Analytics and politics

Anyone else notice how our political divide mirrors the split between baseball’s old school and the analytics crowd?

Ideological purity matters above all, with no greater sin than compromise.

It’s an intellectual construct that leads one to argue that hits are irrelevant or that a pitcher’s win total is the best way to measure their performance.

Or to deny climate change is man-made. Or claim Islamic fundamentalism is no more a threat than Buddhist extremism.


Stat geeks can be just as stubborn as Stewart

Diamondbacks GM Dave Stewart was justly ridiculed this week when he expressed hope free agent James Shields would take less money to play in Arizona for “a true baseball team vs. some of the other teams out here that are geared more toward analytics and those type of things.”

But old schoolers like Stewart aren’t the only execs reluctant to think outside their comfort zone. Stat geeks have long maintained that strikeouts don’t matter, and, despite considerable evidence to the contrary, Astros GM Jeff Luhnow — the darling of the analytics crowd — persists in constructing an offense filled with prolific whiffers.

Newly acquired Evan Gattis struck out 97 times in 106 games last year but, compared to some of his new teammates, he’s Tony Gwynn. Incumbent catcher Jason Castro struck out 151 times in 126 games in 2014. DH Chris Carter struck out more than even B.J. Upton while rookies George Springer and Jon Singleton’s were on pace to strike out a combined 466 times. Returning starters Dexter Fowler, Matt Dominguez and Robbie Grossman also eclipsed 100 K’s.

We know how excessive strikeouts stifled the Braves offense last year, but they weren’t alone. Of the 10 teams with the most strikeouts in 2014, only Washington finished with a winning record. Conversely, of the 10 teams with the fewest strikeouts, seven had winning records and four reached the playoffs.

Granted, the 2013 Braves managed to reach the postseason despite striking out 15 more times than a year ago. But good pitching almost always got the better of the Braves’ all-or-nothing offense.

In the three years that Luhnow has been GM the Astros have finished either first or second in the majors in strikeouts and are certain to maintain that trend in 2015. Obviously Luhnow doesn’t believe making contact is all that important, but those of us who endured the 2014 Braves know better.


Most stat geeks clueless when it comes to drugs

Let me start by saying I’m not close-minded to the importance of advanced metrics in evaluating talent. I’m neither old school or new school — I just hate math. Unfortunately, many of the number crunchers seem driven by a conviction that they are more evolved than those who have better things to do with their time.

Smugness is one thing. Hypocrisy is another. For all their obsessing over stats, most have little interest in facts. Like the indisputable fact that greenies and steroids are not interchangeable.

Fortunately, not all stat geeks are wed to this false narrative used by PED enablers eager to dismiss the cloud hanging over the heads of Bonds, Clemens, McGwire, et al. Larry Behrendt of the Yankee Analysts blog tried to keep them honest in a 2011 post that’s worth your time.

Drugs like amphetamines do not behave in ways that make for convenient arguments about who should (and should not) be Hall of Fame inductees.

For the moment, let’s toss the categories out the window, and look at the facts instead. With all of the facts in hand, it becomes impossible to compare amphetamine use to the use of anabolic steroids.  These two drugs are not remotely similar.  No good can be accomplished by blurring the important distinctions between these two drugs.

Anabolic steroids are drugs that mimic the effects of the male hormone testosterone.  They increase protein synthesis within cells, which helps build larger muscles.  It’s well accepted that use of anabolic steroids, in combination with adequate diet and high intensity exercise, can result in gains in muscle strength.  Whether this increased strength enhances performance in a sport like baseball is an open question – most people assume that it does, but we have no scientific proof. …

While anabolic steroids enhance performance by helping an athlete build muscle, amphetamines (sometimes called “greenies” in baseball circles) affect performance by stimulating the athlete’s central nervous system.  Amphetamines trigger increases in the user’s blood pressure, heart rate, cardiac output and breathing rate.  As a result, athletes that take amphetamines experience increased alertness and wakefulness, and decreased sensation of muscle fatigue.  Studies show that amphetamines can increase reaction time and cognitive function, and improve an athlete’s endurance (at least to the extent that the athlete is willing to work longer and harder without reporting exhaustion).

