Thursday, August 28, 2014

2014: COMPLETE GAMES #85, #86, #87, #88, #89, #90, #91...

Not surprisingly, when one's schedule gets frenetic, the gods go to work and paint you into a corner...so it was with the CG watch last week. In the midst of much furor elsewhere, we were delivered our first 4-CG day of the 2014 season (8/21). Interestingly enough, three of these were complete game losses. (We'll have to do some research to determine what day in baseball history had the most of these.)

Of the four who threw CGs that day, only Brandon McCarthy (#88, NYY vs HOU) emerged with a win--a 3-0, four-hit shutout. Dallas Keuchel (#85, HOU vs. NYY) was his unlucky counterpart in that game, making it into only the second game in which both pitchers went the distance (the other occurrence, as you may recall, was on 7/31, when the Dodgers' Clayton Kershaw beat Julio Teheran and the Braves, 2-1).

Elsewhere, the Padres' Tyson Ross (#86, 2-1 loss to the Dodgers) and recent Tiger import David Price (#87, 1-0 loss to Tampa, his old team, despite allowing just one hit) completed the rare hat trick of same-day CG losers.

So our calendar display of 2014 complete games now adds a new color to signify four CGs in a single day. (Someday we will create a CG calendar display for a pre-expansion year, and see if we can run out of colors...but that's someday.)

We should pause for a moment to note that 2014's CGs are more prevalent in the middle of the week (53 on Tuesdays-Thursdays) than on the weekend (34 on Fridays-Sundays, with just 4 on Mondays). Not meaningful, probably, but worth a passing mention.

Meanwhile, CGs continue to agglomerate: Drew Smyly, the man traded for David Price, gifted his new team (Tampa) with a two-hit shutout (CG #89, 8/22)  over Toronto (Rays won, 3-0). Madison Bumgarner (#90, 8/26) and the unsinkable Colby Lewis (#91, 8/27) both posted their second CG during the month of August.

Lewis's game is notable for being the first time in 2014 that a pitcher allowed four or more runs in a CG and was the winning pitcher. There have been only four such instances where CG pitchers have allowed 4+ R in '14: their record is now 1-3 this season. The WPCT in such games, from 1914 to now as captured in the Forman et fils Play Index, is .435 (7173-9292); since 2000, however, that WPCT value is much lower (30-147, .169), which demonstrates how complete games have increasingly become elite games.


Wednesday, August 27, 2014

GOING OWAR INSTEAD OF AWOL...TOP HITTERS IN HISTORY THRU A SLIGHTLY ALTERED LENS

We are both the first to admit and the first to claim that far too much time is spent by all variety of baseball obsessives in compiling lists of the "all-time greats." We are not immune from this behavior, though we are much more circumspect about it than most.

Oops..wrong Bill James. But you can read
an interesting account of "Seattle Bill's" life and
all-too-brief times as a pitching ace here...
It's a natural temptation in that it allows for the application of varying forms of grandiosity (even amongst those of us who work a well-practiced "aw shucks" act while spilling their opinions all over the rest of us like a drunk with trembling hands).

Bill James made such behavior safe for "the masses" thirty years ago by whipping out the systematizing apparatus attached to his "family jewels" (yes, we get extra points--and, of course, extra "goodies"--for highlighting our salacious phrases in such felicitous bad taste thanks to our sponsor "Fright Quotes R Us", or "FQRU" for short...) and publishing his Historical Abstract. (One of our favorite wags redubbed it the Hysterical Abstract--a shockingly good shoe-fit, in fact--and the wag has been dogging the tale ever since: the 1999 version of the HA re-set the standards and ground rules for grandiosity.)

Bill is also the guy who conjured up the first Wins Above Replacement (WAR) system, one of the tools that allowed him to add grandiosity to gravitas. That was actually a Good Thing, and while it (along with much else) has been used and abused over the past thirty years, the portion of WAR that measures offense is much less bloodied than its defensive counterpart. So after a sufficient number of sidelong glances, we're going to trot it out once again, albeit with a bit of a twist.

