Talking Chop

Innings pitched, strikeouts, walks, and home runs.

These four statistics are all that a pitcher can control, at least according to the FIPsters over at FanGraphs, whose Wins Above Replacement statistic is based on FIP, which in turn is based on these four peripherals. Usually, these few numbers provide for a good approximation of a pitcher's true talents; particularly when one is looking at partial seasons of data, FIP is often a more reliable measure than ERA.

So how on earth did Tim Hudson manage to put up a 2010 ERA of 2.83 despite a pedestrian FIP of 4.09? How did he manage to be consistently one of the top 10 pitchers in the National League while putting up an underwhelming 2.7 WAR (though baseball-reference.com pegged him at a much more impressive 5.4 WAR)? In short, how could a pitcher be so successful when he has average strikeout, walk, and home run rates?

The common answers from the stat-heads were that Huddy was merely having a "charmed season," that he was phenomenally lucky, or some such. I'm a stat-head, too--I use FIP and other advanced stats all the time--but that kind of dismissal of Huddy's 2010 season is, quite simply, a pile of crap.

What it all comes down to is Batting Average on Balls In Play (BABIP), the stat that removes strikeouts and home runs from batting average. One of the core tenets of sabermetrics is that a pitcher cannot control his BABIP; over time, that figure will typically stabilize at the league average (around .300). Indeed, this is true for most pitchers, but it is definitively not true for Tim Hudson.

Over his 12-season career, Huddy has allowed a BABIP of .286. That can't be explained away by luck. Nobody is lucky for nearly 2300 innings. Hudson has also posted 4 full seasons (counting 2010) with BABIPs of .265 or lower, compared to 3 full seasons with a BABIP above .300 (none of those above .309 in his rookie year). In other words, Tim Hudson knows how to induce more poor contact than the average pitcher. When he posts a low BABIP, it cannot be easily dismissed as a fluke.

For Tim Hudson, BABIP is like strikeouts, walks, or home runs for any other pitcher. Sure, some luck is involved, but skill is the overriding factor. If a pitcher sets a career high in strikeout rate or a career low in walk rate, we will praise him for improving his pitching (even though luck almost certainly contributed as well). So why not praise Hudson for setting a career-best .250 BABIP in 2010? He has demonstrated the ability to lower BABIP over his entire career, so why is it so hard to believe that he could step up his game in that department for a season?

Just as if he had exceeded his previous high strikeout rate by 1 K/9, Huddy's .250 BABIP in 2010 will likely prove to be somewhat of an outlier (his previous career best was .261). But a .250 BABIP for Tim Hudson is not the same as a .250 BABIP from, say, Derek Lowe (career BABIP: .300). Hudson's track record indicates that, when he is on his game, batters have trouble hitting the ball with authority. And he was on his game for nearly all of 2010.

The degree of Hudson's dominance is probably not sustainable in 2011 and beyond--in BABIP as with any other statistic, it is difficult to repeat a career year--but our concern here is with the past, not the future. And in 2010, the full weight of the evidence implies that Huddy was a true ace who got a little bit lucky, not an average pitcher who fluked into an ERA more than a run lower than he "deserved."

Some other fun Tim Hudson statistics from 2010:

• He led the majors in ground-ball rate at 64.1% (no other qualified pitcher was above 60%).
• He was extremely clutch. In 81 "high-leverage" plate appearances (as defined by FanGraphs), opposing hitters had 1 extra-base hit (a double by Pedro Alvarez on September 7th). One! He also induced 10 double plays in those situations, or 1 for every 6 balls in play.
• Batters swung and missed at his pitches only 6.9% of the time (a career low); he also threw the ball in the standardized strike zone only 46.3% of the time (also a career low). Combined, these numbers back up the hypothesis that Hudson's goal was to let hitters get themselves out by swinging at pitchers' pitches, rather than to blow them away.
• From May through August, his combined ERA was 2.08.
• He allowed only 2 unearned runs all year, so his ERA sure isn't being propped up by the generous decisions of the official scorer. That figure is all the more impressive considering the huge numbers of ground balls Huddy allowed; his fielders had tons of chances to botch plays but rarely did so. By comparison, Tommy Hanson gave up 11 unearned runs in 26 fewer innings.

Congratulations to Tim on his fantastic comeback season, and for providing us all with an object lesson in both the benefits and the dangers of statistical analysis.