Sorry, Jeff Locke, You Won’t Keep Dominating

Jeff Locke
Jeff Locke says he has “gotten much better” over the last two years.

Jeff Locke, let’s talk.

Heh. Rhymes.

Jeff, your pitching has played a role of significance in the Pirates’ 26-18 start and a spot in the playoffs if Bud Selig took ether and ended the season on May 21. You have pitched five quality starts in your last six outings in the last month, allowing an average of fewer than four hits per start. Combine those with your earlier starts, and you find yourself 11th of 59 qualified NL pitchers with a 2.73 earned run average. I’m not being tongue-in-cheek when I say you should be proud.

It just won’t last.

Nothing personal, Jeff. You are still 25 years old and have a long career ahead, hopefully in Pittsburgh’s rotation instead of some other team’s. You were great in Triple-A last year, and you have definitely earned the spot in the rotation you were so happy to get at the start of the year. If you continue to pitch well, there is no reason you can’t keep that spot when guys like Jeff Karstens and Charlie Morton are ready to join the team.

But your performance won’t continue to match the performance of the NL’s best pitchers. You just won’t keep your ERA below the ERA of guys like Stephen Strasburg, Cliff Lee and Madison Bumgarner. That doesn’t mean you won’t continue to pitch well, but the run-prevention numbers will start to go north. Here’s why, and remember the words “regression to the mean” —

1. You can’t maintain such a low ERA with such a high FIP. No one can.

Let’s start with the basics, Jeff. Your FIP this season is 4.47 compared to your 2.73 ERA. We know that FIP is a better predictor than ERA of future run-prevention ability. Look at it this way, the five pitchers since 2000 with the lowest ERA compared to FIP over a full season:

  • Ryan Franklin, Mariners (2003): 3.57 ERA, 5.17 FIP (1.61 difference)
  • Elmer Dessens, Reds (2002): 3.03 ERA, 4.61 FIP (1.58 difference)
  • Al Leiter, Mets (2004): 3.21 ERA, 4.76 FIP (1.55 difference)
  • Jeremy Hellickson, Rays (2012): 3.10 ERA, 4.60 FIP (1.50 difference)
  • Jeremy Hellickson, Rays (2011): 2.95 ERA, 4.44 FIP (1.49 difference)

Out of the 1,132 pitcher seasons since 2000, not one starter has maintained your 1.74 difference in ERA and FIP over the whole year. Jeff, that alone should be an indicator that your ERA won’t stay this low.

Let’s look at Hellickson, who shows up twice there. In his brief career, the Tampa Bay starter has been a true outlier in outperforming his defense-independent numbers and flouting BABIP-based regression. Why? R.J. Anderson says it has been “some combination of weak contact, defense, infield flies, and park.” It all checks out. Hellickson those a crapload of changeups that induce weak contact, pitches in front of a good Rays defense, draws plenty of pop-ups and plays about half his games in pitcher-friendly Tropicana Field.

Despite all that jazz, Hellickson still can’t keep an ERA more than a run and a half below his FIP. And sorry Jeff, but you’re no Jeremy Hellickson.

Locke’s Response: “I don’t understand what all those things mean anyway, because the people that made them up didn’t play. I just know that if you attack and be aggressive and do what you can do every outing… yeah, maybe the hit totals have been down, but the walks have been up a little bit. It all evens out somewhere. A baserunner’s a baserunner, no matter how they got there. I don’t worry about anything like that.”

2. Your BABIP won’t be that low, even with the Pirates’ great defense.

This season, the Pirates replaced their players’ gloves with high-powered vacuums. Given there was no rule expressly forbidding vacuums, and the fruitful endorsement deal with Hoover, the Bucs have turned an MLB-best 71.6 percent of balls in play into outs.

But bad news, Jeff. Bud Selig found out about the vacuums and has banned them from future games. I had to tip off the Commissioner’s Office. Now your .224 BABIP (third-lowest in the NL) is set to go back up. Since 2000, only one starter has ever kept a BABIP that low over a full season. Can you guess who? Jeremy Hellickson? You’re good at guessing.

You have also stranded more than 82 percent of your baserunners, this year, a number just as unsustainable. You know who pulls that off over a full season? Pedro Martinez, Curt Schiling, Roy Halladay, Randy Johnson, Johan Santana, Jered Weaver, Roger Clemens and young Jake Peavy. I don’t think you will take any offense by my saying you are not those guys.

Locke’s Response: “That’s why there are seven guys behind me. I like to use them all. [The defense] is efficient all the time, a lot of speed in the outfield… You just gotta trust the guys, because they trust you.”

3. You will face tougher opponents.

Jeff, we have all had a lot of fun in May watching you control teams like the Nationals, Mets and Astros, all in the bottom quarter of the league in run creation. Problem is, they are all outside your division, and you have faced the 15th-easiest schedule of 142 eligible starters.