The big, honking, impossibly colorful (and likely blurry, until you find the courage of your convictions--or maybe just your parking tickets) chart that we display here uses OWAR (with data kicked out by the Play Index at Forman et fils, a place a little less like the Land of Nod than the Other Two Alternatives That Dare Not Cross Our Everlovin' Lips...) as a rate stat.

There will be objections. The first one will be global. WAR is a counting stat--or so the claim will go. Rather than bog things down further, we will simply say that we disagree. WAR is also a rate stat--if you bother to do it.

But since it's usually a means to an end even when it's advertised as an end unto itself, it's really OK to use it in unfamiliar (and, frankly, unexamined) contexts. Such as this one.

The second objection is contextual: OK, you're gonna force a rate stat version of OWAR down our throats. That's not your worst offense (and here we thought that those records had been sealed away, even from the prying eyes of the Intelius generation). But couldn't you use something more akin to the way baseball seasons work? Why the eff do you insist on using OWAR per 1000 plate appearances??

There are two reasons. First, we are just as ornery as we've always been, and there really is no better place to demonstrate that than right here. Second, we wanted a rate measure that would produce a bit more conceptual distance between the players being measured.

Nope, no Mona...at least, not yet.
So, with that, let's talk about the chart, which you've already seen above (believe us when we say we'd never make you read this much text without some images to break things up--though we stop short at showing you Derek Jeter morphed into a cheezy version of the Mona Lisa...or do we??).

The chart, OK, the chart. OWAR/1000 PAs scales it on the left, players are festooned according to their most frequently played defensive position. This shows us distributional scarcity of top-end players, and, as you'd likely expect, catchers are the scarcest in the 8+ OWAR/1000 PA configuration.

AM to DM: "The last time you
mentioned me in your goddam blog
THIS is what happened. This is your
last warning, you bum: mention me again
and I'll sue you for child support even
though you're not the father!!!"
Surprisingly (perhaps), it's not SS or 2B (the other representatives at the left extreme of Bill's "defensive spectrum"--as opposed to the "spectrum of defensiveness" that his wig-wag followers have seemingly absorbed into their bloodstream...) who are bereft of top-end offensive performers: it turns out to be third basemen (though none of them, alas, are named Vanessa Minnillo--or Alyssa Milano, for that matter).

Players with less than 6000 lifetime PAs are shown with asterisks (*). Hall of Famers are shown in bold, active players in red. Players still in HoF consideration are shown in italics. And the great hitters who are the game's Meritocratic Pariahs (which happily enough, rhymes with Mariah...) are shown in plain text with boxes around them (yes, that's right, we traffic in narcosynthetic visual metaphors even while we juggle Derek's impossibly frenetic social calendar).

Many of the players in red (Joe Mauer, Andrew McCutchen, Hanley Ramirez, Ryan Braun) will likely dip below 8 OWAR/1000 PAs before they hang up their spikes, but it's interesting to catch them in mid-career just to get a sense of how their upside fits into this continuum.

And then there's A-Rod, destined to join those key Pariahs (Barry Bonds, Joe Jackson, Dick Allen, Mark McGwire) as the greatest hitters with plausible-length HoF careers to be barred from the silent room of plaques.

So--those are the best hitters of all time, according to this application of OWAR as a rate stat. Those who use WAR in its pick-up-sticks incarnation will wave it off, but so be it. (They are no longer the wave of the future anyway.) We just suggest that the compromise between moralizers and modelers be handled by adopting an approach such as this one and simply enshrining all of the hitters who appear on this list (not counting active players, of course). That's right...and that means "King Kong" Keller, too. (No, we are not shilling for the FQRU folks when we say this--though, come to think of it, what a fabulous tie-in...can you say "ca-ching"? Of course you can...hell, you KNOW you can!!)

Monday, August 18, 2014

2014: COMPLETE GAMES #83, #84

We were spared two more "cheap CGs" when the San Francisco Giants' protest was upheld and their 8/19 contest with the Cubs was "recategorized" into a suspended game.

