Pitching Primer: AL Postseason Preview

by Matt Trollo (MTro86) Last Updated 2017-01-11 20:07:25

Welcome to the American League Postseason Pitching Preview, which will be followed by a National League one before they start on Friday. Utilizing the same concepts as the regular season articles each day, we’re going to attempt to decipher first round post-season matchups for each league.

We did this last year too, without the benefit of being able to show the actual stats, which I’ve figured out how to do this year. Note that at the time of this writing, pricing is not yet available, so we won’t be able to get into potential value and home team will be depicted by the team with home field advantage in the series, though, obviously, both teams in the series will have home games, which will be taken into consideration.

We also realize that the importance of every single game in a short winner-take all series dictates that managers usually be more aggressive with bullpens, shortening the leashes on most starters in the postseason, though both starter and bullpen quality determine to what extent. We must therefore discuss opportunity and potential managerial tendency as well as performance here.

Lastly, not all rotations have been finalized at the time of this writing. There will be an attempt to cover all potential or likely options even if it’s just a sentence or two, but we’re not exactly sure which pitchers will be pitching in which order or park yet, though some obvious assumptions can be made.

Most stats are pulled directly from the Fangraphs.com database. If a stat is used that you are not familiar with and want to learn more about, their glossary does a terrific job of explaining all of the advanced stats used here.

Starting Pitching Main Chart

We’re using Seamheads.com three year park factors. Home team is in bold. Team Def = UZR/150. L2Yrs is a rolling calendar. Hm/Rd xFIP is since the start of the 2015 season. Opp team offensive stats are wRC+.

Toronto Blue Jays

The Blue Jays prevailed in a hard fought Wild Card game, although Buck Showalter gave them a gift in holding his best pitcher out of an extra-inning game. Looking at the Toronto rotation as a whole, you realize how deep and balanced it is even if they don’t really have a traditional number one….kind of like the Texas offense. They may have five number twos and threes though. They’re the last team to set a rotation and the most difficult to guess on because they could legitimately run these guys out in any order with little blowback against a perfectly balanced Texas lineup, who might actually be the worst of the AL offenses left. It’s a bit of a wonder how they ended up with home field advantage with what seems like a below average pitching staff and average defense as well.

Texas will not be an easy park to pitch in. Coming into the season, it and Boston had the highest run environment in the American League over the last three seasons, though both were tied for the fourth highest run environment in the league via ESPN park factors for 2016 (1.156).

John Gibbons has shown that he will stay with his starter as long as he is pitching well, as he did with Stroman through six innings. He used David Price in an unorthodox manner last season, but Price had been shaky with a history of post-season difficulties. The bullpen overall is middling (4.11 ERA, 3.98 xFIP, 3.8 fWAR), but could be a problem if closer Roberto Osuna, who had to leave last night’s game, is dealing with some sort of injury.

Aaron Sanchez had just league average strikeout and walk rates, but the fifth highest GB rate (54.4%) in the majors. His strikeout rate dropped below 17% in July and August, but moved up over 20% the last month of the season, though his GB rate dropped to 40% with a hard hit rate above 35% over the last two months. There were frequent talks of moving him to the pen to conserve innings, but they needed him down the stretch. His control suffered too with an 11.4 BB% in September. Frankly, the Blue Jays would probably prefer the ground balls if pitching in Texas, which is not optimal for daily fantasy players, but I’d suspect but I’d suspect he’d either get one game in Toronto or be used as a potential super-reliever if they’re still concerned about his workload. You could also argue he might be their second best pitcher behind Stroman.

J.A. Happ has not been named the Game One starter as of this writing, but I’d suspect that’s where they’re going. His ERA was well below his estimators, as illustrated in his last start of the season, when he walked five Boston batters with just one strikeout, but managed to allow just two runs over 6.1 innings. In fact, he struck out three or fewer in five of his last seven starts and would be a dangerous choice for two games in this series, as Toronto just can’t expect his 79.7 LOB% to hold up. It looked like he was having a second half resurgence for a while with frequent employment of a fastball that was generating swings and misses, but that did not hold up. He ended with similarly league average strikeout and walk rates to Sanchez, but with a much lower ground ball rate and a league average rate of hard contact. He’s been generally over-priced by both daily fantasy sites for most of the season.

