, or between 3 and 3.5 wins.

That would be worth between $12 and $16 million on today’s free market, but Youkilis was not eligible for that yet, still having two years of arbitration left to serve. Past history shows that arbitration awards tend to be worth 60% of market value in their 2nd year and 80% in the 3rd and final year. Throw in the 10% discount for security and expected aging and that means a fair contract for four years would have fallen between $37 and $48 million. The actual contract falls just about right in the middle of that, suggesting that the Red Sox signed a perfectly legitimate contract with Youkilis. It’s not really a win (nor a loss) for either side from a value standpoint. Frankly, I think the projection is a little on the conservative side and that the Red Sox made out well here, but it’s nothing crazy.