Archive for September, 2009

Using Advanced Metrics on The Machine Part 1

In less than a week, Joe Posnanski’s much awaited new entry into the book world officially arrives. The Machine — which as the insanely long subtitle will inform you – is all about the 1975 Reds is probably the most anticipated baseball literature release of the year. I’m guessing most of the readers of this site are also fans of Posnanski, so to rev the hype engine a little harder, how about some graphical looks at how that team stacked up to the other teams in the league?

First up, team wOBA:

reds1

That red dot, that’s the Reds. Creative I know. The Reds were the best offensive team in baseball during this season and it wasn’t all that close. Only Boston surpassed the .340 team threshold. The Reds lead the majors in BB% and were near the tops in ISO (and Speed rating, for whatever it’s worth) which you would expect from any top offense.

How did their player stack up against each other? Glad you asked.

reds2

This consists of every player with at least 100 plate appearances. In case you didn’t know – and holy smokes you better have known – Joe Morgan was a heck of a ballplayer. Morgan’s 21 BB% is jaw-dropping and his .327/.466/.508 line is just incredible. Hopefully Morgan’s buffoonery in announcing hasn’t caused some statistically orientated fans to discount his playing career because that would be an absolute crime.

How often is the second baseman the best offensive player on the team while the catcher is the second best? Johnny Bench’s .283/.359/.519 line is almost equal to George Foster’s .300/.356/.518 and Foster played the corner outfield. That guy Pete Rose was pretty good too, as was Ken Griffey and Tony Perez.

Later on a look at their pitching staff.


Buster Posey Promoted

The wait is finally over.

The Giants promoted Buster Posey to the major leagues today. Day two of September 2009 becomes a day of joyous occasion for Giants fans who grew tired of Bengie Molina’s .261/.281/.437 line through nearly 500 plate appearances. Molina’s allergy of walks is shared by back-up Eli Whiteside.

Obviously differences between the majors and minors exist when it comes to drawing a walk, still, consider that the Giants backstops have a combined 554 plate appearances and 18 walks this season. Meanwhile Posey has just fewer than 550 plate appearances throughout the minor leagues while boasting 70 walks.

Assuming the Giants plop him into the lineup immediately and Posey hits to a league average (meaning an absolutely average MLB hitter, not NL catcher) tune rather than the .321/.391/.511 line in Triple-A Fresno, he’s still a pretty nice upgrade to add this late in the year. With a game separating the Giants and Rockies in the wild card race, any kind of upgrade at this point helps. Do the basic math and assume Posey takes 100 plate appearances away from the bumbling duo and the upgrade is a little less than two runs.

Notice we’re dealing with 100 plate appearances though. That’s a pretty small sample size and anything can happen within a stretch that leaves Posey looking better or worse than league average. It’s not impossible that Posey could actually finish with a worse line than his lacking counterparts. I know, outlandish right? But Matt Wieters is proof that sometimes phenoms don’t light the league on fire immediately:

Wieters: .263/.309/.372
Giants Putrid Crew: .257/.283/.409

Of course that’s not park adjusted or league adjusted, so there’s room for improvement in that comparison. My warning for Giants fans is to not take these 100 (or however many) plate appearances too seriously. Oh, and to not pin the playoff hopes on top of his shoulders.

That being said, I’m cheering for a Giants post-season berth. The folks on the east coast need to experience Tim Lincecum.


Bartlett Finds Power Too

This afternoon, we talked about the surprising power surge of Kendry Morales, who is beating his projected ISO by about about 120 points this year. However, we always knew that Morales had some power – we just didn’t expect his doubles to turn into home runs like this. If we want to look at a guy whose really having an out-of-nowhere power surge, then we have to turn to Tampa Bay, where Jason Bartlett continues to have one of the craziest seasons in recent history.

