“And then I was looking at the little Chinese lady. There was a beauty to her — she was just a tiny little Chinese lady, I was staring at her because I was fascinated by her. I don’t know anybody like her, and I am SO not a little old Chinese lady.

“Then I look and I think, ‘What are her thoughts?’ That’s what I was burning inside with. ‘What is she thinking right now?’ I can never know. And my dumb brain is telling me she’s just thinking: ‘Ching chung cheeng, chung cheeng chaing.’ That’s how dumb I am, that I think Chinese jibberish that I made up is in her actually Chinese mind.”

– Louis CK

I have several mostly done baseball stat posts — a couple of them completely done — but I haven’t posted them yet. There’s a reason for this. I do think that in the next couple of days, I will put up a a pitcher rating system that I’ve developed with the help and inspiration of Tom Tango and Bill James. I will also post my 32 best players in baseball. I will be updating you on my friends awful fantasy baseball efforts too. If you want to start fantasy baseball and do better than my friend then go to the fantasy homepage here.

But the reason that I haven’t posted any of them quite yet, I think, is because it just feels to me like there’s this point I can’t quite articulate about the stat vs. gut argument in baseball. I keep feeling like I’m circling the runway. I’ve tried to write a post about why some sportswriters seem to feel unease — if not outright contempt — toward the advanced stats. I’ve tried to write a post about why some baseball managers and very smart baseball people say things that can be so easily proven false by just the quickest glance at the numbers. But, like I say, it still feels like I’m running in circles.

Saturday, doing my usual Baseball Think Factory reading, I came across Dan Steinberg and his blog about a couple of Washington Nationals radio ads, narrated by manager Jim Riggleman. The first ad is a celebration of old-time baseball (“Smart ball,” in Riggleman’s description). The second is a mild repudiation of stats against the power of humanity, not to mention a radio gala commemorating the competitive spirit of Ian Desmond.

Here’s what Riggleman says in the first commercial:

“How many ball games are won by one run? Last season: 732. A walk, a bunt, a well-placed single and a sacrifice fly. You call it small ball; I call it smart ball.”

And here’s what he says in the second:

“In baseball there’s a stat for every situation. Tie game, man on second, Ian Desmond at-bat. In day games, he’s batting .219. That’s what the stats say. Do you pinch hit for him? Absolutely not. Sometimes, you believe in the stats. Sometimes, you believe in the players.”

These commercials, for whatever reason, kind of clarified something for me that has been foggy in my mind: Baseball people really don’t get at all what people like Bill James and Tom Tango and Pete Palmer and the like are doing at all. They might THINK they know. But in the end, they are just assuming that the Chinese jibberish that they make up is what is actually happening in the minds of the most brilliant sabermetric minds.

I actually don’t think the Riggleman stuff here is all that bothersome. I like Riggleman. And he likes small ball. And he likes to stick with his players (a noble quality, I think). And he likes to go with his gut sometimes as a manager. That’s fine. I mean, no, it’s not really “fine” in that that I wouldn’t want him to be the manager of my team, but that’s because I have a particularly strong distaste for small ball (“smart ball”) and overconfidence in gut instincts. The second commercial is particularly silly along those lines … it is touching that the manager would stick with Ian Desmond in that situation because of his faith in the heart, but it should be noted that Riggleman has twice led the league in pinch-hitters used so he apparently buys into day batting averages more than most.

But it’s that .219 day-game average that stands out in the commercials — no self-respecting sabermetric thinker would ever quote a .219 day batting average. This is exactly the sort of thing people MAKE UP when talking about sabermetric thinker. This is Chinese Jibberish. You will hear people, in their mocking voice, say stuff like: “Oh, what does he hit on Thursdays after full moons during Republican administrations?” This is their terrible impressions of stat people.

Only … one of the fundamental principles of sabermetrics is the principle of sample size. If anything, smart people like Tom Tango have a MUCH LARGER sample size requirement than the average person. They believe that there is ALMOST NOTHING to be learned from a few at-bats during the day time, that Ian Desmond’s .219 batting average in 197 plate appearances during the day tells you ALMOST NOTHING.