These are not opinions. But apparently Craig Calcaterra, who mocks the mere suggestion that Jack Morris is Cooperstown-worthy (for the record, he wouldn’t get my vote), knows something science does not. It’s “lunacy” to keep Bonds and Clemens out of the Hall, he writes.

Players who have either admitted to or have been credibly accused of taking (greenies) include Pete Rose, Mike Schmidt, Mickey Mantle, Ted Williams, Hank Aaron and Willie Mays. And this leaves out all of the drug and/or alcohol users who took things which hindered their performance, which also impacted the competitive nature of the game, albeit adversely to their team’s interests.

Tortured logic aside, none of the legends mentioned by Calcaterra topped his season-best home run record by 24 after his 35th birthday, as Bonds did in 2001. Greenies never caused anyone’s head to literally swell, either.

And they never helped a 34-year-old power pitcher revive his career at age 34 a la Roger Clemens, who rebounded from two mediocre years to strike out 292, a career-best, in 264 innings.

Not that it matters to humorless prigs like Keith Law, who recently tweeted:

Worthy names like Tom Glavine, who remained effective through his 40th birthday without cheating. But to Law and his acolytes, Clemens is more deserving due to a higher strikeout rate. How he achieved it is apparently irrelevant.

Good thing Keith Law doesn’t work for the Braves

Keith Law is to stat geeks what Gene Roddenberry was to Trekkies, but when it comes to evaluating young Braves he’s been anything but (wait for it) logical.

Not that he’ll admit it when he’s off. For instance, he continues to insist Evan Gattis is nothing more than Corky Miller with pop.

From his Thursday chat:

Q: Has Gattis done enough to be the full time starting catcher next year if the Braves let McCann walk in free agency?

Law: Seems like they’re planning that, but I wouldn’t. Play him every day and his inability to get on base gets exposed.

Q: Is Evan Gattis = John Buck (in terms of a career path) a fair statement, Keith? And that’s not an insult as some might take it… I mean Buck’s going to have a decently long career when all is said and done and has shown bursts of power.

Law: I don’t think that’s a crazy projection, although I’d still bet the under on that.

He’s using a small sample size to judge Gattis’ “inability to get on base.” In 74 games last year, Evan posted a .389 OBP, and in 832 minor league ABs his OBP was .374.

Law was previously way off about Mike Minor. Granted, he wasn’t alone, but Klaw treated him as if he was a poor man’s Mark Redman. Minor didn’t even place in Law’s ranking of the 60 best potential picks prior to the ’09 draft.

A back-end starter at best, he said about the Braves’ ace, who has followed up last season’s 1.154 WHIP with one south of 1.000.

Law was not as cold on Freddie Freeman, but has consistently rated him below Brandon Belt, who has a .760 OPS in 741 big league AB’s.  The gap with Freddie (.786) isn’t as wide as I assumed, and this year they are separated by .001 point. But that’s with Freddie hitting only 2 homers, a number sure to rise.

I doubt you’d find a big league GM who’d trade Freddie for Belt.

Look, talent evaluators are often wrong. But I’ll take the Braves’ track record over Keith Law’s.

Advanced metrics should sue for slander

Jerry Crasknick’s chat on today featured an increasingly common meme (sorry, I hate that word too) concerning Fan Uggla:

According to advanced metrics Dan Uggla has been an above average offensive 2B since he has been with the Braves. He is overpaid, but he still needs to start every day.

Crasnick responded appropriately:

CJ, Feel free to cite advanced metrics, but that doesn’t make it any easier to watch Dan Uggla bat .180 while trying to hit a home run in every plate appearance.

At least CJ conceded Fan was overpaid. Last December, the folks at Capitol Avenue Club wrote that “Uggla has indeed outperformed what he has been paid the first two seasons.”

And don’t give me this, “he draws a lot of walks” nonsense. Fan’s OBP since he joined the Braves is a below-average .324.


Spring stats not totally irrelevant

Would you rather him give up four homers in an outing, like Teheran did last spring against Detroit? Nothing wrong with a young pitcher gaining confidence, is there?