Bartlett’s ISO by season, 2005 to 2008: .094, .084, .096, .075. He had 12 home runs in about 1,700 plate appearances. He was consistent in his lack of pop, which is kind of what you expect from someone with his physical build and defensive chops. Bartlett was a classic glove first middle infielder who drew some walks, stole some bases, and tried not to kill too many rallies.

This year, however, has been a totally different story. He’s at .338/.398/.522, and his ISO is .184. He’s doubled his career home run total in 109 games, and it hasn’t just been balls flying over the wall – he’s got 32 doubles+triples, one off his career high, which he set in 2007 with 120 extra plate appearances.

Even an injury can’t slow him down. He had a monstrous start to the season, then missed three weeks in June with an ankle injury. After coming off the DL, he looked like he was reverting to previous form, hitting just one home run and slugging .402 from mid-June through the end of July. But he’s found the power stroke again as of late, hitting .349/.432/.560 since the beginning of August.

As a 29-year-old with no track record of this kind of ability, it’s hard to figure what to make of Bartlett. For all the talk that Ben Zobrist’s breakthrough offensive performance has gotten, he’s not the only former slap hitter now whacking the ball down in Tampa. With Reid Brignac waiting in the wings and newly acquired Sean Rodriguez in the fold (plus the aforementioned Zobrist and Akinori Iwamura), the Rays have enviable middle infield depth. If they’re convinced that they’ve fixed Bartlett somehow, then he’s an all-star that they should be hanging onto.

More likely, however, is that Tampa’s going to try to shop him this winter coming off a career year and one season away from free agency. It will be interesting to see how many teams believe this version of Bartlett is for real.


AFL Preview: Mesa Solar Sox

The Arizona Fall League rosters were announced by Major League Baseball last week. The league allows up-and-coming prospects (usually from high-A and double-A levels, as well as recent high draft picks) to continue honing their skills away from the fall instructional leagues held by each organization. Play will begin in early October and run until late November with the six teams – each one made up of five organizations’ players – continually facing each other.

Over the next week, we’ll take a look at some of the more interesting names on each team. The rosters that were recently released are preliminary rosters and some players will be added, while others could be removed. On Monday, we took a look at the Phoenix Desert Dogs. On Tuesday, we had a gander at the Scottsdale Scorpions roster. Today, let’s view the Mesa Solar Sox roster.

The Mesa Solar Sox (Click HERE for the entire roster)
Minnesota, Chicago (NL), Los Angeles (AL), Florida, Boston

Andrew Cashner | RHP | Chicago
Drafted as a college reliever, Cashner has made a successful conversion to the starting rotation. He opened the 2009 season by allowing just 31 hits in 42 high-A innings. The right-hander then moved on to double-A where he’s had similar success, although his walk rate has risen from 3.21 to 4.56 BB/9.

Tommy Mendoza | RHP | Los Angeles
This right-hander had a breakout season at the low-A ball level in 2006 at the age of 19. The next two years, though, were a disappointment. Mendoza’s value has risen again in 2009, although the ceiling is a little more modest this time around. Still only 22, he profiles as a No. 3 or 4 starter.

Welington Castillo | C | Chicago
With Geovany Soto tanking with a capital ‘T’ there is suddenly a desperate need for catching depth in the Chicago system. However, Castillo chose a really poor time to have an off-year at the plate. He is a very promising defensive player with an excellent arm behind the dish. Unfortunately, Castillo has proven for two straight seasons now that he cannot hit right-handed pitchers, which significantly hinders his chances of ever playing on an everyday basis.

Hank Conger | C | Los Angeles
Conger has the opposite problem to Castillo. The slugging catcher can mash with the best on them, but his defense is bad enough that no one believes he can remain behind the dish. After struggling with injuries in 2007 and 2008, the switch hitter finally played a full season and is currently hitting .294/.365/.423 in 428 double-A at-bats. A former first-round pick, Conger is still just 21 years old.