Everyone in the Bass family is
"O-fer" vs. Rick Porcello...
Clayton Kershaw's fifth CG of 2014 (8/16, #83; keep in mind that one of these is, in fact, a "cheapie"...) proved to be the sixteenth losing distance-going effort in 2014. Kershaw gave up a couple of homers, allowing the Brewers to emerge victorious over the Dodgers, 3-2.

And last night (8/20, #84) Rick Porcello got his groove back with a three-hit shutout vs. the Rays at Tropicana Field in Tampa. The Tigers, now chasing the Royals, won 6-0 and climbed back to just a half-game back.

(Yes, some of you will notice that we cheated with the date stamp. When the post begins with an asterisk, all bets are off!)

Currently the WPCT when pitchers toss a CG in 2014 is .809 (68-16). And the current pace for CGs now sits at 108--which is an exact match for the lowest total in a season set back in 2007.

THE RANDOMIZING OF HIGH LEVERAGE*

*coming soon!

Friday, August 15, 2014

MEET THE NEW BOSS...SAME AS THE OLD BOSS

Oh, to be a fly on the wall. Fortunately, though, we've got that covered for you--with that truly unique baseball correspondent Buzzy the Fly, who not only survived a near-death experience at the hands of Commissioner-elect Rob Manfred, but has been buoyed by his participation in a special "pesticide resistance program" (which was not funded, despite rumors to the contrary, by our shadowy sponsors "Fright Quotes R Us").

We really need winged journalists these days, given that baseball's coverage often seems as "embedded" (there are those FQ's again--damn, but it's habit-forming!) as what we had to endure during the Iraq fiasco. Reading the news articles at MLB.com is the perfect cross between the Wall Street Journal editorial page and Pravda.

Buzzy, of course, has a somewhat different story about how things went down in the Baltimore putsch, much of it revolving around the real reason why resistance to Budzilla's template for corruption finally reared its ugly head.

But the bottom line is that it's still "the bottom line" that prompted the mini-revolt. Buzzy's recordings indicate that the "petulance quotient" for billionaires has followed the general trend of the stock market over the past five years.

Jerry Reinsdorf, "light-headed" back in July, was not
really the instigator of the "mini-revolt" against Rob Manfred
earlier in the week.
It's not enough to have a wired monopoly with carefully calibrated bylaws that skirt the level of scrutiny that baseball deserves to have placed on it (though, to be fair, one could say this about all too many aspects of American business).

Territorial rights and old grudges were not the true battleground in Baltimore, as has been widely reported. Buzzy's tapes indicate that it's the future of media that was instrumental in whipping up a last-minute frenzy of factionalizing. And part of the fallout in that area has to do with who will control the creation of "advanced data."

Now, none of these folk can be seen as the "good guys" (FQ alert). But it was Budzilla who decided to leave the "advanced data/media" issues in the roiling region of endless entrepreneurial kerfuffle. And his hand-picked successor Manfred will do the same. As certain owners have discovered, that is actually more of a constraining scenario for the use and growth of such information products than it is an opportunity or a strategic advantage.

This situation was not "resolved," it was simply tabled. But what it means for the next seven years is that we'll have more of the same with respect to the blighted vision of the game that has prevailed. The reign of an approach utilizing ineffectual committees whose answers are known before the questions are even posed will continue. Manfred will do his best to live up to The Who's dismay in "Won't Get Fooled Again": "Meet the new boss...same as the old boss."

But it's not an issue that will stay tabled for long. We may need to have Buzzy train an army of winged correspondents to keep up with this.

2014: COMPLETE GAME #82

The Royals have gotten hot again, and it's hard to argue with the fact that they've been doing some of that against good teams (though it's possible to wonder if Billy Beane didn't leave his offense just a little too depleted--particularly from the right side of the plate--in his reaching out for Jon Lester). The KC strengths (bullpen, defense) have been boosted of late by an uptick in hitting (particularly in clutch situations).

And now Jason Vargas. Like Ervin Santana last year, Jason has found a way (thus far) to keep the ball in the park--something that's not really been his forte in the past. Oddly, however, he's not really pitched all that well at Kauffman Stadium this year (4.27 ERA vs. 2.16 on the road).