This is one of the real ironies of stat vs. gut — the gut people often make fun of stats and yet they are the ones most likely to rely on the least telling of them, the ones with small sample sizes (day batting averages, 10-at bat matchups with pitchers, batting average with runners in scoring position) that they probably don’t mean a thing. This is not some side-thing either with sabermetrics; this is one of the founding and fundamental beliefs. BEWARE SMALL SAMPLE SIZE.*

*Advice I wish my beloved SI would have taken before putting Jeff Francoeur on the cover with “The Natural” label.

Take Tango’s invention “FIP.” FIP means “Fielding Independent Pitching” and it’s an effort to measure a pitcher’s performance based specifically on things he is responsible for — these would include strikeouts, walks and home runs.

FIP has taken a beating in the gut-based community because it doesn’t FEEL right. It reduces pitching to its skeleton shape, and we have come to know pitching as something much larger than a skeleton, something beyond just strikeouts and walks and home runs, something ethereal, something artistic, something grand. Reducing pitching to strikeouts, walks and home runs feels, in a way, like Dr. J. Evans Pritchard’s attempts to reduce poetry to a chart in “Dead Poet’s Society.” This is the sort of thing that baseball people despite. This is the sort of thing that suggests to them that stat people are trying to take the humanity and poetry out of baseball.

Two problems with that sort of thinking. One, the FIP concept is right: Pitching IS largely strikeouts and walks and home runs. People seem unable to believe it — I’m often unable to believe it — but starting pitchers have very little control of anything else. There’s a very simple statistic people call BABIP — Batting Average on Balls In Play — and it gives you exactly what it promises, the batting average of a hitter on balls hit into the field of play.

Here are a few pitchers with their career BABIP. See if anything surprises you:

– Mario Soto, .255
– Eric Show, .267
– Jim Deshaies, .271
– Nolan Ryan, .271
– Scott Elarton, .277
– Johan Santana, .278
– Bob Forsch, .278
– Dan Quisenberry, .280
– Pedro Martinez, .282
– Bob Walk, .282
– Eric Milton, .285
– Neal Heaton, .285
– Roger Clemens, .286
– Greg Maddux, .286
– Brandon Webb, .291
– Kevin Brown, .293
– Roy Halladay, .294
– Roy Oswalt, .300

Yep, Jim Deshaies gave up fewer hits on balls in play than Nolan Ryan. Bob Walk had a lower BABIP allowed than Roger Clemens. Hit a ball against Dan Quisenberry’s slow sinker and you had less of a chance for a hit than managing to hit Pedro Martinez’s electric stuff.

And those are CAREER numbers, which means we have bigger sample sizes. If you talk about individual seasons, well, forget about it. The season numbers are stunning and illogical and prove the point that success after hitting a ball in play is largely due to chance and the alignment and skill of the defense. Who was the best pitcher in baseball last year? Probably Roy Halladay. Is this because Halladay broke a lot of bats and forced a lot of easy ground balls and constantly coaxed hitters to put the ball right where the fielders were standing? Nope. Halladay’s BABIP last year was FIFTY SECOND in the National League, behind, among others, Rodrigo Lopez and Brad Bergesen.

Josh Johnson, who led the National League in ERA, had an even higher BABIP — he allowed batters to hit .301 on balls in play.

The numbers point to the simple conclusion: Pitchers don’t have much control over balls in play. It seems impossible to look at the numbers and not draw that conclusion. Even a reliever you would assume does control balls in play, Mariano Rivera, really doesn’t. In his 12 best seasons, the BABIP against him has ranged from .212 to .296. And, believe it or not, the .296 year might have been better than the .212 year.

2003: BABIP .296. 40 saves, 1.66 ERA.
1999: BABIP .212. 45 saves, 1.83 ERA.

Pitchers — especially starting pitchers — have so much less control than we want to believe. This is true in large and small senses. People used to say, with all seriousness, that pitching (meaning starting pitching) is 90% of baseball. Bill James detonated that cliche in one of his funnier essays, but the fact it became a cliche — and the fact that every now and again you will still hear it — tells you its power.

Starting pitching in 2010 is about 25% of baseball. It’s easy to figure that percentage.

Step 1: Figure that run scoring and run prevention are each 50% of the game.

Step 2: Bill James figure that pitching is about 75% of run prevention with defense the other 25%. You can adjust this if you want, but it won’t change the overall number much. Anyway that seems about right.

