Lincecum is best option for Giants in NLCS Game 4

Tim Lincecum has shown flashes of his old self during three relief outings this postseason. (MCT/Landov) The Giants drew even with the Cardinals in the

Tim Lincecum has shown flashes of his old self during three relief outings this postseason. (MCT/Landov)

The Giants drew even with the Cardinals in the National League Championship Series on Monday night, and as the series shifts to St. Louis, manager Bruce Bochy is considering restoring Tim Lincecum to the rotation to start Game 4. On the heels of a season in which he put up a 5.18 ERA, the league's highest among qualifiers, Lincecum was bypassed in favor of Barry Zito for the Game 4 start in the Division Series against the Reds. The choice between the two former Cy Young winners — both a long way from their award-winning form — was rationalized by the latter's late-season run; the Giants won Zito's final 11 starts, and he allowed three runs or fewer in his final five, taking the win in each.

Alas, Zito lasted just 2 2/3 innings in his start, and the Giants needed 4 1/3 innings of strong backup work from Lincecum in order to hold on to their early lead. Lincecum has thrown two other two-inning relief stints as well, one in each series, and overall he has allowed just one run in 8 1/3 innings, allowing three hits, walking one and striking out nine.

Starting Linceum is the right move, in part because the Cardinals tee off on lefty pitching. As I noted several times during the Division Series, four of their five 20-homer hitters — Allen Craig, David Freese, Matt Holliday and Yadier Molina — are righthanded, while Carlos Beltran is a switch-hitter. Craig, Holliday and Molina each slugged above .600 with OPSes over 1.000 against southpaws this season, but none of those righties exceeded Craig's .482 slugging percentage or Molina's .833 OPS against righties. Furthermore, the team has shown a significant platoon split both in the regular season and in the postseason:

Split	AVG vs LHP	OBP	SLG	AVG vs RHP	OBP	SLG	OPS Dif
Regular	.276	.338	.450	.270	.338	.410	+.040
Post	.253	.353	.455	.229	.309	.392	+.107

Within the small sample of seven postseason games, the gap between the Cardinals' performance against lefties and righties is almost three times what it was in the regular season in terms of points of OPS. Furthermore, Zito showed a huge platoon split  this year (.209/.259/.299 against lefties, .281/.355/.468 against righties), and a more typical split in the three years before that (106 points of OPS higher against righties). Lincecum actually showed a reverse platoon split this year (.232/.330/.392 against lefties, .282/.352/.462 against righties), which does rate as a concern, but in the three years prior he was basically platoon-neutral (two points of OPS higher against lefties).

Not that Lincecum is exactly an ideal choice against an offensive juggernaut given that he posted career-worsts in home run rate (1.1 per nine, up from 0.6 from 2007-2011) and walk rate (4.4 per nine, up from 3.3 from 2007-2011) this year. His strikeout rate was still a respectable 9.2 per nine, and even expressed as a percentage of plate appearances (reflecting an additional number of batters faced that he wasn't retiring), it wasn't a huge step down, 23.0 percent compared to 26.5 percent prior. But when combined with the upticks in his walk rate and batting average on balls in play (.316, up from .298 prior), his decreased ability to miss bats did have an impact.

What went wrong for Lincecum in 2012? Some have blamed struggles with his unorthodox mechanics following an offseason in which he lost 30 pounds. Pitching coach Dave Righetti told USA Today back in June, "His whole body changed, and that weight can affect your mechanics as well. I don't give a damn what his weight is, but I know what too big is and I know what too little is. He showed up too little." Lincecum said he lost the weight because his knees were sore, and he threw very little over the winter.

The difference was noticeable. According to the PITCHf/x data at BrooksBaseball.net, his average fastball velocity of 91.0 MPH was down 2.1 MPH from last year and 3.7 percent from his 2008 Cy Young campaign, when his velocity was at its peak. His other pitches lost zip as well, costing him movement, and while he actually got around the same rate of swings and misses as last year, his breaking stuff was creamed; in 2011, his slider and curve were both hit for home runs 0.4 percent of the time, the slider left the yard 1.0 percent of the time this year, the curve 1.2 percent.

Others blamed Lincecum's woes on his mental approach. In May, his father, who designed his unorthodox delivery, told the San Jose Mercury-News that his son didn't have a lot of confidence, and that he needed to be more aggressive by pitching inside. In June, Lincecum told USA Today, "It used to be, two runners on and no outs, and it was like, 'I'll get a double play and strike out the next guy.' That was the arrogance I had, knowing I could dig deep and get out of it. Now I give up a couple of walks, and it's like, 'How am I going to get out of this? Is this the inning they get me?' That doubt creeps into your mind. It's almost like you're waiting for something bad to happen."

Whether his woes were physical or mental, Lincecum did improve somewhat as the season went on, but the change wasn't as dramatic as his won-loss and ERA splits by half (3-10, 6.42 before the All-Star break, 7-5, 3.83 after) would have you believe:

Split	IP/GS	HR/9	BB/9	SO/9	ERA	FIP	BABIP
1st	5.4	1.0	4.7	9.7	6.42	2.94	.338
2nd	6.0	1.2	4.0	8.7	3.83	3.16	.291

Lincecum's home run and strikeout rates actually moved in the wrong direction after the All-Star break, and much of the difference in his performance owed to a 47-point drop in BABIP; his FIP — an estimate of his ERA based upon his peripherals — was actually higher in the second half. Perhaps the BABIP drop that helped lower his ERA was simply luck, perhaps that reflected a better ability to place the ball where he needed to place it with more consistency thanks to improved mechanics; the answer probably lies somewhere in between.