PED apologists violate their own liturgy

Familiar language from ESPN’s Christina Kahrl, who claims the HOF is already compromised by PEDs.

I mean, c’mon, no Mike Schmidt or Hank Aaron in the Hall of Fame? By their own admission they broke the same baseball rule on the books that Bonds did, and they did so for the same reason — to enhance their performance.

She’s talking about amphetamines, which were once doled out like Morrison’s peppermints in most, if not all, of baseball clubhouses. That doesn’t make it right, but they weren’t consumed in the shadows. Eddie Mathews wasn’t snorting lines with Hank in a toilet stall, for instance, a la Canseco injecting McGwire. Greenies didn’t give one player a significant advantage over another.

Besides, it’s ridiculous to compare the banned substances.  The proof is in the stats, yet the apologists ignore the evidence. Perhaps because it totally destroys their argument.

What else explains Bonds’ production in the twilight of his career? Bonds’ lowest OPS, in four seasons from ages 36-39, was 1.278. His highest OPS in the prime of his career, from ages 26-29: 1.136. He had 69 more homers from ages 36-39.

Fortunately, someone else crunched the numbers typically required by the statistically obsessed.

Below are the top 15 OWPs of all time, regardless of age. Before 2001, no player had reached .924, Bonds’ OWP for the whole period that covers ages 36-39. Notice how unusual it is for someone aged 36-39 to have such a great OWP. It appears that no one has aged as well as Bonds.







Barry Bonds





Barry Bonds





Barry Bonds





Mickey Mantle





Babe Ruth





Fred Dunlap





Ted Williams





Barry Bonds





Babe Ruth





Babe Ruth





Ted Williams





Babe Ruth





Ted Williams





Pete Browning





Babe Ruth




Dare I mention the freakish guns and engorged head?

Apparently none of this is sufficient proof for the likes of Kahrl, who writes of “the purported performance-enhancing benefits of PEDs.”

This from the group that sneers at those who ignore the irrefutable evidence found in the numbers.

Tonight’s reading assignment

Terrific column by’s Howard Bryant — so good he almost makes up for Rick Reilly — on this year’s Hall of Fame balloting. Much to recommend here, particularly this excerpt:

[B]ecause of the steroid era, the baseball writers are going to guess who deserves enshrinement based on who had big muscles or who had a suspicious career year. Thus, goes the thinking, the system must change. It is a disdainful mindset that doesn’t just miss the bull’s-eye, but the entire target altogether. It is the great MacGuffin of the game, and reveals a complete lack of respect for voters who for years have done the work, covered the games, and taken the privilege seriously.

The truth is that the writers are reduced to being a mop, left with cleaning up a colossal mess created by Major League Baseball and the Major League Baseball Players Association for enormous profit. The fans also must take their share of responsibility simply because professional sports franchises respond only to loss of revenue. To the people watching, steroids were always someone else’s problem, not an issue to get in the way of the fun and games — until their guy was accused or their team wronged. The journalists whose job it was to hold the institution accountable failed, too, for too little reporting allowed a corrupt culture to flourish. The emerging Generation M, influenced by its Godfather, Bill James, and his capo, Billy Beane, is also deeply culpable for allowing their calculations to blissfully ignore steroids and, through that omission, attempting to legitimize the whole dishonest era (and themselves) by attempting to make the game revolve around only numbers. It is no surprise, then, that two of the Gen M standard bearers, power and on-base percentage kings Manny Ramirez and Jason Giambi (directly linked to Beane and James) were both disgraced by steroids.

What galls me about the stat geeks, outside of the smug uniformity, is their willingness to rationalize away fraud. The game deserves better than that.

The insufferability of stat geeks

The condescension doesn’t help, nor does the fact many (not all) take themselves so very seriously. Witness this sampling of Keith Law tweets culled from the past 24 hours:

That must’ve been some game Ted Williams had back in 1947.

No, Wright had a fantastic first half. His OPS since is .735, .005 points higher than Dan Uggla.

I think that’s what they call a small sample size.

Miguel Cabrera’s .333 BA, 41 HR, 130 RBI: Irrelevant. Just like …

Only people who played the game would believe in something so stupid. If it can’t be analyzed with stats, it can’t be.

