The Upcoming Mike Leake Mistake

With the Winter Meetings officially in high gear, some early themes and trends of the 2015-16 offseason are establishing themselves. Foremost among them is the substantial investment of years and mega-millions in the legion of starting pitchers on the free agent market. Sure, the game is flush with cash — and, certainly, there is quite a bit of high- to middle-end pitching talent on the market — but the first few officially consummated deals are coming in above projections in terms of years and/or dollars.

Zack Greinke, $34.3 million per year through his age-37 season? David Price, $31 million per year through his age-36 season? Hisashi Iwakuma, who has exceeded 180 innings pitched exactly once in his four seasons in the U.S., and who also endured nagging shoulder injuries during his time in Japan, guaranteed $15 million per year through his age-37 season?

John Lackey and Jeff Samardzija came in a bit closer to year/dollar projections, but they both bring their share of risk to the table. Another half-dozen starters are likely to put pen to paper in the next few days, and it’s a pretty safe bet that most of the contracts aren’t likely to end well.

Mike Leake, who just completed his age-27 season, could be one of those. He’s been connected to a handful of clubs, and is often mentioned as a consolation prize for the clubs missing out on the Greinkes, Prices and Johnny Cuetos. Though his age and durability would appear to be significant plusses, he seemingly oozes league-averageness. How much, and for how long, are clubs going to be willing to guarantee a pitcher with a round career ERA- of 100?

I was in my first season as a member of the Seattle Mariners front office during Mike Leake’s draft year. The draft had a clear first overall pick in Stephen Strasburg and, most felt, a clear second overall pick in Dustin Ackley. There was another group of a handful of players, including Leake, in the draft’s next tier. I saw him pitch a couple of times at Arizona State during major league spring training, and really liked what I saw.

He lacked the top-of-the-scale fastball velocity and imposing physical stature that you look for at the top of the draft, but he had just about everything else. He pitched above his stuff, commanding four pitches including a plus slider, and had enough changeup to handle opposite-handed hitters. His athleticism was apparent; he could really hit and field his position, and had a knack for stepping it up in the late innings of close games. Leake might sit 89-91 mph with his fastball for six innings, and then pull 93 out of his back pocket to get a big out late. The performance numbers were great, the scouts spoke glowingly of his makeup. If anything happened to Ackley on the way to draft day, Leake would have been a perfectly serviceable backup plan.

When you’re drafting a player, you’re only drafting his six controllable MLB seasons — if he’s lucky enough to get to the majors in the first place, let alone last that long. Leake leapt straight to the majors in his first pro season, logged only 7.1 innings in the minors overall, and has recorded a 9.7 WAR in his career going into free agency. That translates to a solid, if unspectacular #3 overall pick in 2009 by the Reds. Leake ate up innings, qualified for five consecutive ERA titles from 2011 to -15, and produced annual WAR totals between 1.4 and 2.3 over that span.

Clubs often need to be reminded that they are paying players for what they are going to do in free agency, rather than what they have already done. It’s pretty rare for players to hit free agency with the sort of youth that certain members of this offseason’s free-agent crop — including Leake, Jason Heyward and Justin Upton — still possess. Does this make Leake more attractive than his free agent pitching compatriots? Or are there mitigating factors that severely limit his upside and value? Let’s take a look his 2015 plate appearance frequency outcome and relative production allowed by ball-in-play (BIP) type data to get a better feel. First, the frequency information:

Plate Appearance Outcome Frequency, 2015

Metric	%	REL	PCT
K	15.3%	76	19
BB	6.3%	85	34
POP	2.2%	75	36
FLY	24.4%	81	17
LD	21.6%	102	54
GB	51.8%	113	80

There are clear plusses and minuses in Leake’s frequency profile. On one hand, he is a ground-ball pitcher who doesn’t walk many batters. His ground-ball percentile rank of 80 is quite good, and exactly matches his average grounder percentile rank from 2010 to -15. He’s never been lower than 64 or higher than 91 in that department. Leake’s walk rate percentile rank of 34 is actually his highest since his rookie season, when he posted a 60 mark.

On the negative side, his strikeout rate is his most obvious limiting factor. In six major league seasons, Leake has never posted an average or better strikeout rate percentile rank, and his 2015 mark of 19 was sharply down from 40 in 2014, and just above his career low of 16 posted the previous season. Mike Leake allows an awful lot of contact.

One more subtle flaw in Leake’s profile only becomes apparent when you take a multi-year view at the data. His 2015 liner rate percentile rank of 54 doesn’t exactly jump out, for a couple of reasons. First, it’s barely above league average, and second, liner rates are relatively random, fluctuating much more from year to year than those of other BIP types. There are, however, pitchers who possess a true talent either of yielding or preventing line drives. Leake, unfortunately, appears to have a knack for yielding them, as 2015 was the fourth straight season in which he has allowed a league average or higher liner rate, including marks of 85 and 91 in 2012 and 2013, respectively.