Starlin Castro | SS | Chicago
The Cubs organization was aggressive with Castro in 2009 after he hit .311/.364/.464 at rookie ball in 2008. The 19-year-old shortstop hit .302/.340/.391 with 22 steals (in 33 attempts) at high-A in 2009 before being promoted to double-A, where he’s held his own in 24 games. Incumbent MLB shortstop Ryan Theriot needs to be looking over his shoulder.

Matt Dominguez | 3B | Florida
Three prep third basemen were nabbed near the top of the first round during the 2007 draft: Dominguez, Josh Vitters, and Mike Moustakas. All three players have had their struggles but Dominguez (who was drafted last at 12th overall) is the only one of the trio to make it to double-A. Although he’s struggled a bit with the bat, Dominguez has the edge in the field.

Josh Vitters | 3B | Chicago
Vitters has had troubles adapting to new levels in his brief pro career. After starting out very well in 2009 at low-A (His second attempt at the level), he has struggled mightily upon a promotion to high-A. At the root of the problem is his lack of patience at the plate, which has manifested itself in the form of a dismal 2.4 BB% in 2009.

Bryan Petersen | OF | Florida
I highlighted Petersen prior to the 2009 season as a possible breakout candidate in 2009. He hasn’t made me look like a genius, but he’s held his own in double-A as a 23 year old. His power has taken a step back in 2009, with his ISO dropping from .200+ to .120. He’s also struggled on the base paths with only 12 steals in 24 attempts. On the plus side, he’s shown a consistent ability to hit for average and he continues to trim the strikeouts. If he cannot regain some ground with his power and speed numbers, he’ll develop into a fourth outfielder.

Michael Stanton | OF | Florida
Petersen’s teammate should need no introductions. Stanton is one of the biggest mashers in the minors and fans are constantly asking for updates on his ETA, especially after he spent the first part of 2009 demolishing high-A pitching. The holes in the talented outfielder’s plate approach have been exposed by double-A pitchers, though. He’s currently hitting .234/.311/.453 with a strikeout rate of 33.6 K% in 274 at-bats. Stanton will likely opened 2010 in double-A with a mid-season promotion to triple-A possible if he makes the necessary adjustments, which have so far eluded him. At the age of 19, time is very much on his side.


Feliz Shows no Signs of Slowing Down

By now everyone knows that Neftali Feliz is having an incredible run as a reliever for the Rangers. After last night’s two inning four strikeout performance he has racked up 26 strikeouts to just one walk over 19 and two thirds innings.

For Feliz it starts with his blazing fastball that averages over 96 mph and is regularly in triple digits. But unlike another triple digit flame thrower he can locate the pitch in the strike zone. He throws it early in the count, when he is behind (which is rare), and at times with two strikes up in the zone as a strike out pitch. Here I plot the fastballs with the swinging strikes outlined in black and called strikes opaque.

pitch_loc_fa

Again he is always around the zone (amazing for how hard he is throwing the ball). If he misses it is rarely by much, expect for up (and those are mostly when he is ahead in the count). The swinging strikes are, as expected, mostly on pitches up in the zone. While the called strikes are for the the most part on the outer half of the plate.

Feliz also throws a changeup and curve. He uses both generally later in the count, the curve more to RHBs and change more to LHBs.

pitch_loc_other

You can see that curve has been death to righties. Tons of called strikes in the down-and-away corner of the zone, and tons of swing strikes outside of the zone. To lefties he has kept the change either far away or very low in the zone.

Just an incredible month-long dominance, striking out 26 and only walking one batter.


Angels re-sign Teixeira… sort of

Over the winter, the Angels lost out on a bidding war to retain Mark Teixeira and watched him end up in pinstripes. In order to fill the hole on their offense, they gave their first base job to… Mark Teixeira?

