StrategyJuly 18, 2012

Post to Twitter

Selling Low: Jon Lester - 5 comments

By Ian Devine

Numerous fantasy analysts have touted Jon Lester as a buy-low candidate, and given his poor first half, there is certainly a course correction in order. This course correction will not result in improved fantasy contributions, however, and Lester’s owners would be prudent to take advantage of current perceptions. Think of Lester as a bubble stock: he has value, but the superficial indicators mask broader trends and obscure an impending collapse. With so many analysts promoting the value he does have, astute owners have a last opportunity to capitalize on his decline.

The case to acquire Lester is clear. In 112.2 innings, Lester has gone 5-6 with a 4.49 ERA and 1.35 WHIP. ERA indicators paint a different picture, showing his underlining skills to be in line with better results: a 3.57 FIP, a 3.60 xFIP, and a 3.69 SIERA. A quick glance at Lester’s profile and performance would seem to indicate ERA improvement of nearly a full run.

Lester also benefits from the halo effect of a couple seasons as a truly elite fantasy option. In 2009 he went 15-8 with a 3.41 ERA and 1.23 WHIP in 203.1 innings, with 225 strikeouts for a ratio of 9.96 K/9. In 2010, he was, if anything, better, with a line of 19-9/3.25/1.20 with the same 225 strikeouts in 208 innings for a K/9 of 9.74.

Problems arise, however, when we examine Lester’s career trajectory. Looking first at his strikeout rates (all rates below are K/9) we see a clear trend from 2009 to the present:

2009: 9.96
2010: 9.74
2011: 8.55
2012: 7.53

With regard to this year specifically, Lester has posted a tERA of 4.52, nearly identical to his actual ERA and a strong indication of what we can expect in the second half. Metrics like FIP and xFIP brush pitchers in broad strokes by attempting to neutralize batted ball data like home run rate, while tERA accommodates the fact that hit trajectories are, to some degree, within the pitcher’s control. Lester’s tERA is adversely affected in comparison with his FIP and xFIP because his line drive percentage has jumped from 15.9% last year to 23.3% this year. In other words, hitters are not only hitting Lester, they’re squaring up and hitting him hard, and this seems unlikely to change.

Lester’s velocity has declined steadily since he entered the league, which in itself is not concerning. All pitchers see their velocity decline in an absolute sense, so Lester’s movement from 93.5 MPH with his fastballs in 2009 to 92.2 MPH this season is not at issue. What does affect his pitching is the differential between his pitches. While his fastball velocity has declined, his other pitches have remained static. In particular, his change-up velocity is marginally higher than it’s ever been, at 85.9 MPH compared to 85.7 MPH and 85.6 MPH in his 2009-2010 peak. With virtually identical change-up velocities, the decline in his fastball has resulted in less separation, and his change-up, formerly an asset, has become much less effective. Lester also stopped throwing his slider after the 2010 season, favoring instead a sinker, which is much less effective per pitch values.

These changes have manifested most directly in his platoon splits. While his performance has declined against batters from both sides, right-handed batters have gone from .235/.299/.350 and a .292 wOBA in 2009 and .216/.304/.316 and a .285 wOBA in 2010 to .276/.328/.455 with a .335 wOBA in 2012.

The conclusion is clear: with the decline in velocity differential between his fastball and change-up, Lester no longer has a weapon to confront opposite-handed batters. With the transition from devastating slider to relatively ineffective sinker, his ability to challenge same-side hitters has also declined.

Since the start of June, Lester has been moderately more effective than in the first two months of the season, but the improvement is a mirage. Looking at his game logs, he gave up two runs in six innings at home against Baltimore. He had three mediocre outings at home against Washington, at Chicago (Cubs) and at home against Atlanta. He was roughed up by Toronto, gave up one run in 6.1 innings at Oakland, and was tattooed by the Yankees.