Marco Estrada is an extreme fly ball pitcher (0.69 GB/FB), who probably shouldn’t be pitching in Texas, though it’s really been more run friendly than power extreme the last several years. He actually led the staff with a 22.8 K% this season, but his 9.0 BB% was also highest of all starters (including Liriano since joining the team). He has a 2.93 ERA in the first half of the season, but with a 4.11 FIP due to a .193 BABIP and 80.3 LOB%. Some may blame his second half decline (4.27 ERA, 4.19 FIP) on a back injury, but the team claims it was a non-issue and there really was no decline if you believe his estimators, just expected regression with a .285 BABIP and 71.4 LOB%. His hard hit rate actually dropped five points (33.7% to 28.2%) in the second half. He too, is often over-priced in a daily fantasy sense.

Francisco Liriano was originally omitted from this list originally because the dummy writing it saw four well regarded starts, who each made 29 or more starts for the Blue Jays and came to the quick conclusion that those were obviously the options, forgetting the pitcher that may have done some of the best work on the staff down the stretch since coming over at the deadline. In 47.1 innings, including eight starts, Liriano had a 16.7 K-BB% (which would have led the rotation over a full season) and although his hard hit rate (33.6%) did not decline, he did keep the ball on the ground 52.7% of the time. While there are really no safe options in such a dangerous park and the highest hard hit rate on the staff doesn’t make him any more appealing, the combination of ground balls and strikeouts should probably make him a serious consideration to start one of the first two games. After all, it’s not like the Jays have a strong control and weak contact guy in the rotation.

Marcus Stroman is their best overall pitcher. He led the majors with a 60.1 GB% and struck out has a 22.7 K% in the second half of the season. Ideally, he’d be starting this series and pitching two games in Texas if needed, but was needed in the Wild Card game to get here. He and Liriano would probably be the most attractive in daily fantasy sense as they have been generally underpriced due to second half resurgences in their peripherals that their price tags have yet to catch up to.

Having completely forgotten about Liriano until half-way through this writing, I now believe the Jays may deploy Sanchez out of the bullpen, which would make him a weapon (especially if Osuna is out), but one that can come in and throw multiple innings, giving the offense a completely different look from most of the starters. It may also shorten the leash for several of these pitchers.

Late Edit – Estrada has now been named as the Game One starter, which is unfortunate for daily fantasy players (from a pitching standpoint) and potentially the Blue Jays as well. All things being equal, I might call him the fifth best starter on this team right now.

Texas Rangers

The Rangers might have been rooting for the Orioles Tuesday night or at least they probably should have been because the Blue Jays have a double digit walk rate against LHP (though surprisingly, a higher HR rate against RHP – 15.0 HR/FB). Two potential starters are left-handed with walk and/or hard contact issues, while two of the three right-handed options have HR issues of their own. They would have matched up much more strongly against a Baltimore offense that struggled greatly with LHP.

The Texas bullpen was ranked 24th by fWAR (1.7) with a 4.40 ERA and 4.38 xFIP this season, but were fifth best (2.7 fWAR) with a 3.50 ERA and 3.51 xFIP in the second half of the season, led by the Bush, Claudio and Dyson monster, though the closer was actually the worst of the trio by peripherals over that span (9.0 K-BB%).

Jeff Bannister would probably like to ride his top starters as long as possible, though the inconsistency of Hamels may not allow him to do so. Ranger fans can hope that stubborn faith in his arm won’t hurt them in this series, but they really don’t have that extra arm like Sanchez, who may have the capability of coming in and throwing multiple innings of dominant relief. Beyond the top two, the other four pitchers probably shouldn’t be starting in a post-season series and will likely have very short leashes in Games Three and Four.

