Don’t Call It a Lincecomeback

On Tuesday night, Tim Lincecum earned his first career save, coming in on his regular throw day to record the final two outs of the Giants’ 9-6 victory over the Phillies. His familiar motion looked out of place as he warmed up mid-inning; his only previous regular-season relief appearance had come in April 2008, when a forecast for rain convinced Bruce Bochy to start Merkin Valdez instead of the promising young Lincecum, who stayed in reserve for an extended outing in the middle innings. Lincecum topped out at 98 miles per hour in that earlier appearance, his first of a season in which he would claim most of the majors’ black ink and win the first of two consecutive Cy Young Awards.

Lincecum’s most recent relief appearance prior to Tuesday, however, came less than two years ago, during San Francisco’s second victorious trip to the World Series. Lincecum pitched well that time, too, throwing 2.1 scoreless innings in the Giants’ tight Game 3 win over Detroit, but factors other than weather had kept him off the mound for first pitch. Lincecum made only one start that postseason, because after struggling in the rotation all summer, he’d gotten bumped by Barry Zito (who took three turns) and Ryan Vogelsong (who took four) and seemed to be in danger of being switched to a relief role long term.

Yet San Francisco stuck with him, out of loyalty and a lack of alternatives, even as Lincecum’s problems lingered last season and into the start of this year. Late last month, he impressed by no-hitting the Padres, which earned him plenty of attaboys and dredged up more than a few memories of a time when he often appeared to have unhittable stuff. That was as far as the post-start celebrations went, however; having floundered for a protracted period, Lincecum was far past the point where one dominant start could spark predictions of a real return to form. The 30-year-old had entered the game with a 4.90 ERA on the season and a 4.79 ERA since the start of 2012, despite having made half his starts in one of baseball’s best pitcher’s parks, and despite pitching in a league in which the average starting pitcher had posted a 3.92 mark over the same two-plus-season span. Further sapping the excitement, the no-hitter came against San Diego, one of the 10 worst-hitting teams since World War II, which had come up empty against Lincecum in another no-hitter less than a year before. One had only to look at the 4.72 ERA that Lincecum had posted between those two no-nos to understand that his June 25 achievement was no guarantee of subsequent greatness.

Shortly after Lincecum got Will Venable to ground out to end the second historic start, SB Nation’s Grant Brisbee — who’s seen enough Lincecum starts to commit each of the righty’s individually identifiable mustache hairs to memory — dismissed the predictive power of the no-hit performance with two clear-eyed lines: “This no-hitter doesn’t mean that Lincecum’s good again, that he’s going to be a solid contributor for the rest of his Giants career. It doesn’t mean anything but nine innings and no hits.”

Given Lincecum’s recent track record, Brisbee’s was the only reasonable conclusion to reach. Since that outing, however, the former two-time Cy Young winner has delivered four consecutive strong starts (albeit another one against weak-sauce San Diego). Pull back a bit, and it’s possible to paint an even rosier picture of Lincecum’s season. Since the end of his ugly April, which he finished with an ERA just under 6.00, the righty has recorded 11 quality starts in 15 attempts — a success rate that would be among the best in baseball if he’d sustained it for the full season. In 95 post-April innings, he’s held a weak assortment of opposing batters to a .194/.283/.312 line and posted a 3.03 ERA.

Adjusted for today’s offensive environment, that performance can’t come close to touching prime-period, elite-level Lincecum. Still, for almost three months, the Giants’ erstwhile ace has been an above-average starting pitcher, at least in terms of his rate of runs allowed. By the same standard, he was well below replacement level in each of the past two seasons. In that sense, Lincecum’s comeback is convincing, if far from complete.

[mlbvideo id=”34042305″ width=”500″ height=”280″ /]

The wrinkle in the revival is that by some measures, Lincecum never slipped as far as his ERAs would suggest. Even during his down years, Lincecum’s Fielding Independent Pitching (a measure of what a pitcher’s ERA should have been, assuming a league-average ability to prevent balls in play from becoming hits) remained respectable and his xFIP (a regressed version of FIP that replaces a pitcher’s actual home run total with the number of homers he would have coughed up had his fly balls cleared fences at a league-average rate) hovered more than a run below his ERA. Both metrics tend to predict future ERA more accurately than ERA itself, and sure enough, Lincecum’s ERA entering Tuesday was exactly the same as his combined xFIP over the previous two years.