Tex, 2009: .280/.380/.541, .392 wOBA, +4.0 wins
Kendry Morales, 2009: .314/.355/.597, .398 wOBA, +3.8 wins

Morales doesn’t have the same style of production as T-Rex, but in terms of overall value, the Angels have essentially matched what the Yankees got from Teixeira, and they did it with a guy who was viewed as a bust coming into the season. It wasn’t like Morales was a guy with projections all over the map, either – CHONE had him as a .327 wOBA guy, ZIPS at .325.

This is, of course, way out of line with anything Morales had done before. These are better numbers than he put up in Triple-A last year, when he posted a .387 wOBA. The main difference has been the power, where Morales is posting a .287 ISO, tied fort he seventh highest mark in baseball. That’s 120 points higher than the projections had him at, and really the only area where he’s exceeding what we should have expected by any large margin.

Power is one of those things that stabilizes fairly quickly, however, so while there are examples of Brady Anderson style flukes, we have to assume that our previous beliefs about how much thunder exists in Morales’ bat were wrong. It’s really, really hard to luck your way into 69 extra base hits in 506 plate appearances.

ZIPS still projects a pretty heavy regression for Morales going forward, as his rest-of-season projection has his ISO dropping to .202 – 40 points better than the preseason projectoin, but 80 points lower than his current ISO. But the projections have changed enough that, given what we know right now, we can say the Angels certainly made the right choice in letting Teixeira go and replacing him with Morales. Given their relative price and performance, Morales is the clear choice.


Geovany Soto’s Sophomore Jinx

Geovany Soto is struggling and he doesn’t know why. The reigning NL Rookie of the Year has scraped together a measly .212/.318/.371 line and is now losing playing time to his illustrious backup, Koyie Hill. Lining up this season’s stats to his ROY campaign, I can see why he’s baffled.

First let’s take a look at his batted ball types.

3707_C_season_blog_9_20090831

(Blue is fly balls)

Uh-huh. No major changes, really. He’s hitting just a few more fly balls, but less are leaving the yard – down to 10.2% compared to 14.7% last year. That says something about why his power production is down, but it doesn’t explain why his batting average is hanging around the Mendoza line. Consider also his plate discipline numbers. Soto’s walks are up and his strikeouts are slightly down. He’s swinging at fewer pitches out of the zone – 20.1% last year, down to 18.1% this season – and he is making more contact when he does swing –74.7% last year, 77.4% this year. So he’s being more selective and making more contact when he does pull the trigger. It’s just as if he is almost always hitting the ball right at someone.

Soto has had some issues with his shoulder and oblique this year, which could be the culprit at least in part, but listening to Soto talk, he doesn’t seem to think that’s the case. Soto seems to attribute his woes to good old fashioned bad luck, and I’m not quite sure I can blame him.

3707_C_season_blog_7_20090831

Playing around with THT’s xBABIP calculator, Soto’s expected BABIP is .314. His actual BABIP: .245. Last year Soto was playing a bit over his head. This year he appears to be suffering some sort of horrid luck that Billy Sianis wouldn’t wish upon his worst enemy.

There may be more to it than luck, of course. Feel free to fill me in, someone.


Sean Rodriguez, Victor Zambrano, and Mark Langston Attend a Wake

On Monday I addressed the Scott Kazmir deal and portrayed the player to be named later someone who probably wouldn’t impact things too much.

I was wrong.

He is Sean Rodriguez, confirmed by a minor league teammate on Twitter yesterday and shortly thereafter an official team press release.

He’s a 24-year-old middle infielder with a ridiculous line of .301/.402/.622 in Triple-A. It comes in a hitters park where the team average line is .273/.342/.434, but Rodriguez still possesses some skills that appear to separate him from a figment of the ballpark. Baseball America had him amongst the top 10 prospects in the Angels system in their 2008 pre-season rankings and didn’t appear on the 2009 list because he received more than 150 plate appearances last season causing him to lose his prospect status.

He has some questions as to his contact rate and where exactly he’ll play. At the same time he’s close to the majors – as in, he’ll make his Rays debut within the next four weeks – and should contribute in a meaningful manner almost immediately.