Step 3: Starting pitchers averaged six innings per start in 2010. You have to go back almost 25 years, to 1988, to find a year when starting pitchers averaged even 6 1/3 innings per start.

Do the math ((.50*.75) *.66) and it means that starting pitching as a whole is about 25% of the game.

I hope that people start using that as the cliche, but they won’t. We want to infuse pitchers with bigger roles and larger purpose. That’s why we assign to them wins and losses. That’s why there seems a visceral reaction to stats like FIP.

I said above that there are two problems with the sort of anti-stat emotion out there — one is that many people don’t seem to realize how logical and well thought out these baseball stats are. They tend to create Chinese jibberish and believe that’s what the stats are really saying.

The second problem is that this stuff isn’t really foreign to baseball people. Tom Tango wasn’t the first guy to attempt to separate the contribution of a pitcher and his defense. No, that attempt goes back more than 100 years.

Introducing the statistic: ERA.

Think about ERA for a moment. Why was it invented? We were already counting runs allowed for pitchers, that was easy. But Henry Chadwick — back when the game was very different — did not want to blame the pitcher for runs that were clearly the fault of the defense (and defense was much more than 25% of run prevention then). Yep, he wanted to give pitchers credit (or blame) for what they did, and defenses credit (or blame) for what they did. This led to the invention of the error and, if you think about it, the ludicrous way that we actually go back and try to reinvent history (“if he makes that play, then the runner doesn’t go to second, and the next guy doesn’t hit the single that drives him in”) and parse runs into earned and unearned categories.

How silly is this? Well, what if I tell you that at the same time the error was invented there was also an attempt to label something called the “Good play.” Here’s how the “good play” would have worked: Someone in the press box would have determined whether or not a defensive player made a good play — that would be a play made that was above and beyond ordinary effort. And if a good play saved a run, that run would be charged to pitcher as an “unscored run.” Yes, we would charge pitchers for runs that did not score.

The previous paragraph, as far as I know, is complete fiction … I just made it up. I don’t think there ever was any effort to popularize the good play. But the good play is just the opposite of the error. It’s another bizarrely simplistic and subjective attempt to separate pitching from hitting. We’ve been living under the quirky nature of ERA all of our lives, and few complain about it despite its obvious biasses and general mindlessness. We credit pitchers for every run they saved except ones when a fielder makes a mistake so obvious that someone can notice it from the press box? That’s how we do it? Really?

Really. And yet, when someone like Voros McCracken discovers through the numbers that pitchers don’t control much beyond strikeouts, walks and homers … when Tom Tango invents a stat that gives us a much clearer view of pitchers … when Bill James invents something called “the strikeout-walk win-loss record” and really shows us that you can judge pitchers fairly and quite accurately based only on those two stats … many people tend to shudder and bellyache about these stat people who care only about how left-handed pitchers do on grass fields recently mowed four days after national holidays against right-handed batters with three syllables in their names and …

Yep. Chinese jibberish.

Back to Riggleman for a moment. In the first commercial he talks about the importance of those manufactured runs in a one-run game. But even in the commercial, even on his own terms, he gets it wrong. He says: “A walk, a bunt, a well-placed single and a sacrifice fly. You call it small ball; I call it smart ball.”

First off, if you give up an out to bunt the runner over to second, you would hope — HOPE — he would score on the well-placed single. It’s like Riggleman doesn’t even get his own manufactured run scenario right.

Second, if you have runners on first and third with one out and only get one run — as the sacrifice fly suggests — you are actually doing worse than league average. Teams tend to score about 1.2 runs in those situations. You are not GAINING runs in smart ball, you are losing them.

Third, they have been playing baseball for more than 100 years. And for more than 100 years, more runs have scored with a man on first and nobody out than with a man on second and one out. This has been true EVERY SINGLE SEASON for more than 100 years. Every single one.

Not only that, but since expansion in 1969, your chance of scoring a single run is better with a runner on first and nobody out than with a runner on second and one out. Get that? Your percentages for scoring ONE RUN is better.

Now, a manager may believe that these so-called numbers are wrong, that hundreds of thousands of innings and at-bats and situations are wrong, that what is right is the manager’s own instinct for avoiding the double play and putting his RBI guy up in the right situation. I don’t begrudge a manager for thinking that or a team for believing in that manager or fans for wanting it to be true. I just wouldn’t call it smart ball.