If the rotations holds up as scheduled (which it won’t, but play along), you’ll face the Brewers at the Miller Launchpad, the power-packed Tigers, and the above-average Braves and Giants. These teams can score runs in bunches, Jeff. Godspeed.

It’s Not So Bad!

All of this doesn’t necessarily mean you will go back to the same fringy starting pitcher of the last two years. Pitchers evolve. You appear to have a higher release point on your pitches, you are throwing more sinking two-seam fastballs and getting a few more ground balls hit toward those quality infielders.

Catcher Michael McKenry certainly thinks highly of you, Jeff, and he loves catching you.

“He knows what he needs to do, knows what he can do and he puts his trust in us,” McKenry said. “We’ll go back there and we’ll improvise and change some things when we need to, but most of the time he’s gonna attack with the gameplan he has… Once he starts to learn himself and grow a little bit, his possibilities are endless.”

Endless, Jeff! So take heart: your run-prevention numbers will go in the wrong direction. I’m almost certain of that. But you’re a young pitcher still entering his prime. You have a great defense behind you. You are showing good deception with your pitches. And you’re a left-hander in a ballpark that helps you get outs and swallows potential home runs from right-handed hitters. I think you will be okay in the end.

And sorry for telling Selig about the vacuums.

0 0 votes
Article Rating
Oldest Most Voted
Inline Feedbacks
View all comments

He’s a better option than James McDonald.
That’s really the question.


It way too early to be looking at his FIP as an indicator for where Jeff Locke’s season is going. Homeruns really drive up FIP. He gave up 4 in his first 3 starts. He has given up only 1 since. Anyone care to look at what his FIP is if you factor out those first 3 starts? I crunched the numbers and it comes to 3.19. I agree Locke won’t keep a sub 3.00 ERA all season. But I don’t think he is going to come crashing down either.


And if you fail to acknowledge that the one game he surrendered 3 of the 5 HR allowed doesn’t drastically skew his FIP then that is your prerogative. Don’t get my wrong, I don’t think Locke will maintain a sub 3 ERA either. But this is really misapplication of FIP you are using to support that theory.


I think a lot of people are really missing the point of FIP, it’s a predictive stat, not one that explains what has happened. That’s what ERA does. So, based on FIP, if Locke continues to pitch the way he has this season, his result on the field will be worse than they have been thus far.

As far as FIP to ERA – beyond the 5 mentioned, there are pitchers who consistently outperform their FIP – Jim Palmer is a famous example – but none of them consistently did it to the extent that Locke is this season. And obviously, we need a bit more data on Locke before we start comparing his “overachieving” ERA to that of Palmer’s.

FWIW, Hellickson has a FIP this season that is right around his career averages, an xFIP that is significantly lower than his career averages, and a huge ERA (well over 5) in 9 starts. So interpret that how you like.


FIP is a stat that should be calculated at the end of the year, it is not a stat for a few games. Siera would be a better stat to use for Locke right now IMO.

Tim Williams

His SIERA is 4.75. His FIP is 4.48.

Lee Young

again…take out his first 3 starts and his SIERRA MADRE drops a whole point.

So…your point is?


Tim Williams

I don’t think you can chalk his FIP up to a small sample size, then remove the bad starts and make the sample size even smaller.

I also agree that it’s a SSS. But it’s also a SSS for ERA, which means we shouldn’t be giving much credit to the early season numbers. And if we’re going to use anything to predict future performance, FIP has been proven to be a much better indicator than ERA.

Jeremy J Stein

Before I started looking into the numbers I totally agreed with this article but now I’m not so sure.

I’m not sure you can’t remove the bad starts..

First 3 starts -> 5.17 BB/9 and 3.44 K/9
Last 6 starts -> 3.16 BB/9 and 6.32 K/9

Truely a tale of two pitchers. I expect the BB/K numbers for the rest of the year to look similar to his last 6 starts. His BABIP and Runners stranded % will return to normal so a FIP of 3.45 does appear to be more accurate. I personally think his ERA will eventually sit between 3.50 and 4.25.

Lee Young

But you’re missing the point that he is NOT the same pitcher he was in the first 3 games!

Based on his stats AFTER those first 3 games, his FIP is around 3.45, not too bad of a regression.

I’m going with the new and improved Jeff Locke (and keeping my fingers crossed….lol )




Well, I would argue that Locke is neither pitcher, but somewhere in the middle. But you are correct that FIP – or any predictive stat of this nature – can’t take into account a player “changing,” per say (for example, a pitcher adding a pitch or velocity) . I’m just not sure I see that in Locke.

Much of this discussion still falls under the “small sample size” category, anyway, and it would be unwise to draw too many conclusions just yet.

Lee Young

agree on the ‘conclusion’ part. Another reason I don’t get excited about FIP in May.