Who knows where he takes the diamond for the Rays. If I had to guess, I’d peg him for a super utility role to begin 2010. Meaning he’ll play some outfield and a lot of infield in the same manner that Ben Zobrist did this season before becoming Ben Zobrist. This gives the Rays some added roster flexibility with regards to Akinori Iwamura’s pending option and Jason Bartlett who will enter his second year of arbitration off a career year.

Consider this: the Rays turned a few seasons of Victor Zambrano into six years of Scott Kazmir. They’ve now turned two-three seasons of Kazmir into six of Matthew Sweeney, Alexander Torres, and Sean Rodriguez. Obviously there’s a non-zero chance at each of them busting or working out, but that’s quite a haul for a pitcher with a career FIP of 4.95.

Maybe Victor Zambrano could be the gift that keeps on giving for another few years, much like the Mark Langston deal for the Mariners. For those who don’t know, Langston was dealt in May of 1989 for Randy Johnson, Gene Harris, and Brian Holman. Johnson was dealt in 1998 for Freddy Garcia, Carlos Guillen, and John Halama. Garcia was dealt for Miguel Olivo, Jeremy Reed, and Mike Morse. Morse brought back Ryan Langerhans; meanwhile Reed was in a three-team deal that brought back a boat of players that included Aaron Heilman who was dealt for Ronny Cedeno and Garrett Olson. Cedeno was then dealt in a package for Jack Wilson and Ian Snell. That deal was made about 7,405 days ago and still may add branches.

Maybe someone will write about the Zambrano deal in another 7,405 days – sometime in 2029.


Put Away the Torches

Add the White Sox to the list of teams people will call quitters since they effectively “threw in the towel” by dealing Jim Thome and Jose Contreras last night. The White Sox are six games back of first place Detroit and two and a half back of second place Minnesota. To date they’ve played a little under .500 ball and with 30 games remaining, CoolStandings has their playoff odds at 6%. That is to say, for every 1 million simulations, the White Sox make the playoffs 60,000 times while they miss out the other 940,000 tries

You don’t have to run simulations to understand the chances weren’t good anyways. The Tigers have 32 games remaining, and if they simply played .500 ball Chicago would have to go 22-8 to tie Detroit for first. That’s without taking Minnesota into account. If Detroit plays like they have all season, at a .531 clip, then Chicago has to go 23-7 to tie, 24-6 to win and this is again without taking Minnesota into account.

The bigger offense here might be missing out on a chance to land draft picks or potentially costing them a pick. Per Eddie Bajek’s latest Elias rankings, Jim Thome is a Type-A while Jose Contreras appears to be on the outside looking in on Type-B status. Maybe the Dodgers have agreed against offering Thome arbitration, but otherwise if the White Sox plan to bring the slugger back into the fold next season, it would cost them – or whoever signs him, for that matter – a draft pick.

The draft pick angle should be focused on more while the playoff potential angle is focused on less. Either way, I don’t think Kenny Williams should take much heat for doing what amounts to a favor for Thome.


From the Department of Advanced Statistickery

Earlier this year, resident ubermensch Dave Cameron submitted for the Reader’s consideration a piece in which he introduced another of FanGraphs’s cutting-edge metrics — namely, Hyperbole Index. Hyperbole Index, or HI, took the world of baseball analysis by storm*, leaving breathless readers of all descriptions, from ESPN Senior Writer Rob Neyer to cunning linguist Snoop Dogg (the latter of whom gushed, “That’s shit’s off the hizzy!”).

*Denotes possible use of hyperbole.

Because I try to have as few original thoughts as possible throughout the day, I’ve opted, in this space, to ride the frig out of Dave’s equally well-pressed and metaphorical coat-tails and share with the Wide Readership some very experimental metrics on which I’ve been working as part of my contribution to FanGraphs’ Department of Advanced Statistickery.

Two notes before I unveil said metrics.