Lester is now a matchup option and little else. He performed adequately against National League teams and at one of the most pitcher friendly parks in baseball, and suffered against stronger offenses. Looking forward to the second half of the season, Boston has one of the most challenging schedules of any team. Buster Olney ranks their schedule the second most difficult in all of baseball. Starting Friday, the Red Sox play at Tampa Bay, the White Sox and Blue Jays at home, at Texas, at New York, and at home against Detroit. After a series with the Twins, they go back to Texas, play at Cleveland and Baltimore, and then again in New York before a home series with the Angels. With that schedule, Lester won’t have more than one favorable matchup in the next month, and depending on how the rotation lines up, he may miss that series. Lester’s season line could well get worse before it improves.

Lester has reached a new level of performance. He is no longer a fantasy ace, and is not even a must-start in standard leagues. Unless and until he either defies the linear progression of time and regains his lost velocity or slows his change-up to regain the fifteen MPH differential between his fastball and change-up, and until he either masters his sinker or returns to his slider, Lester will not be a top starting pitcher.

Lester will probably have another nice season or two in which everything breaks right and he helps fantasy teams, but he’s simply no longer reliable. If an owner in your league looks to his improved recent performance or the fallacies of his FIP, xFIP and SIERA, you should move him while you have the opportunity.

Rate this article: DreadfulNot goodFairGoodVery good (3 votes, average: 4.00 out of 5)
Loading ... Loading ...

Want to write for the Cafe? Check out the Cafe's Pencil & Paper section!

Post to Twitter

Related Cafe Articles

• Other articles by Ian Devine

No related articles.

5 Responses to “Selling Low: Jon Lester”

  1. derekmal says:

    I could not agree more with this article. I have been laughing at the “experts” urging others to go out and acquire him for the last month. My advice is to stay away from ALL red sox starters for the duration of the season. That team is a sinking ship if I ever saw one.

  2. User avatar J35J says:

    Nice article and I agree.

  3. Ender says:

    Well i would not buy low because I don’t like the big drop in K%. LD% is one of the more worthless in season stats though so I’m not buying the tERA thing. LD% fluctuates wildly year to year just like BABIP so it is really hard to put a ton of stock into it. In Lester’s case specifically his career xFIP and FIP more or less nail his ERA and his career tERA is way higher so tERA seems to be the less valuable metric for him and the one I would put a lot less faith in.

  4. Ian Devine says:

    @Ender: I went back and checked how closely Lester’s career tERA tracked with FIP and xFIP and the correlation is something like 90%. His tERA from 2008-2011: 4.16/3.36/3.34/3.93. Following his SIERA from the same period: 4.16/3.15/3.25/3.59. FIP: 3.64/3.15/3.13/3.83. xFIP: 4.03/3.09/3.18/3.62. tERA: 4.16/3.36/3.34/3.93. In each case, in 2008 Lester was young and figuring things out, turned into an ace in 2009, remained dominant, though slightly less so in 2010, and performed much worse in an accelerating decline in 2011.

    Through last year all four measures tracked nearly identically. This is the first year with a significant discrepancy, and it’s worth thinking about where that discrepancy originates.

    And this is where I take issue with your comment. You are right, of course, that BABIP is one of the last stats to stabilize. It takes upwards of 650 plate appearances, meaning by the time a sample is large enough, the performance has no validity because the underlying skills may have changed.

    Line drive rate, however, is actually one of the first hitting stats to stabilize. With a sample of 150 PA, line drive rate is reliable and has predictive value. Lester reached that point long ago.

    As the discrepancy between his tERA and other metrics is entirely driven by a meaningful sample of batted ball data, we have to accept it as indicative.

    Contrary to your assertion, line drive rate is one of the first statistics to have meaning within a season or a subset of a season. It should be the first in which you invest.

  5. Ian Devine says:

    @Ender: Ender: I just realized where your perception came from. Line drive rate for batters takes much longer to stabilize – more than 650 PA, and per some studies, perhaps as many as 2000. Line drive rate for pitchers stabilizes much more quickly and can be used in season.


Leave a Reply

You must be logged in to post a comment.