Eagerly awaiting the clever ripostes

Former Brave is ‘worst player in baseball,’ but not as bad as Jerry Royster in ’77

According to Fangraphs and Baseball-Reference’s WAR (Wins Above Replacement) calculation, Jeff Francoeur is the the worst player in the game, though he’s not the worst ever.

Jerry Royster‘s 1977 with Atlanta, a -3.7 FanGraphs and -4.1 B-R debacle. The utilityman hit .216/.278/.288 and, the metrics say, played brutal defense.

Baseball Reference ranks Frenchy’s WAR at -3.0 — the 11th worst season for an offensive player since 1901.

With a dreadful August and September, Francoeur could threaten the season both sites agree is the worst ever.


Bill James determined to become a pariah

The godfather of sabermetrics appears to be the last person not named Paterno defending the late Penn State coach (excerpted from an interview with ESPN Radio’s Doug Gottlieb):

James: [Paterno] knew less about [Sandusky] than everyone else there … [the thought that] everything revolves around [Paterno at Penn State] is total nonsense. He had very few allies. He was isolated. He was not nearly as powerful as people imagine him to have been. …

James: “they kept it quiet because they had no idea what was happening … they just thought they were dealing with a little misunderstanding …” …

James: “Paterno is one of the very few people who saw Sandusky and saw a coach who wasn’t doing a job anymore, rather than a hero … people who are responsible for it are the media. The media created this smokescreen behind which Sandusky operated, and then they’re trying to blame Paterno.”

So much for being magnaminous

The most annoying thing about the stat geeks is their arrogance, their firm belief that they’re right and everyone else is a hapless Neanderthal.

Exhibit Infinity 

If you think Venters is pretty much the same pitcher he was last year, minus luck, then be my guest. If you want to postulate that Derek Lowe was every bit as good as Huddy in 2011, that’s your prerogative. And if you choose to believe that Warren Spahn was primarily a knuckleball pitcher, well, I guess you’re entitled to your fantasy.

But don’t suggest that you know more than everyone else just because you have an algorithm that proves what should be as opposed to what is.

The statistically inclined need to realize it’s not so much the message that’s rejected but the messenger. If you really want to convince doubters, cool it with the condescension.

Otherwise you’ll be stuck in an eternal circle jerk with the Keith Laws of the world, sneering at us unenlightened folk who actually believe what we see with our own two eyes.

Tommy Hanson proves the strikeout is overrated

Today was vintage Tommy Hanson, and that’s not altogether positive. He struck out 9, bringing his K/9 ratio to nearly a strikeout an inning, roughly equal to his career mark. Yet he only pitched 5 innings.

He’s reached the 7th inning only twice this year and last season, in 22 starts, he made it into the 7th 7 times. The last time he pitched into the 8th? Sept. 27, 2010. He has 1 CG for his career.

Contrast that to Brandon Beachy, whose K/9 ratio has dropped from 10.7 in 2011 to 6.6 this season. He completed 7 innings only once last year but has done it three times already this year.

I’ll take the innings.

Baseball takes time, in more ways than one

This is in no way a slam against our sabermetric-inclined friends. I just had to chuckle after reading this paragraph recently on Capitol Avenue Club. The post is insightful, and I completely agree with it. But it just reminded me anew that it takes some work, work I have neither the time nor the inclination to do, to keep up with the latest in advanced baseball stats.

That is to those guys’ credit. We have traded barbs on this site more than once. But I’ve come to appreciate that most of the devotees of the newfangled numbers sincerely want to devise new and better ways to understand what’s happening on the field. It’s just that sometimes, in my opinion, they border on ripping the soul out of the game.

Anyway, here’s the paragraph. I of course get the innings pitched stats, but the rest is a complete mystery, and I’ve followed baseball closely for 40 years.

Last season, the Braves received 119.2 innings out of the combination of Scott Linebrink (54.1), George Sherrill (36.0), and Scott Proctor (29.1). While Sherrill was actually reasonably effective as a left-handed specialist, the other two provided very little value during their Braves’ tenures. Proctor recorded a 6.06 FIP while Linebrink sported a 4.30 mark. Combined, the three totaled -0.5 fWAR.

