Today on Blogcritics
Home » Culture and Society » When a Six-Run Inning Isn’t the Pitcher’s Fault

When a Six-Run Inning Isn’t the Pitcher’s Fault

Please Share...Print this pageTweet about this on Twitter0Share on Facebook0Share on Google+0Pin on Pinterest0Share on TumblrShare on StumbleUpon0Share on Reddit0Email this to someone

In Saturday's game versus the Rays, Jon Lester got off to a pretty good start.  After giving up a two-run homer to Evan Longoria in the 1st, Lester regained his form and threw three scoreless innings.  Until the 5th inning, that is, when everything fell apart.  When the bleeding stopped, the Rays had scored six runs and had taken an 8-1 lead.  The funny thing is, those six runs weren't really Lester's fault.

How can a pitcher give up six runs and not take most or all of the blame?  You have to fault the defense, right?  Nope, not their fault either.  The Sox didn't commit an error in the inning, and they only made one little mistake that could be called a defensive "error".  If not the defense, then, who was responsible for those six runs?


Very few people are aware of the great role that randomness plays in baseball.  Players and sportscasters like to pretend that every hit, homer, strikeout and double play is a function of skill, or at least 95% so.  Sure, there are a few exceptions, but "luck evens out."  So says the old cliche.

Well, no, luck doesn't always even out.  Yes, the longer the season goes on, the less likely it is that someone is just lucky.  It's easier to get lucky for 20 games than 162.  But there's no magical point in the season where all luck disappears from the baseball stats.  Some players have been able to maintain their performance based mainly on luck for a whole season, in fact.

Separating luck from skill has long been an obsession of baseball performance analysts.  If a pitcher has a bad game, was it bad luck or was it a poor pitching performance?  If a hitter goes on an 0-12 streak, does it mean anything?  Is Emilio Bonifacio for real (well, okay, we know the answer to that one)?

A great leap forward in the study of randomness in baseball was made by amateur sabermetrician Voros McCracken, who proved that — statistically speaking — a pitcher has no ability to prevent hits on balls in play.

If you've never come across McCracken's theory, it could shatter your basic understanding of the game .  Don't misunderstand; McCracken isn't saying that all pitching is random.  He's saying that a pitcher's actual skills rely almost exclusively on  strikeouts, walks, and keeping the ball in the ballpark.  But if the ball is in play — that is, put into fair territory without leaving the park — the odds of a ball being a hit or an out are pretty even.  Conclusion:  A pitcher has almost no ability to limit hits on balls in play; it's almost all luck.

In the years since McCracken's discovery, some caveats have been added to his theory.  It was found that knuckleballers have more of an effect on balls in play than do other pitchers.  And pitchers do, as previously believed, have certain tendencies toward allowing more groundballs than flyballs, or vice-versa.  However, whether you give up groundballs or flyballs, you still can't control where they're hit — the rules still apply.

This knowledge was a great help in assessing which pitchers are really good and which ones are lucky.  Pitchers who have a very low ERA, despite poor walk, strikeout and home run numbers, are almost guaranteed to see their luck run out.  No one knows when; their luck may last for a season or just for a few weeks.  But it will end.  Many pitchers who "take a step forward" actually just get lucky one season, much to the chagrin of the team that drops a big free agent deal in their lap.

Another bit of baseball wisdom confirmed by McCracken's theory was that  the best pitchers are those who get strikeouts.  Limiting walks and homers helps as well.  If you want to see a list of the greatest pitchers in the league at any given moment, take a look at the strikeout leaderboard; it's the most informative simple pitching statistic out there (better than wins and even ERA).

Now a lot of people think this theory is hogwash.  They will start naming lots of people who succeeded with control and command rather than strikeouts — and Greg Maddux is usually at the top of the list.  Maddux wasn't a strikeout king, was he?  No, but he struck out more people than anyone remembers.  Maddux topped 200 strikeouts once (1998) and struck out more than 150 batters in eleven seasons.  People have a hard time accepting this argument, but if you've got the Baseball Encyclopedia with you, you'll win the discussion in no time.

Granted, you don't have to strike out 300 guys a year like Randy Johnson to be effective; but it's almost impossible to succeed as a pitcher with a below-average strikeout rate.  Look at all of the great pitchers in history; even if they didn't lead the league in K's, they almost definitely struck out more than the league average.

This new theory of pitching also helps explain some of the great fluke seasons of modern times.  Famous flukes like Greg Hibbard, Dave Fleming, Randy Jones and 1970 NL Rookie of the Year Carl Morton can now explain their flash in the pan as the cost of pitching with an unsustainably low strikeout rate.  Such problems inevitably come back to haunt you (witness the downfall of Carlos Silva).

But there is a time, as I mentioned before, when neither pitching nor defense can explain a pitcher's failure to prevent runs.  This is randomness; by that, I mean that it reflects the inconsistencies of baseball.  A time, for example, when a player does everything right but gets a negative result.  Examples include:  a screaming line drive hit right at a fielder ("atom balls"), a perfect breaking ball on the outside corner that the hitter somehow punches into the hole, any play involving a bad hop or wet grass or Astroturf, "seeing-eye singles," a lazy 320-foot fly ball that lands in the "short porch" for a home run, or the 390-foot can o' corn hit to the deepest part of the ballpark.

We could probably think of a dozen more cases where the skill of the players involved is superseded by bad (or good) luck.  True, it is rare that you'll have a lot of these happen to you in a row.  But if you play 162 games a year, you're going to see some freaky things.  All you can do is hope that the breaks go your way more often than not.

In other words, hope that you're not Jon Lester . . .