Yu Darvish needs to be starting the first game of this series and pitch twice if necessary. He started his season strong after missing over a year and a half, did have some issues with run prevention late August to mid-September, but finished with at least eight strikeouts in each of his last four starts (38.4 K%) and struck out nine or more in eight of his 17 starts overall. From a daily fantasy standpoint, he’s the highest upside arm in the ALDS. Though Darvish did go at least six innings in 11 of his last 13 starts, Hamels could be considered more durable and more easily able to come back on three days rest for Game Four as the Rangers may best be served going with a three man rotation only because they can’t possibly just go with two. If Darvish started Game Two, he’d be able to return on normal rest for Game Five with two off days. (Hamels has since been named Game One starter.)

Cole Hamels can succeed and even thrive against any lineup on a good day, but those have occurred less often than his ERA would indicate this year. His strikeout and swinging strike rates are right on career averages, but a 9.1 BB% that’s a 50% increase over his career rate and the highest he’s ever allowed was masked by an 83.2 LOB% in the first half of the season. In the second half, he was still inconsistent, but did improve nearly five points to a 15.4 K-BB%, allowing his ERA to remain below three and a half despite the strand rate dropping 73.7%. A drop in HRs may have played a more significant part, allowing just eight of 24 in the second half of the season. Hamels walked at least three in 15 of 32 starts and Toronto was fourth against LHP with a 10.2 BB%, though their 13.0 HR/FB against southpaws was surprisingly average. While it will be difficult to endorse faith in Hamels in this series, he certainly has the upside to be considered a strong contrarian play depending on cost. Though, the problem might be that he could end up the chalk as few people may really understand how much of his success was dependent on a high strand rate for much of the season.

Neither Martin Perez (RHBs .341 wOBA, 2.0 K-BB% this season, .337 wOBA, 4.0 K-BB% career) nor Derek Holland (RHBs .346 wOBA, 34.7 Hard% this season, .338 wOBA, 32.1 Hard% career) should sniff a significant workload in this series.

That creates an interesting scenario in this series with “interesting” not necessarily meaning good for the Rangers. As mentioned, the Blue Jays had a higher HR rate with more hard contact against RHP. A.J. Griffin allowed 28 HRs in 119 innings. Only Chris Young (28) and Taijuan Walker (27) came close to the same amount of damage in under 150 innings this year. He allowed multiple HRs in eight of his last 10 starts. However, I believe he should probably get the Game Three nod because RHBs had just a .306 wOBA with 10 HRs against him (still not great with a 0.68 GB/FB and 12.2 HR/FB), while LHBs had a .406 wOBA and 18 HRs (21.4 HR/FB, 44.4 Hard%).

The other option might be Colby Lewis, who allowed six HRs in 18.1 innings returning from a two and a half month layoff in September. He’s another extreme fly ball and occasionally HR prone pitcher, but the difference is that with little platoon split, RHBs hit him just as well as LHBs, which could be a major drawback here.

Despite ending the season with an American League high 95 wins, Yu Darvish may be the only usable pitcher on this staff from a daily fantasy standpoint, but the good news is he may also be the top overall arm still available in the American League.

Boston Red Sox

Cleveland may have home field advantage in this series, but the Red Sox are probably the best team in the American League at this point, especially with the Tribe down two of their top three starters and the health of Corey Kluber in question.

Oddly, both teams are starting their star pitchers in Game Two, though it might make more sense if you believe Porcello’s 22 wins and Price’s post-season troubles are predictive of future performance (I don’t). For Cleveland it’s understandable if Kluber needs an extra day of rest, but also very concerning.
While Boston is tied with Texas for the highest run environment in the AL over the three years before this one, Cleveland (1.207) and Boston (1.199) were the two most run friendly environments in the American League this season according to ESPN. One of those is a surprise.

These were also the top two home offenses in baseball (Boston 121 wRC+, Cleveland 120), which may have had something to do with those park factors. While Boston dropped just 16 points of wRC+ on the road, the Indians dropped 35 points, giving Boston a significant advantage for their two home games and probably the pitching advantage in Game One with Porcello against Bauer. Cleveland was a bit more vulnerable to LHP, mostly with significantly less power.