There are two possible perspectives on the discrepancy between Lincecum’s abysmal rate of runs allowed and his relatively robust peripheral stats over the last two seasons. The first is that he was simply getting hit harder, and that FIP and xFIP, by treating Lincecum’s balls in play as they would any league-average pitcher’s, were failing to account for his opponents’ increased quality of contact. According to that interpretation, Lincecum, while still missing bats, was making more mistakes in the strike zone and paying the price, perhaps as a result of reduced velocity. That would have made him the opposite of longtime rotation-mate Matt Cain, who became Internet famous for keeping batters off base and in the ballpark by inducing weak contact.¹

^1.

And whose elbow issues this season have made Lincecum’s resurgence matter all the more.

The second perspective boils down to “small sample.” Human beings, confined as we are to one infinitesimal slice of the cosmic calendar, are overly inclined to hunt for insights in smaller and smaller sub-slivers, ascribing meaning to recent samples that might be too tiny to be statistically significant. As Lewis Pollis argued at Baseball Prospectus in June, even a qualified pitcher who has underperformed his FIP for two straight seasons is less likely to be a true underachiever than he is to bounce back to normal the next year. The BP database contains records of 62 expansion era pitchers who, like Lincecum, had ERAs at least 15 percent higher than their FIPs in consecutive seasons of at least 150 innings pitched. In the following season, those same pitchers (all but two of whom remained in the sample) posted ERAs that were only 4.5 percent higher than their FIPs — a difference, but one much smaller in magnitude. In support of Pollis’s point, Lincecum’s ERA is lower than his FIP thus far this season, and identical to his xFIP.

However, there’s plenty of evidence to back up the theory that hitters were smoking Lincecum’s mistakes. In 2012, his worst season, the righty allowed a career-high rate of line drives, the type of batted ball that most often lands for hits. He also allowed a career-high rate of “hard-hit balls,” according to ESPN’s TruMedia tool. Both batted-ball classifications and hit hardness are subjective, imperfect determinations made by stringers or video scouts, but they’re the best we can do without access to more sophisticated, proprietary motion-tracking technologies.

If Lincecum is allowing weaker contact this season, as his hard-hit average suggests, it’s not because he’s throwing any harder. In fact, his fastball velocity has continued a gradual decline …

… that may have contributed to his decision to use his four-seamer at a career-low rate.

Maybe mechanics can offer an answer. “Lincecum’s delivery is highly dependent on his lower half, and he is much better at coordinating his delivery (timing-wise) when he gets the 80-grade momentum that has become his trademark,” Doug Thorburn, an expert on pitching mechanics for Baseball Prospectus and coauthor of Arm Action, Arm Path, and the Perfect Pitch: Building a Million-Dollar Arm, writes in an email. “For the past few seasons, though, he has lacked the usual burst, with a slower delivery that was inconsistent, all of which pointed to physical/conditioning issues. The momentum has been back for the past couple of months, with a quicker overall pace to the plate that he can line up more consistently to find his ideal release point.”

Lincecum has made conditioning more of a priority since he packed on extra weight late in 2011, then overcorrected by dropping 30 pounds prior to 2012, which left him too drained to repeat his delivery. Last offseason, Lincecum rented a warehouse in Seattle, where he stayed in baseball shape, throwing off a mound over the winter for the first time. If Thorburn is correct, and Lincecum’s enhanced fitness has paid mechanical dividends, then the righty’s command should have also shown signs of improvement.

We can check that, thanks to COMMANDf/x data provided for this piece by Sportvision and MLBAM. COMMANDf/x, which has been active since 2010 and draws upon the same camera-based system that tracks thrown and batted balls for Sportvision’s other f/x offerings, determines the distance between the position of the catcher’s glove at the moment when the pitcher releases the ball and the pitch’s actual location, using that output as a proxy for pitcher accuracy. Its use is generally restricted to fastballs only, since the catcher’s glove isn’t always an accurate indication of where a pitcher intends to locate a breaking ball.

