After giving up five runs in five innings in Seattle on Saturday, Tim Lincecum‘s ERA now stands at 6.19. However, his 3.87 FIP and 3.82 xFIP stand in stark contrast to that mark, and his walk rate, strikeout rate, and home run (or fly ball) rate suggest that his process and results haven’t lined up in the season’s first two months. You can essentially trace the cause of his runs allowed to a .330 BABIP and an 11.4% HR/FB ratio, both of which are categories that have wide variation and less predictive value than more core metrics. In fact, many times, you’ll hear a pitcher who is at one extreme in either BABIP or HR/FB be referred to as being either lucky or unlucky, depending on which side of the spectrum they’re residing on at the moment.
That’s no question that FanGraphs is a contributor to those kinds of statements, as we’ve long supported the idea that pitchers should be evaluated more by their ability to throw strikes and either miss bats or get batters to hit the ball on the ground than they are by the outcomes of when a batter does make contact. We’ve used the term luck to describe fluctuations in BABIP or HR/FB, so any critiques I make about the use of that word are as introspective as they are anything else. But, that said, I do want to suggest that using the word luck to describe Tim Lincecum’s early season results is probably doing a disservice his actual performance.
When we say a pitcher has been lucky or unlucky, it usually brings up notions of a well placed pitch getting deposited just out of the range of his defender’s gloves, or a guy giving up a line drive into the gap only to have it get run down by a fantastically talented outfielder. Or, with home run rate, it’s more along the lines of allowing a weak fly ball to right field in Yankee Stadium that just keeps carrying, or throwing a meatball down the middle in San Diego and watching it get run down at the warning track. In these instances, there’s a significant disconnect between what the pitcher did and the outcome with which he was credited. In that sense, he probably was “lucky” or “unlucky” in the truest sense of the word.
But, is that really what’s happening to Tim Lincecum this season? As a team, the Giants have allowed just a .279 BABIP, so it doesn’t seem like the context in which he’s pitching has been a significant problem for his teammates. And, as has been widely discussed, San Francisco pitchers have long held some of the lowest HR/FB rates in baseball, and AT&T Park is one of the toughest places in the game for a left-handed hitter to pull the ball over the fence, so it’s not like Lincecum is routinely having to pitch in environments where he can throw a good pitch and watch it just barely creep over the fence. If the separation between his ERA and his FIP/xFIP is due to something besides poor defense or small parks, is it really bad luck?
For instance, let’s look at the two home runs he allowed to the Mariners in the first inning on Saturday. Here’s the location of the pitch he threw to Casper Wells that ended his at-bat in the first inning.
What happens when you throw a 91 MPH fastball on the inner half of the plate at belt level? This happens.
According to ESPN’s Hit Tracker Online, that ball went 390 feet and would have left every park in baseball. Lincecum threw a terrible pitch and it got hit a very long way. There wasn’t much in that scenario that you could consider bad luck. He got what he deserved for throwing a power hitter a meatball middle-in.
Now, let’s look at the same information, just from Jesus Montero two batters later.
Hey, that looks better, right? The last pitch of the at-bat is an 85 MPH slider on the inside corner towards the bottom of the strike zone. Nothing wrong with that, at least in theory. Well, here’s the pitch location (and movement, sorta, as best as you can see it from Seattle’s camera angle) in real life:
On second thought, that location wasn’t so hot after all. Yes, the pitch has some downwards movement and was lower in the zone, but it was closer to the middle of the plate than anywhere it could tie Montero up, and it essentially broke right into his wheelhouse. Montero destroyed that pitch, hitting it 445 feet into the second deck in left field, and obviously, that also would have left every ballpark in America.
If you had to grade both of these pitches on degree of difficulty for the batter to hit, you’d probably give them both an F. Maybe the slider to Montero gets a C- for being down in the zone, but given that the count was 1-2 and Montero is an aggressive hitter who will gladly chase pitches out of the zone, that’s probably not where Lincecum actually wanted to throw that pitch. In neither case would you look at the outcome and say Lincecum was “unlucky” in the way we usually think about the term. He threw bad pitches and they got whacked. That’s bad pitching more than bad luck.
And this is where DIPS theory really breaks down for a lot of people. If we tell someone that Lincecum has just been unlucky, and then they watch him throw those pitches, they’re not going to agree with you, and they probably shouldn’t. The word luck undermines the actual point of DIPS, which is not that the pitcher has no control over the outcome of a ball in play, but that variations in outcomes on balls in play don’t tell us much about what to expect from a pitcher’s future outcomes on balls in play. It’s not so much about being lucky as it is about doing something that isn’t likely to repeat in future. It’s more about repeatability or sustainability than what most people consider to be luck.
Saying that Tim Lincecum has been unlucky is probably not true. He’s struggling with his command, falling behind in counts more often, and throwing pitches that are rightfully getting crushed based on movement and location. If Wells had fouled off that fastball on Saturday, that would have been luck, so maybe you could argue that Lincecum is suffering from a lack of good luck (in that it’s quite possible that hitters aren’t missing his mistakes as often as they used to), but that’s not the same thing as suffering from bad luck.
And that’s why we should probably try to reduce our usage of the word luck to begin with. Yes, there are bloopers that fall in, broken bat squibs that find holes, or times when a defender just falls down and the pitcher gets blamed for his defensive miscue. There are definitely instances of luck in baseball, and they do effect the results that a pitcher is credited with. I’m not arguing against DIPS theory – I’m just saying that perhaps we should try to do a better job of talking about it when a guys results aren’t lining up with his process because he’s throwing bad pitches that hitters aren’t missing.
What Voros McCracken and the others who followed his research really showed us wasn’t that pitchers have no control over batted ball outcomes, but that the things that cause those gaps don’t hold up over time. Lincecum can be doing things that are causing him to give up a lot of runs now but history suggests that he won’t keep doing those things in the future. He’s either going to figure out how to fix his command or he’s going to change his approach to pitching, and he’s not going to keep locating 91 MPH fastballs middle-in at the belt with regularity. Maybe hitters will start missing his mistakes more often. Maybe he’ll start making fewer mistakes. Whatever the cause is, the effect is likely to be that Lincecum is going to get better results in the future than he has in the first two months of the season.
But that doesn’t mean he hasn’t earned his poor results to date. The word luck absolves him of blame for the outcome, which shouldn’t be what we’re trying to do. Blame Tim Lincecum for throwing terrible pitches – just realize that it doesn’t mean that he’s going to keep throwing terrible pitches in the future.