A question for stat enthusiasts

Was I just lucky for 21 years? (No)

And I mean this sincerely — why the reliance on BABIP? How is it so much better than ERA? Some pitchers, like Greg Maddux, pitched to contact with great effectiveness. All balls batted in play are not equal. BABIP assumes that everything is a line drive and it’s partly luck as to where they land.

How many dribblers were hit off Mad Dog and Glavine? Were they merely lucky for two decades or are there other ways to retire a batter than a strikeout?

There’s plenty of other examples. Hall of Famer Jim Palmer averaged just 5 strikeouts per 9 innings over his career — Warren Spahn, only 4.4 K/9. That’s a lot of BABIPs. Granted, Palmer had Belanger and Brooks Robinson behind him, but Maddux and Glavine had some of their best years backed by mediocre defenders like Jeff Blauser. I don’t have Johnny Logan’s UZR on me but one can assume that Spahn wasn’t backed by Gold Glovers for the duration of his 21-year career.

Perhaps I’m being too simplistic. If so, correct me, but please leave the Keith Law-esque snideness at the door.

Stop telling me Derek Lowe didn’t suck

It’s hard for me to take advanced statistical measurements seriously when sabermetricians keep insisting Derek Lowe wasn’t all that bad.

Obviously, Lowe didn’t pitch anywhere nearly well enough to earn his $15 million dollar salary. But….he wasn’t nearly as awful as you might think. Sure, his ERA was 5.05, but his FIP was 3.70, and his xFIP was 3.65. Those are damn good numbers, actually better than Jair Jurrjens‘ on the season.

I’ve had it with the persistent dismissals of JJ’s performance. When healthy he’s been a rock but the statistically obsessed continue to attribute his good work to luck. Apparently JJ’s one lucky dude.

Now know acknowledging measurements like ERA and innings pitched qualifies me as a dinosaur but these are not irrelevant numbers.

Despite making 11 more starts Lowe pitched just 35 more innings than JJ. The soon-to-be highest-paid middle reliever in baseball never saw the 8th inning this season and made it through 7 just three times. Jurrjens completed two games and pitched past the 7th on six occasions.

Altogether JJ has pitched 671 innings for the Braves — nearly 100 more than Lowe. But Lowe has allowed 12 more hits. Their BB/9 IP ratio is virtually identical. Lowe’s WHIP: 1.463. JJ’s: 1.291

Apparently I’m easily fooled.

Based on Lowe’s production as a Brave (7.8 fWAR in three seasons), he’s slightly overpaid, assuming that 1 fWAR = $4.5 million. So he’s been worth $35.1 million, and made $45 million.

So Lowe’s worth $12 million a year!?! If that’s how the numbers crunch it’s hard for me to take the number crunchers seriously.

Whitlock: Stat geeks ‘ruining sports’

Amen Brother Whitlock, one of the better sports columnists out there.

There’s a stat for nearly every action in baseball. Little is left to the imagination. Sports were never intended to be a computer program, stripped to cold, hard, indisputable, statistical facts. Sports — particularly for fans — are not science. Sports, like art, are supposed to be interpreted.

It’s difficult to interpret baseball these days. The stat geeks won’t let you argue. They quote sabermetrics and end all discussion. Is so-and-so a Hall of Famer? The sabermeticians will punch in the numbers and give you, in their mind, a definitive answer. …

The nerds are winning. They’re stealing the game from those of us who enjoy examining the gray areas of sports. We’re about 10 years away from a computer program that will write stats-based opinion pieces on sports.

How do the stat geeks respond? Typically.

If you want to read a dumb, reactionary column about how statistics have ruined sports and that people who use statistics should “STFU,” by all means, go read Jason Whitlock’s latest thing over at Fox.  Just know ahead of time that it  is aggressively stupid, profoundly lazy and provides no insight whatsoever.  Even if you hate stats and are looking for ammo in that argument, you’ll find nothing there. It says a lot about Jason Whitlock’s personal aversion to thinking hard about sports, but not much else.

But I mention it anyway because I really find myself wondering what should be done when such drivel is encountered.

(Thanks to Caz for the link)