Boston had the ninth best bullpen by fWAR (4.8) with a 3.56 ERA and 3.67 xFIP, but the third fewest innings thrown this season (Blue Jays bullpen threw the least). While Kimbrel did strike out 37.7% of batters, he was more human this year with a 13.6 BB% and Boston really doesn’t have an otherwise dominant arm out of the bullpen, but they will should have Drew Pomeranz, who would definitely be a weapon against this offense for several innings at a pop.

Rick Porcello won a major league high 22 games this season. Or rather, his team won 22 games that he started and left with the lead after at least five innings. This was certainly the best season of his career with career best strikeout (21.2%) and walk (3.6%) rates, though he also generated a career low 43.1% ground ball rate with a nearly league average rate of hard contact (30%). He’s also thrown a career high number of innings by 20 (223), but has shown no ill effects does the stretch. Prior to his last two starts, he had gone at least 7 innings in 11 straight starts and I have no doubt that his manager is looking for that many in each post-season start. The problem, from a daily fantasy standpoint is that the Wins are basically priced in at this point and a RHP facing the Tribe in Cleveland might be the most difficult spot on the board in Game One.

The most curious aspect on this side of this matchup might be how John Farrell decides to handle David Price. Porcello may get a longer leash than the performance has earned (though he has been good) and you would think Price has earned that as well, but how will the perception of his post-season issues affect his workload in this series. Considering that the third and fourth starters are shakier with some bullpen questions, my thought would be that Price is probably the best arm they have and they spent over $200 million on him this off-season not to treat him cautiously around this time of year. That Price was able to finish with a 3.99 ERA says a lot after an incredibly rough start. Both his peripherals and traditional results fluctuated throughout the season, finally putting it all together some time in August, finishing with a 3.43 ERA and 18.5 K-BB% over his last nine starts, allowing more than three runs just twice, though he did allow eight of 30 HRs over his last five starts this season. Cleveland has shown much less power (10.0 HR/FB) vs LHP. He’s probably a better choice than Porcello in Cleveland, a park that favors LHBs more than RHBs, should it come to a Game Five.

Clay Buchholz getting the call in Game Three is a bit of a surprise considering an ERA and estimators around five this season. He was moved back into the rotation in September, allowing two runs or less in four of five starts, but with a 9.3 K-BB% and 36.1 Hard%. His leash will absolutely be the shortest of the starters with Pomeranz likely in waiting should he run into trouble early. If there is a reason for optimism here, Buchholz was in line with his league average career SwStr% this season, despite a significantly lower K%. He’s a tough and highly risky sell against a good offense in a dangerous park.

Eduardo Rodriguez is a highly regarded prospect, who finally caught fire down the stretch, striking out 29 of his last 69 batters (42%), though he did walk five Blue Jays in his last start. He did have a 6.02 ERA and 5.56 xFIP at home this season with 10 of his 16 HRs allowed coming to RHBs at Fenway. Roster Mike Napoli that day. He did not allow a HR to Toronto in that last start at home, in which he walked five though. Depending on cost, this might be an arm players might look to for low cost upside against a low power offense vs LHP, though the leash will likely be shorter if Boston trails in the series at that point.

While we can be confident that the Red Sox probably want Porcello to go deep, Price is in the best spot to get the job done here and Rodriguez may have more upside than some realize. If players are buying into the Price big game narrative, all the better for us potentially getting him at lower ownership levels on a four game slate on Friday with Kershaw, Scherzer, Darvish, Kluber and Lester going as well.

Cleveland Indians

The Cleveland staff could be the AL favorite at full strength, but with just Kluber left and even him ailing against the top offense in the majors, they may be an afterthought. None of the other starters can be trusted. Bauer is erratic and showing less upside recently, Tomlin is a HR machine and…well….Clevinger is listed, but I can’t actually see them giving him a start, even as untrustworthy as Bauer has been. It’s just really sad that Carrasco and Salazar are not available here.