QMAX (aka the Quality Matrix) shows us that Jason's 2014 numbers aren't really that much better than his lifetime performance. His QWP is only .520. His QERA+ (our version of adjusted ERA) is 104.

But what can one say after a three-hit shutout in which he retired the last 23 batters he faced? What Sandy Koufax said in his autobiography--we paraphrase: "[The .500 pitcher] always looks better than he really is, because his good starts are so good."

The Royals are looking to make the playoffs with eight of their ten regular hitters producing OPS+ values below league average. Their starting rotation is overachieving (QMAX and FIP, as is often the case, are in agreement here). But they have a favorable schedule down the stretch. It won't only be the "midwestern angsters" who'll be holding their breath.

Getting back on topic for the post: CGs have slowed down again in the past week. The projected total for 2014 has slipped back to 109. (The record for fewest CGs in a season, you might remember, is 108).

Wednesday, August 13, 2014

ELIAN EASES INTO THE RECORD BOOKS

It's precisely because baseball is as close as it gets to a pure meritocracy that the game's marginal players are so interesting. And that brings us to the saga of Elian Herrera, who managed (against long odds) to get himself into the baseball record book a month ago.

Herrera, now 29, has only a little bit of pop in his bat, mostly doubles, and it took him five years to get out of Class A ball after being signed out of the still-burgeoning Dominican baseball factory back in 2003. The Dodgers didn't even put him into rookie ball until 2006. It became clear that Elian's only path to the big leagues would be as a jack-of-all-trades type, so the natural second baseman quickly volunteered his way into playing the outfield. Over the next four years, the Dodgers would move him all around, providing him with enough polish at six defensive positions to make him into a viable utility man.

He finally made it to Albuquerque (AAA) in 2012, and the Coors-like conditions at his home field proved more  than congenial: Elian hit .400 at home and found himself up with the big club in early May.
God bless him...Elian Herrera is STILL struggling out there
in the outfield.

For about thirty games, it looked like it might be a Cinderella story for Herrera, who was not only hitting over .300, but was getting on-base at greater than a .400 clip. But reality set in at that point, and Elian went into a major tailspin. He struggled in the outfield and at the plate, and the Dodgers sent him back to Albuquerque in early July.

After Elian spent another year there in '13, the Dodgers opted for Justin Turner as an upgrade at utility player, and Herrera found himself on waivers. Former Dodger organization man Ron Roenicke, now manager of the Brewers, tipped off his club that Herrera was worth a flyer, and so Elian got a chance to discover life in the Midwest.

That path of discovery began in Nashville, and has proved to be a veritable roller-coast ride so far. Elian has been up and down between Nashville and Milwaukee four times in '14, operating as the 26th or 27th man on the Brewers' roster. It was his most recent return to the big club, though, that permitted him to get into the record books.

Just how did he do that? Why, by slapping out five hits in one of his rare starts. On Sunday, July 13th--"getaway day" for the All-Star break--Herrera collected four singles and a double, scoring three times and driving in two as the Brewers clomped the Cardinals, 11-2. While there are likely close to three thousand players who've had five or more hits in a game (the total at Forman et fils, going back to 1914, stands at 2220), it's still an elite accomplishment for a major leaguer.

The wrinkle for us here is not just the mundane magic of a man on the margins. It's located in the spot in the batting order in which Elian managed his feat. Herrera was batting eighth in the Brewer lineup on  7/13 when he rapped out his five hits.

So that leads us to wanting to know just what the distribution of 5+-hit games looks like when we apportion them across the batting order.

And the answers can be found in the chart at left, where you'll see what likely makes perfect sense. There is a clustering effect at the top of the batting order, with a high preponderance of these games in the #1 and #2 slots.

Makes sense, yes? These guys are the likeliest to get a fifth or sixth chance to bat in a game. So even though they may not be the best hitters on their team--and often it's not close--the vagaries of small sample size distribution and the innate structure of the game works in their favor.