So we have a pitch-to-contact guy who allows a bunch of grounders (good thing) and liners (bad thing). The important piece of information yet to be evaluated is the authority of the contact allowed by Leake. To do so, let’s take a look at his relative production allowed by BIP type data:

Relative Production Allowed by BIP Type, 2015

Metric	AVG	OBP	SLG	REL PRD	ADJ PRD	ACT ERA	CALC ERA	FIP	TRU ERA
FLY	0.183		0.704	140	198
LD	0.613		1.022	94	106
FLY + LD	0.466		0.913	116	137
GB	0.172		0.181	43	112
ALL BIP	0.287		0.465	83	111
ALL PA	0.240	0.288	0.388	86	112	3.70	3.42	4.20	4.44

The actual production allowed on each BIP type is indicated in the batting average (AVG) and slugging (SLG) columns, and is converted to run values and compared to MLB average in the REL PRD (or Unadjusted Contact Score) column. That figure is then adjusted for context, such as home park, team defense, luck, etc., in the ADJ PRD (or Adjusted Contact Score) column. For the purposes of this exercise, sacrifice hits (SH) and flies (SF) are included as outs and hit by pitchers (HBP) are excluded from the on-base percentage (OBP) calculation. One quick note here: I have presented this type of analysis many times, but only recently have I begun to show fly ball and line drive line items both separately and combined.

The most important factor to note here is that, across the board, Leake’s Unadjusted Contact Scores are substantially lower than his Adjusted Contact Scores. This means that, on all BIP types, the actual damage allowed was less than the damage he “should have” allowed based on BIP authority. His actual combined FLY/LD Unadjusted Contact Score of 116 is bad enough; only seven NL ERA qualifiers posted worse marks. That 137 FLY/LD Adjusted Contact Score, however, was the worst among NL ERA qualifiers, by a considerable margin.

The only saving grace, to this point, is that Leake doesn’t allow many fly balls, and his relatively strong batted-ball mix keyed by his high grounder rate conceivably could carry the day despite the relatively loud nature of his FLY/LD contact allowed. Leake allowed actual grounder production of .172 AVG-.181 SLG last season, good for a minuscule Unadjusted Contact Score of 43. Once you take authority into account, however, that figure jumps to 112. Yes, he allows harder-than-average grounder authority as well. On all BIP types, there is a huge gap between Leake’s 83 and 111 Unadjusted and Adjusted Contact Scores, respectively. That is by far the largest negative differential among 2015 MLB ERA qualifiers; in other words, he was the luckiest starter in baseball last season.

Add back the Ks and BBs, and those figures creep a bit higher to 86 and 112. His “tru” ERA of 4.44 (equivalent to a 114 ERA-) is adjusted for BIP authority and is quite a bit higher than his actual ERA of 3.70 (95 ERA-), and a bit higher than his 4.20 FIP (108 ERA-). Note his very low calculated component ERA of 3.42, which is based on actual performance allowed, including that incredible dose of grounder luck. Despite what his traditional numbers might say, Mike Leake was a well-below-average MLB starter qualitatively last season. Among the 38 NL ERA qualifiers, only one (Aaron Harang) had a higher “tru” ERA.

Going back a couple seasons, his 2014 “tru” ERA of 3.57 was better than his actual 3.70 ERA, while in 2013 (4.07 “tru” vs. 3.37 actual ERA) the calculus was similar to that of 2015. His Adjusted Contact Score was a slightly above average 95 in 2014, a slightly below average 105 in 2013. If you’re feeling charitable, he is an average to slightly below-average contact manager who is hurt further by his low K rate.

Here’s one more table to bring it all into perspective:

K/BB Contact Score Multipliers (2009-15)

MLB	K > 2	> 1	> 0.5	AVG	< (0.5)	< (1.0)	< (2.0)
BB > 2	68.9	77.2	81.8	88.5	95.8	102.8	—
> 1	72.3	81.2	89.1	94.6	101.6	106.2	112.2
> 0.5	77.5	85.6	89.0	96.9	103.1	111.8	118.4
AVG	82.1	87.5	92.1	100.4	106.9	113.2	117.6
< (0.5)	80.3	90.7	95.3	103.0	113.0	115.6	—
< (1.0)	—	93.4	99.0	106.1	114.8	123.1	—
< (2.0)	—	97.3	99.4	109.1	120.7	—	—
ALL	74.8	85.5	91.6	98.4	105.7	110.0	115.8

The K/BB Contact Score Multiplier is the average number by which a pitcher’s contact score is multiplied to derive his calculated component ERA. K rates increase from right to left in the above table, walk rates decline from top to bottom. Each cell is valued by the number of standard deviations better or worse than league average. For instance, Leake had a K rate of over a standard deviation worse than league average and a BB rate of over one-half standard deviation better than league average in 2015. This gives him a K/BB Contact Score multiplier of 111.8. This means that given an average contact score of 100, Leake’s ERA can be projected at 111.8, or nearly 12% worse than league average.

The 2015 season represents Leake’s best case scenario given his current K/BB status. He lucked into an 83 contact score, which kept his ERA in check. There have been two past seasons in which both his K and BB rates were a bit better, nudging him into the 101.6 multiplier cell. That’s his overall best case scenario: basically league average. That is not a pitcher into whom you want to sink big years, and big dollars.

I worked in the Brewers’ front office when we signed Jeff Suppan to a four-year, $42 million contract. It was not exactly my favorite transaction. On one hand, Suppan was quite a bit older (32) than Leake, but on the other, he was living in almost the same Multiplier table cell (113.2) at the time of his signing. Leake is probably going to get more years and even after adjustment for inflation, potentially more dollars per season than Suppan, who earned 2.1 WAR over the life of his contract.

Mike Leake is not a bad pitcher, but he is an awfully limited one. He will answer the bell regularly, but there simply isn’t any real upside above league average. An irrationally exuberant starting pitcher market is going to work out in his favor. He has had the good fortune to have some pretty strong infield defenses behind him throughout his career, and any club in hot pursuit would do well to follow suit, to raise the floor on its investment.