Two lackluster years

John Lackey has always been a little overrated. Praised for his postseason success and competitive demeanor, Lackey used to have a reputation for being one of the best starters in the American League. In 2007, he even contended for the Cy Young award, managing an impressive third-place finish.

This esteem helped him land a monster five-year, $85 million contract from the Boston Red Sox. A surprising deal at the time, the signing is as perplexing now as it was then. In only one season had he bested an xFIP- of 87, and by the time he signed the contract, that was a few years in the past.

Paid like a top-tier starter, Lackey has been anything but. Replacement level*. Bust. Flop. Call it want you want, he has not lived up to his contract. Aside from moving to a more offense-friendly environment and a tougher division, what has been driving his mediocrity?

*According to Baseball Reference WAR.

Being easy to hit is bad

According to my database, Lackey had a whiff rate (swing-and-miss per pitch) of 8.1 percent from 2008-2009 and 7.3 percent in 2010-2011. Despite an ostensibly small difference, if we perform a significance test between the two proportions, we find a significant difference at a 95-percent level for a one-sided test.


This graph shows the whiff rate of all of Lackey’s pitches based on horizontal pitch location, split by batter handedness and time frame. Gray bands indicate confidence.

For righties, we see that he is really the same in both time ranges. This is affirmed when we break down his whiff rates by batter handedness.

Years      Handedness  rv100  Whiff  Swing  Contact
2008-2009       L      -0.69   0.07   0.44     0.85
2008-2009       R      -0.18   0.10   0.45     0.78

2010-2011       L       1.87   0.05   0.44     0.88
2010-2011       R       0.88   0.10   0.47     0.79

rv100 = linear weight run value per 100 pitches
Whiff = whiffs / pitches
Contact = 1 – (whiffs / swings)

As you can see, the whiff dropoff is coming against lefties. When we look at the graph, we see that there is a significant difference between his whiff rates on pitches middle inside between the two time frames. In fact, it almost appears as if his whiff rates to lefties in that area of the zone have been halved.


This graph shows the whiff rates of his pitches by vertical location, split by batter handedness and time frame. The vertical borders of the strike zone are marked by the two dotted lines. As the graph shows, his whiff rates are much lower on pitches below the bottom of the strikezone than before.

Lackey has always relied on his breaking balls. His fastball has never had elite velocity nor spectacular movement. His changeup is usable, but not special. Lackey has always made his money with a very effective curveball- and slider-heavy repertoire.

As soon as batters stop having trouble with his breaking balls, Lackey stops having success, and that’s exactly what we see here. The opposition is simply no longer swinging and missing at his breaking balls below the zone. His trademark curveball now only garners whiffs on about 9.5 out of every 100 pitches—a rate significantly below the league average.

Interestingly, batters haven’t stopped chasing pitches below the zone.


Batters are chasing pitches below the zone at a similar rate as before; they are just making more contact on these pitches. All of this may have a compounding effect. Since his breaking balls are less effective than before, it’s possible that he has more trouble getting ahead of batters and staying in pitchers’ counts, an effect that would reduce the effectiveness of all of his pitches. But when we look at the distribution of his counts, this is not what we find.


  0-0    0-1    0-2    1-0    1-1    1-2    2-0    2-1    2-2    3-0    3-1    3-2 
0.2741 0.1368 0.0604 0.1013 0.1072 0.0927 0.0296 0.0516 0.0762 0.0090 0.0188 0.0425 


A Hardball Times Update
Goodbye for now.
  0-0    0-1    0-2    1-0    1-1    1-2    2-0    2-1    2-2    3-0    3-1    3-2 
0.2607 0.1286 0.0598 0.0975 0.1120 0.0959 0.0302 0.0576 0.0790 0.0086 0.0210 0.0491

The distribution of counts against Lackey is, surprisingly, almost identical in each time frame. The only count that shows a difference greater than one percentage point is the proportion of 0-0 counts, which is higher in 2008-2009 than 2010-2011, implying that his at-bats in 2010-2011 were longer than before.

Location, location, location

Until now, we really have only seen notable differences with Lackey’s breaking balls; batters are no longer struggling with breaking balls below the zone. But his fastball has been very ineffective, as well, despite very similar movement and velocity. This suggests that another cause, such as command, may be at play here. To test this, let’s look at the density of his horizontal pitch locations for his fastballs, split up by batter handedness and year.


In 2011, it appears that many more of his fastballs were down the middle (marked by the dotted lines) than in previous years. This is not a rigorous examination of his command, but it does suggest that he was having command issues with fastballs to lefties in 2011.

Finishing thoughts

Lackey has struggled, with a major contributor being the reduced effectiveness of his breaking balls. His slider and curveball are simply not as challenging as they used to be, especially below the zone. As shown above, there is also evidence suggesting a drop in fastball command in 2011.

Overall, though, not a lot has changed. Lackey started throwing more sliders than curveballs in 2011, but there is nothing in the data that suggests pitch selection is at fault here.

A lot of his poor performance is likely due to bad luck; he has significantly underperformed his defense independent pitching statistics (DIPS) in the past two years, which is a characteristic that he did not display before. He did post an xFIP of 116 in 2011, which is bad, but not horrible. Going forward, it is doubtful Lackey will ever be worth what he signed for, but the tools are still there for a solid contribution.

References & Resources
*PITCHf/x data from MLBAM via Darrel Zimmerman’s pbp2 database and scripts by Joseph Adler/Mike Fast/Darrel Zimmerman

Newest Most Voted
Inline Feedbacks
View all comments
Dave Studeman
12 years ago

Great stuff, Josh.  But I’m not sure I recognize your format for xFIP.  Those are usually expressed as ERA equivalents. but your figures look normalized to 100, I think.

Josh Weinstock
12 years ago

Yea, sorry that’s xFIP- , not xFIP. Yea, it’s normalized to 100 where above 100 is worse than average, etc.

12 years ago

I love stuff like this. However, when I do a very quick comparison of K rate, xFIP and a few others between 2010 and 2011, his performance looks like it declined more against RHB than LHB, even though your article implies it was the lefties that primarily caused his demise. Thoughts?

Josh Weinstock
12 years ago


I was actually looking at 2008-2009 vs. 2010-2011 to get more data, so I wasn’t really focusing so much on differences between 2010 and 2011. There’s also a lot of variability in year to year platoon splits for pitchers. I mainly just went off of what I observed in the PITCHf/x data.