First is that it’s important to recognize that the stats which follow are very much in the development stage and therefore lack the precision and accuracy of a finished product. Much like the Master Painters of yore, I’m only ever involved in the conception of a new stat. The grunt work I pass long to my apprentice statisticians, whose responsibility it is to f a whole bunch with Excel and crap.

Second, I’d like what follows to serve as a bit of a repost to certain readers who’ve called into question the empirical rigor of my first couple-few submissions to FanGraphs. While most readers have been supportive, there are some — and here I’m thinking particularly of user PlayOnWords at D-Rays Bay, who suggested that I’m “turtlenecking”* FanGraphs — there are those who would break that most important of commandments, “Let a player play.” I believe the following work will serve as a giant “Booyah” to those critics.

*Which, if I were to do that to FanGraphs, I’d at least have the decency to buy it dinner and stuff first.

Here now are five metrics on which I’ve been working:

Stat
Douche Factor

For Short
DF

What It Measures
Probable douchey-ness of a player, expressed as a percentage from 0% (very probably NOT a douche) to 100% (almost definitely a douche).

How It’s Calculated
While I’m still experimenting with the degree to which they ought to be weighted, I’ve identified, by means of extensive testing and research, what the main components of the metric are.

ChrisBritton2DF considers the three following components:

1. Stupid Jewelry Factor — In which the number of Phiten-brand “energy necklaces” is considered.

2. Facial Hair Factor — In which the of facial hair is considered. In most cases, the goatee is douchey. Clay Zavada is merely comical and is not penalized.

3. Eye to Jaw Ratio — In which the distance between the eyes is weighed against the distance across of the jaw. Close-together eyes paired with wide-set jaw denoted probably douchey-ness.

Probable League Leaders
Chris Britton (see photo)
Joba Chamberlain
Brad Penny

***

Stat
Whiffle League Equivalents

For Short
WLEs

What It Measures
While a lot of work has been done to assess how minor league performance might be predictive of future major league performance (see: MLEs), very little has been done to investigate how minor or major league production might translate to wiffleball. As you can imagine, the wiffleball lobby isn’t happy about this one bit, and have recently made it their business to harass Yours Truly. Not that I can blame them. As an avid wiffleballista myself, I’m shock-and-awed at the degree to which wiffleball’s place in history has been ignored.

How It’s Calculated
I don’t care, just do it.

Probable League Leaders
This guy.

***

Stat
WOPS

For Short
WOPS

What It Measures
Contrary to what the Reader might think initially, WOPS isn’t Weighted OPS or Walks-plus-OPS or anything involving OPS at all. Rather, WOPS is an attempt to measure the relative Italian-osity of major/minor leaguers. If you don’t get immediately how hilarious this is, click here.

How It’s Calculated
By dividing the the total number of letters in a player’s name by the number of vowels. The closer the result is to 1, the higher the WOPS.

Also, by going to a player’s house/apartment and seeing if his couches have plastic on them.

Also, by considering the model year of the player’s Camaro.

Probable League Leaders
Mike Napoli
Silvio Berlusconi
Val Pascucci

***

Stat
Total Recall Bases

For Short
TRB

What It Measures
No idea, actually. It’s just I thought it might be possible to do some promotional tie-in work with the film of the same name. I’ll have my people call someone else’s people.

How It’s Calculated
Dollar signs.

Probable League Leaders
Give me money.

***

Stat
Enthusiast Quotient

For Short
EQ

What It Measures
The degree to which a player either (a) has warmed or (b) is likely to warm the cockles of the Enthusiast’s heart.

How It’s Calculated
Sir Philip Sidney once quothed*, in re the ideal writing process: “Look into thy heart and write!” To calculate EQ, you do pretty much the same thing, except instead of “write” at the end, instead you “assign a number with which you feel comfortable.”

*Which, that’s the only way people used to speak, I’m pretty sure.

Probable League Leaders
Mark Bellhorn
Jeremy Hellickson
Charlie Haeger