 … back to the game.  It's the top of the 5th inning, and Tampa Bay is leading Boston 2-1.  Still, Jon Lester is doing a pretty good job.  The 5th shouldn't be too difficult.  First up is Akinori Iwamura.  Iwamura takes a 2-1 fastball and grounds it up the middle for a base hit.  Next up, Dioner Navarro punches a single through the hole on the left side, moving Iwamura to second.  Shortstop Julio Lugo gets a glove on it, but just barely.  It's ruled a hit, but could have gone either way.  Still, this is not a huge crisis.  Except for falling behind both hitters, Lester hasn't really done anything wrong.  Both grounders just found holes.

The next batter, B.J. Upton, lays down a bunt.  Lester takes a little too long getting there, and Upton is called safe on a close play at first.  That's not a pitching mistake, but really a defensive mistake.

Carl Crawford comes up and sends yet another ball past a diving Lugo into left field.  Iwamura scores, making it 3-1.

Next up is Evan Longoria, who hits a towering fly ball to left field.  It bounces high off the monster and plates two, with Longoria ending up at second.

The Rays lead 5-1, but how much can we blame on Lester?  Longoria's fly was high but not very deep; by the time it hit the wall, the ball was coming almost straight down.  With another ten feet of space in front of the wall, that ball lands cleanly in Jason Bay's glove.  Similarly, if Longoria hits the same ball to right field (where the fence is much deeper), there's no way it's a double.  Rocco Baldelli catches it and the Rays settle for a sac fly (if that; Dioner Navarro doesn't have much raw footspeed at third).  Carlos Pena comes up, and Lester finally records the first out with a three-pitch strikeout.

If you've read this far in the article, you can probably guess what happens to poor Jon Lester next.  He induces another groundball (off the bat of Pat Burrell) and watches again as it rolls between the shortstop and third baseman into right field for a hit.  Crawford and Longoria score, making it 7-1.

Next, in what can only be described as an ungodly coincidence, Jason Bartlett singles through the left side of the infield, moving Burrell to second.  You can almost see the path being worn down on the infield grass by the flock of baseballs rolling by.

At this point, Terry Francona takes Lester out of the game and brings in Hunter Jones.  The first batter Jones faces is Gabe Kapler who — that's right, fans — hits a grounder through the hole into left field.  This loads the bases and proves that whatever deity Jon Lester pissed off is also mad at Hunter Jones.

Next, Aki Iwamura steps up for the second time in the inning and, as if you hadn't already guessed, hits the ball on the ground.  By some miracle, though, this one is actually hit at a fielder, Julio Lugo.  Lugo throws to second for the force out, but they can't turn the double play.  So Burrell scores on the fielder's choice, making it 8-1 (all eight runs are earned and charged to Lester).

The misery ends for the Sox when Dioner Navarro pops out to short.

Let's recap this fifth inning:

HITS:  8 (Seven on the ground, one off the Monster)

HITS ON BALLS UP IN THE ZONE:  3 (1 on a bunt, 1 off the Monster)

That's not to say that Lester was perfect.  He did leave some balls up, so there's that.  And he did fall behind to the first two hitters he faced, which is especially bad when they're the #8 and #9 hitters in the order.

But other than that, didn't Lester do exactly what a pitcher is supposed to do?  He got one strikeout and one pop-up.  He also allowed a fluke double that would have been a fly out to the left fielder in any of the 29 other big-league ballparks.  Other than that, he kept the ball on the ground.  That's exactly what you want from a pitcher in this situation.  It's what every commentator says a pitcher should do with runners on base.  He got ground ball after ground ball; he just couldn't control where the ground balls went.  He also couldn't pick the fielders behind him.  Replace Julio Lugo with Adam Everett and maybe some of those grounders don't make it to the outfield.

No losing pitcher should ever face the media and say, "It wasn't my fault!"  But you know, I really couldn't blame Lester if he did.

The larger lesson to be learned here is, I think, that we have to be very careful when we start making assumptions based on just a box score.  It's something we statheads are often accused of, but everyone is guilty of it from time to time.  Instead, start considering the randomness factor when you look at the game reports.  Just because a pitcher wins the game (capital "W") doesn't mean he pitched well; it simply means that he was on the mound when the team took the lead for good.  The same goes for the losing pitcher (capital "L"); a loss doesn't necessarily mean a bad game, it just means that he allowed what turned out to be the deciding run.  In fact, wins and losses have such a loose connection with a pitcher's actual performance that I've all but given up on them.  They're not utterly irrelevant, but there's nothing a win-loss record can tell you that another stat can't tell you with more clarity and accuracy.

Also, keep in mind that all of those warnings about luck also apply to ERA.  ERA is a much better measure of a pitcher's performance than wins or losses, but it still isn't perfect.  This is especially true this early in the season, when we're trying to judge a pitcher's skill by just eight or nine starts.

Although Voros McCracken's theory wasn't as absolute as it sounds, the basic idea is still as true as ever:  if you want to learn about a pitcher, look at walks, strikeouts and home runs.  If you look at those numbers and put them in context based on the situation, they'll tell you 80% of what you need to know about a pitcher's performance.

And here's the perfect example:

J. Lester (L):  4.1 IP, 1 BB, 6 K, 1 HR.  This tells you how well Lester pitched.  The 10 H and 8 ER, on the other hand, are a reflection of the defense and the luck as much as Lester.  Something to remember the next time you're browsing the box scores.

Powered by

About Aaron Whitehead

  • Matthew T. Sussman

    Yes, but, fans can’t boo randomness, luck, or the energy of the earth as much as they can a pitcher.