Lance Lynn, AL Pitcher WAR Leader

In the winter of 2017, Lance Lynn was coming off a season with a solid 3.43 ERA but poor peripheral numbers and couldn’t get the multi-year deal he desired, eventually settling with the Minnesota Twins. Lynn got off to a rough start, but from May on he put up a 3.34 FIP and a 4.13 ERA with the Twins and Yankees (following a trade), with the former number making Lynn one of the top-15 pitchers in the game and the latter number befitting an average-to-slightly above-average innings-eater. Heading into 2018, Lynn was paid based on his poor FIP and not his solid ERA, but heading into 2019, Lynn received a contract based on his average ERA and not on his very good FIP. Lynn agreed to a three-year deal worth $30 million to pitch for Texas, and 15 starts into in his Rangers career, the MLB pitching WAR leaderboard looks like this:

MLB Pitching WAR Leaders
Name IP K/9 BB/9 ERA FIP WAR
Max Scherzer 106.1 12.4 1.9 2.62 2.19 4.2
Lance Lynn 93 9.9 2.3 4.16 3.00 3.2
Hyun-Jin Ryu 93 8.2 0.5 1.26 2.51 3.1
Matthew Boyd 88.2 11.4 1.7 3.35 3.00 2.8
Chris Sale 90.1 13.0 2.1 3.49 2.80 2.8
Lucas Giolito 85.1 11.0 3.1 2.74 3.08 2.8
Gerrit Cole 96.2 13.8 2.3 3.54 3.10 2.7
Jose Berrios 97.2 8.7 1.6 2.86 3.52 2.6
Frankie Montas 82 9.7 2.3 2.85 2.89 2.6
Charlie Morton 87.1 11.0 3.2 2.37 2.88 2.5
Jacob deGrom 91 11.1 2.0 3.26 3.21 2.5
Jake Odorizzi 76.1 10.0 2.8 2.24 2.99 2.4
Stephen Strasburg 96 10.8 2.3 3.75 3.27 2.4

There’s Max Scherzer at the top, and right behind him is Lynn with 3.2 wins above replacement on the season. While some might have the urge to point to Lynn’s 4.16 ERA and insist there is something wrong with WAR, particularly at FanGraphs, I would request fighting against any such urges. First, I’d like to note that over at Baseball-Reference, Lynn’s 2.7 WAR ranks 12th in all of baseball and isn’t too far off from the one above. As for that ERA, Lynn has put together an unusual season with respect to runs allowed. First, Lynn has no unearned runs on the year. While most pitchers’ earned run totals are around 90%-95% of their runs allowed, Lynn’s runs have all been earned. Indeed, Rangers pitchers outside of Lynn have earned run totals that are 93% of their total runs allowed. While it is possible that Lynn has benefited from great defense, that’s unlikely as we’ll get to below. In any event, that explains roughly 0.3 of Lynn’s higher ERA.

As for the main culprit for Lynn’s higher ERA, look no further than his .345 BABIP. For his career, Lynn has given up a .305 BABIP and his ERA and FIP are an identical 3.62. There’s little reason to think that the .345 BABIP is an accurate reflection of Lynn’s talents, but given that Lynn’s BABIP a year ago was .336, it is a topic worth exploring. Taking a quick look at Statcast tells us that Lynn’s expected BABIP this season is about 30 points higher than the results on the field. Perhaps not coincidentally, Rangers pitchers have a 34-point differential between their expected batting average on ground balls (.262) and their actual batting average on ground balls (.294). Whether it is poor shifting, poor defense, or simply bad luck, when Rangers pitchers have allowed ground balls, they have turned into outs at a much lower than expected rate.

Interestingly, Lynn is a ground-ball pitcher some of the time and but also a fly-ball pitcher depending on the situation. When the bases are empty, Lynn’s ground-ball rate is 37%. All eight of Lynn’s home runs allowed have been solo shots, and he has posted a 28% strikeout rate with the bases empty. With runners on, Lynn’s strikeout and walk rates go down slightly and his ground-ball rate goes up to 45%. While we could chalk it up to randomness, Lynn’s pitch usage is different with runners on base.

Lance Lynn Pitch Usage Percentages
Runners on Bases Empty Difference
Two-seam 25.9 15.5 10.4
Curve 10.6 10.1 0.5
Change 3 4.3 -1.3
Cutter 16 19.4 -3.4
Four-seam 44.5 50.7 -6.2
Through June 18, 2019

With runners on base, Lynn shifts his pitch mix slightly to try and induce ground balls. The strategy works and simultaneously doesn’t work. It works because he hasn’t allowed a home run with runners on, which has led to a very low 2.02 FIP so far this season, second-best in the game to Scherzer’s 1.97 mark. Lynn’s xwOBA of .263 with runners on would seem to indicate that some of the lack of homers is earned, or at least deserved. That he’s received an actual wOBA of .290 with runners on is an indicator of either bad luck or poor defense. I’m not definitive about either one, but it seems unlikely to be Lynn’s fault. There’s an argument to be made that Lynn’s aggressive approach with the bases empty (his expected BABIP is nearly identical to his BABIP in those situations) leads to more men on base, and that even neutral luck might allow those runners to score at a higher rate than his 2.99 FIP might indicate, but his approach with runners on should be leading to a lot more outs.

As for Lynn’s success generally, his pitch splits indicate almost no difference from his career numbers.