Here are Lincecum’s fastball totals and average distances from the target, as well as the standard deviation² of those distances and the percentage of his pitches that missed by more than 15 inches. Keep in mind that a league-average pitcher’s offerings have an average distance from the target of 13.8 inches, which means that the typical fastball misses its mark by a little more than a foot.

^2.

A quick explanation of standard deviation: Say Lincecum throws three pitches, and they miss the target by 10 inches, 11 inches, and 12 inches, respectively. Their average distance from the target is 11 inches, and the standard deviation of those distances is only one inch (because all the distances are clustered tightly around the average). Say a different pitcher throws three pitches that miss the target by one inch, 11 inches, and 21 inches, respectively. Their average distance from the target is still 11 inches, but now the standard deviation is 10 inches, because the pitcher is all over the place — sometimes he’s barely missing, and sometimes he’s missing by a lot. Lincecum gets less like the latter guy every year.

Great as he once was, Lincecum has never been blessed with pinpoint control. In 2010, when he led qualified starters in strikeout rate for the third consecutive season, his average fastball missed its target by over an inch more than the typical pitcher’s. However, his command has steadily improved over the past several seasons, to the point where he can now claim to have average fastball command. And he needs that precision, because he no longer has the speed to survive his mistakes.

We can also see some improvement in the percentage of “grooved” pitches Lincecum throws — the deliveries right down the middle that have the greatest potential to be hit hard.

Lincecum has grooved a lower percentage of each of his pitch types this season, relative to 2013. This graph gives us a good sense of why the former ace had such a hard time over the past two years. From 2007 to 2008, Lincecum grooved seven out of every 100 fastballs; in 2009, he upped that rate to eight. That’s the same number he was throwing in 2012 and 2013, with one crucial difference: Between 2011 and 2012, Lincecum’s fastball had lost a full 1.5 miles per hour. A mid-90s heater like the one he threw during his first couple of years in the big leagues has a chance to beat a batter even if it’s right over the plate; a grooved four-seamer that barely breaks 90 is a pipe shot. So far this season, Lincecum has removed two batting practice fastballs from his repertoire for every 100 four-seamers he throws, as well as a few hangers for every 100 breaking balls. Essentially, he’s serving up five fewer pitches per start that are begging to be hammered.

Instead of those meatballs in the middle of the plate, Lincecum has kept the ball down, following advice from Tim Hudson that he took to heart in February. In his post-no-hitter file, Brisbee wrote: “I was tired of watching the 0-2 doubles and the 1-2 homers.” Turns out Timmy was tired of them, too: His pitch rates in the middle of the zone with two strikes (the two right-most columns in the table below) are the lowest they’ve been in any year tracked by TruMedia. In light of Lincecum’s superior command and location, it’s a lot less surprising that he’s managed to lower his BABIP, hard-hit average, and HR/FB rate:

As long as Lincecum avoids the center of the strike zone, he seems like a safe bet to be serviceable, which is more than many analysts expected when the Giants surprised the industry by bestowing a two-year, $35 million extension on the faltering former star last October. The question is, as always, can he keep it up? We won’t know for a while, but the indicators are all positive: the steadily improving command; the realization that with diminished stuff, he has to work and train harder to win; the pursuit of “crappy contact”; and a newfound willingness to read scouting reports and exploit opponents’ weaknesses, which he picked up from Chad Gaudin by way of Jake Peavy and Greg Maddux.

“Lincecum with [Bartolo] Colon’s command would rule the world, even if he’s pitching at 90-92,” Brisbee concluded last month. “This isn’t to say that Lincecum is now Colon … It’s just a reminder that we’re not stupid to dream. It’s about looking at what this cat can do with his fastball when it goes where he wants it to.”

We’ve been waiting a while for Lincecum to learn that a power pitcher’s approach doesn’t suit someone whose arm has lost its elasticity. It’s a tale at least as old as Lefty Grove, who 75 years ago this month said, “A pitcher has time enough to get smarter after he loses his speed.” Lincecum isn’t going to get smart enough to win another Cy Young Award. For the first time in ages, though, it looks like he’ll last.