The good news is that the Cleveland bullpen finished the second half with 3.3 fWAR (third), a 3.35 ERA and 3.40 FIP. Andrew Miller was the star, but this bullpen had five guys throw more than 20 innings in the second half with a FIP lower than 3.05 and an ERA below 2.35. Terry Francona should and probably will be going to his bullpen early and often in this series. Andrew Miller may pitch more innings than one or two of the starters. Unfortunately, this makes none of their starters trustworthy from a daily fantasy standpoint.

Trevor Bauer struck out 32 of 117 batters faced in a series of four starts in June. Since then, he has just a 9.9 K-BB% over 17 starts, striking out more than five just five times, though three of those occasions have come in his last four starts. There’s nothing otherwise interesting about his profile. He’s a league average pitcher facing the best offense in baseball. Occasionally, he’s a bit better than that, but often a bit worse too.

Corey Kluber would be the one interesting arm here, but a quad injury limited his workload over the last couple of weeks and when you add that risk in facing this offense against the other super arms scheduled to go on his day on Friday, the reward just doesn’t seem worth the risk.

Josh Tomlin had the lowest walk rate in baseball (2.8%) by almost a full point. That allowed him a 13.5 K-BB% with just a 16.3 K%. He was also one of just three pitchers to allow more than 35 HRs this year. The other two were Jered Weaver and James Shields. If you have to start him against the Red Sox, perhaps it’s better off in Boston, a park that suppresses LH power more than most people realize. RHBs do have a wOBA 30 points higher for his career while his 119 HRs allowed have been split almost dead evenly.

Mike Clevinger can miss some bats, but had a 12.5 BB% this year.

With Kluber potentially not 100% and down two All-Star quality starters, none of these arms may provide enough daily fantasy value to consider rostering them against the best offense in baseball in highly positive run environments. Additionally, potentially the best bullpen remaining in AL contention should shorten the game for most of these arms.

K/BB Chart

Pitcher and Opponent K% & BB% for titled splits, similar to the Main Chart.

K/BB Chart – Opponent

Batted Ball Chart

Pitcher and Opponent Batted Ball stats.

Batted Ball Charts – Opponent

K/SwStr Chart (2016 LG AVG – 20.1 K% – 9.5 SwStr% – 2.12 K/SwStr)

Getting called strikeouts can be a skill, but it’s usually not a sustainable one at a large deviation from the league rate (catcher framing and other factors may make some difference here). K% correlates heavily with SwStr% though. Look for a large difference and you might find a potential adjustment before anyone else.

While players should have no interest in one of the two pitchers who under-cut their SwStr% this season, Clay Buchholz might be slightly interesting in that he’s retained a league average swinging strike rate, almost right on his career number, despite the drop in overall strikeout rate.

ERA Estimators Chart (2016 LG AVG – 4.35 ERA – 4.31 SIERA – 4.25 xFIP – 4.31 FIP)

How a pitcher’s ERA matches up against his defense independent estimators.

To see them all listed here, it’s quite interesting how all of the Toronto starters were far separated from their estimators. I’d love to see someone smarter than me look into something like that, but I doubt any Fangraphs writers are reading. The Toronto defense wasn’t exceptional by many metrics, but were very good in terms of BABIP suppression. Estrada’s extreme fly ball profile may have helped, but Stroman and Sanchez both had top five ground ball rates in the majors, neither really limiting hard contact all that much.

BABIP Chart (2016 LG AVG – .298 BABIP – 20.9 LD% – 9.5 IFFB% – 87.3 Z-Contact%)

A few years back, both Dan Rosencheck and Steve Staude separately found that high Infield Fly Ball (IFFB) rates and low Zone Contact (Z-Contact) rates correlated well with lower BABIP for pitchers. I won’t pretend to know how much of the variation in BABIP can be explained by these factors, but since they seem to have some effect, here they are. See if you can use it to your advantage.

It’s presented as the difference between team and pitcher BABIP allowed because team defense can explain a lot of the variance from league average on its own. A pitcher with a much lower BABIP than his team allows is a red flag absent further supporting evidence, while a pitcher with a much higher BABIP than his team allows may have something to offer in the future, especially with the right indicators.

You can find me on twitter @FreelanceBBall for any questions, comments, or insults.

NOTE: Button for pitcher salary chart above opens in popup window