The chart also tells us that Elian Herrera is one of just fourteen players to slap out five or more hits in a game since 2000 while hitting eighth in the batting order.

Of course, it's even rarer to pull this off while batting ninth (just 20 times in one hundred years). And, prior to the DH, we are talking about virtually non-existent. It has happened a total of six times since 1914. The last pitcher to slap out five hits in a game? Mel Stottlemyre, who did it on September 26, 1964.

That's nearly fifty years ago. And it's likely to be another fifty before a pitcher does it again.

But let's not let that distract us from a kindly nod in the direction of Elian Herrera, who found a way to take advantage of his second chance in the big leagues.

PASSING OF A RETROSHEET GIANT...

Just a very brief note to memorialize one of the seminal figures in baseball research whose untimely passing is a significant loss for all of us. Clem Comly was just 59, far too young to be lost both to friends (who were legion) and from the ongoing task of reconstructing the day-by-day brick-and-mortar of baseball history (at which he excelled).

Comly's years of yeoman service in Retrosheet and the Society for American Baseball Research are the type of shining legacy that deserves some kind of official commemoration. We can trust that his friends and colleagues will find a way to create an appropriate ongoing remembrance in his honor.

A more detailed memorial notice can be found at the SABR site via this link.

Monday, August 11, 2014

2014: COMPLETE GAMES #76, #77, #78, #79, #80, #81...

We've been extrapolating the 2014 complete game total for four months now, noodling around with the concept as part of a look at baseball's "endangered species" (and let's take time to bow/scrape to our "sponsor," that august but shadow-frought meta-enterpreneurial collective "Fright Quotes R Us")...

...but we hadn't taken the time to examine the underlying pattern in the recent CG data to see if delving into even smaller sample sizes could prove informative. We will do that in this entry...eventually.

But first let's note that August has started out with a bang (feel free, however, to substitute your own loud noise or representation of same here...) with 10 complete games in the first ten days of the month.

Our old pal Johnny Cueto returned to the CG column (#76, 8/5) with a five-hitter against the Indians as the Reds score a 9-2 win. It was Johnny's fifth CG of the year, and his first since 5/15.

On that same day, the Rangers' Colby Lewis went the distance (#77) with a six-hit shutout as Texas routed the White Sox, 16-0. Colby is vying for the slot on the CG list that is more than a bit unseemly: he's in the running for "pitcher having the worst season who threw a CG." And an "FQRU" to you, too, Mr. Lewis...

The Royals' August surge has been fueled in part by starting pitcher performances--with two of these being complete games. Jeremy Guthrie (#78, 8/7) was in the thick of things as the Royals cleaned up in interleague play (the hapless D-Backs and the struggling Giants), constituting a turnaround from his subpar work in recent weeks. And nominal staff ace James Shields (#81, 8/9) contributed a four-hit shutout a couple of days later.

Meanwhile, Madison Bumgarner (8/8, #80) became only the fourth pitcher in 2014 to have back-to-back CGs--but, like Jordan Zimmerman earlier in the year, his second route-going game resulted in a loss (again, to the white-hot Royals).

And finally, the A's received a plum performance from latest acquisition Jon Lester (#79, 8/7), who threw a three-hit shutout at the Twins. Lester's next start will be against those streaking Royals later this week.

So the long chase for record-breaking scarcity continues...and these ten CGs in the first ten days of August makes us wonder just how likely is it that the 2014 season will come in under the low mark (108) set in 2007. (Remember, we've removed CGs under eight IP for our count, so looking at Forman et fils will give you different numbers.)

The chart at right is an attempt to at least anatomize (if not answer) the question. Monthly totals for CGs are displayed: high totals for each month (covering years 2007-14) are shown in orange; low totals are shown in green.

When we do the arithmetic, we find that 31% of all CGs in 2007-13 occurred during the last two months of the season. If 2014 matched that average, the total of CGs for the year would come in at 103. The current extrapolation for the year, however, based on the current ratio of CG/G, works out to 112.

Looks like we may go right down to the wire...