Lance Lynn Pitch Results
Career wRC+ 2019 wRC+ Career SwStrk% 2019 SwStrk%
Four-seam 98 106 10.7% 11.6%
Two-seam 124 147 6.9% 6.0%
Cutter 88 40 10.9% 11.7%
Curve 78 92 12.3% 12.7%
Through June 18, 2019

The cutter has gotten better results, and he’s throwing it harder on average this year than he has in the past, though Lynn is a pitcher who has varied his speed on his pitches so it is hard to glean too much from the harder average. What Lynn has done with the cutter is throw it a lot more often.

Lance Lynn Pitch Usage
Season FA% FC% SI% CH% CU%
2011 59.9 % 1.8 % 13.4 % 2.7 % 22.2 %
2012 49.1 % 5.3 % 21.0 % 6.7 % 17.9 %
2013 52.7 % 12.8 % 20.5 % 3.9 % 10.1 %
2014 52.8 % 10.5 % 25.8 % 2.7 % 8.3 %
2015 56.5 % 8.0 % 27.5 % 3.1 % 4.8 %
2017 38.8 % 11.8 % 42.2 % 2.6 % 4.6 %
2018 44.8 % 11.3 % 32.7 % 2.1 % 9.1 %
2019 48.4 % 17.6 % 19.5 % 4.3 % 10.3 %
Total 49.4 % 10.3 % 27.2 % 3.5 % 9.5 %
Through June 18, 2019

Lynn has dropped his sinker usage considerably, back closer to 2013 and 2014 levels, which also happened to be Lynn’s best years with the Cardinals. He pitched injured for half of 2015, then missed 2016 recovering from Tommy John surgery before that 2017 season that produced poor peripherals but a solid ERA. What Lynn has done to make himself even better is use that cutter much more often. It’s been one of his better pitches in limited use throughout his career, and replacing a bunch of sinkers, which might prevent homers but don’t do much else, with cutters has helped lead to better results, including more strikeouts. It’s been a strange trip for Lynn since the end of that 2015 season when he pitched with pain ahead of surgery, but he’s finally back to the pitcher he was five years ago. With increased use of the cutter at the expense of the sinker, he’s the best he’s ever been. He’s probably going to allow a few more homers than he has at this point, and some of them are going to happen with runners on base, but Lynn has come up with a formula that has made him one of the best pitchers in baseball for nearly half a season.





Craig Edwards can be found on twitter @craigjedwards.

38 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
sadtrombonemember
4 years ago

Just to follow up on Craig’s point: According to Statcast, Lynn is 5th in the percentage of barrels, at 2.8%–just above Syndergaard, Ryu, and Morton; and just below Soroka, Castillo, and Fried. He has gotten phenomenally unlucky, and his position at the top of WAR leaderboards is completely deserved.

BROD
4 years ago
Reply to  sadtrombone

Scherzer: 4th in “Expected wOBA” (among 108 SPs with at least 250 batters faced)
Ryu: 2nd
Lynn: 27th

sadtrombonemember
4 years ago
Reply to  BROD

That’s because the primary determinant of expected wOBA are walks (which Ryu never does) and strikeouts (which Scherzer always does). Expected wOBA partially duplicates FIP, which is what fWAR is based on (and the main point of this article).

It’s relevant to BABIP, but not as much as it seems at first glance.

Jimmy von Albademember
4 years ago
Reply to  sadtrombone

I think the fact that Lynn has 0 unearned runs and plays in what is typically an extreme hitter’s ballpark – which accounts for bWAR/RA9-WAR rating him highly as well – is far more convincing as justification than his batted ball metrics. Presumably he will find far more success in terms of run prevention if he continues to pitch like this, but you’re not going to convince fans of using run prevention for WAR that we should account for batted ball luck in his value.

sadtrombonemember
4 years ago

I’m not trying to convince them. I’m trying to convince you. Unless you’re referring to yourself, in which case, what are you doing here?

Jimmy von Albademember
4 years ago
Reply to  sadtrombone

I mean, I did say we. And it’s almost as though you can appreciate 99% of Fangraphs’ content while disagreeing with using FIP for assessing value.

sadtrombonemember
4 years ago

So you’re against trying to decompose BABIP into batted ball metrics? You can be against using FIP for assessing value and still understand the value of that.

Jimmy von Albademember
4 years ago
Reply to  sadtrombone

You absolutely can. It’s a philosophical disagreement over what “value” means. Looking at batted ball data is great for assessing what performance will be like going forward. That doesn’t mean you have to value it over run prevention for assessing value provided so far.

sadtrombonemember
4 years ago

It’s not as big a difference in philosophy as you imply. You have a belief that it’s possible to remove the effect of the defense by controlling for data we have on defense. I believe we don’t have anywhere near the data on defensive value to do that, and that most attempts to do are exercises in overfitting (which is why it fits it so badly going forward…because it did a bad job fitting it to past data).

That said, I’m cautiously optimistic about BP’s attempts with DRA, much more so than bWAR which is a hot mess.

TheGarrettCooperFanClub
4 years ago
Reply to  sadtrombone

i was looking into this the other day. his barrel and solid contact rate is the 6th best in baseball and his swing and miss rate on the four seamer and cutter are in the top 10. that combo of weak contact and swing and miss ability is what is driving this. like craig mentions, the homers look like they have to regress, but so far the numbers do suggest that he has not allowed a lot of solid contact, so maybe not