IMO, some stats are very useful, but stats used in this article are not.
First of all there is a lot of regression for every pitcher as the year goes on, that is why the teams are so pitch count happy, they are trying to combat the stats.
Locke is very young and he actually should get better, most of his success this year is do to accuracy, he does what any winning pitcher does, he throws the ball where he wants to, when he stops doing that, all the stats in the world, all the defense in the world won’t matter, his ERA will go up. Somebody trying to predict when he will stop throwing the ball where he wants to is simply playing with numbers.
As far as the defense is concerned Locke has a strong defense up the middle but a below average defense on the infield corners, numbers do not tell the whole story on the Pirates defense.
In Summary, if Locke ends the season with an under 4 ERA, that will be acceptable and probably make him a 3 in most rotations, how he manages to do it does not really matter.


I agree that the ERA is going to rise, but i think that the FIP will lower as well.

I feel like we still have a small sample size of the K and BB numbers, as well as the ERA numbers. While the ERA is unsustainably high, I think the FIP is a little too high as well. his K and BB numbers are both worse than what he’s done in the minors and majors. By the end of the season, I bet both ERA and FIP will be in the high 3s or low 4s somewhere.

Thom Kay

James is dead on.

Great pitchers usually have lower than league average BABIPs. But Jeff Locke’s is crazy low, and it can’t stay that low for long. Catfish Hunter was maybe the best all-time at inducing bad hits, but his .239 BABIP, a career record for starters, was considerably higher than Locke’s.

Locke is walking too many batters and not striking enough out to maintain a sub 3.00 ERA. He’s been great, but it won’t last unless he cuts the walks in half or starts striking out more guys.


I think there area a couple of points to be taking into account, on the Optimistic side:

1- You have to consider the maturing process this year for Locke. If we assume that his first 3 starts he was getting his feet wet and learning how to pitch at the majors from guys like Burnett (Jeff has said openly what a great influence it has been pitching right after AJ), what would be the difference in the FIP after those 3 starts??? (sorry, Im not that confident in doing the math…)

2- After those 3 first starts, besides playing the Mets / Astros, Locke faced also the Phillies, Cards, Nationals / Brewers. NOT exactly Light hitting teams!

Again. Sorry for not doing the Math, but I believe you find a much more consistent MLB pitcher as Locke has becomo more confortable at this level!

Lee Young

as I wrote above:

According to this:

Locke’s FIP is .233+a constant, 3.2, which equals 3.45.

Just a little bit of regression and not much more. He is not the pitcher he was in the first 3 games, even with that good one thrown in.


I’m not gonna check it out, but what was AJ’s FIP after his 10 run debacle last year?

FIP is a moving targetand I have never been a fan of it until later in the season.

Lee Young

that’s AFTER those first 3 starts.

Jake Bauer

Do these statistics take into account the type of contact a hitter is making? I don’t care what his FIP is. The fact is that his ERA is 2.73 and, in the end, runs are the only thing that matters.

Mark Ludwig

Nobody who trumpets defensive independent stats like FIP are saying that a player who is outperforming his peripherals has not been successful in the past. The thought is that going forward, a guy who has a higher FIP or xFIP is more likely to regress and be less effective in future outings.

It’s the same reason that AJB looked like a good bet to succeed in Pittsburgh. His ERA in New York was ugly but his xFIP was great. He gave up more HRs than he deserved to by playing in the bandbox in NY. Then he came to Pittsburgh and his HR rate. That alone would probably account for the ~1 run improvement in his ERA predicted by his xFIP. On top of that, he’s walking fewer hitters so there are fewer baserunners against him and he’s playing in front of a better defense. Poof! Now he’s pitching like an ace. As long as he continues to strike people out and limit his walks, he’s a good bet to keep it up. If Locke doesn’t start to do the same he’s a good bet to see his ERA rise.

Lee Young

According to this:

Locke’s FIP is .233+a constant, 3.2, which equals 3.45.

Just a little bit of regression and not much more. He is not the pitcher he was in the first 3 games, even with that good one thrown in.


Lee Young

Jeff Locke walked 8 batters in 9.2 innings, 2 of his first 3 starts.

What is his FIP MINUS those two starts? He is averaging 2.93 walks, 6.1 Ks and Hits /9.



Mark Ludwig

When he’s only made a handful of starts, it’s not really accurate to just cherry pick his better starts to eliminate the rough ones. What I will say is that his FIP was better in May (3.49) and so was his xFIP (4.15…higher b/c he has allowed so few HRs). The thing is, that’s to be expected b/c he’s striking out more batters and walking fewer.

None of this changes the underlying point behind this article and the DIPS. If Locke strikes out more and walks fewer, then he’s a good bet to pitch well the rest of the season (although he’s still not likely to sustain a sub-3 ERA). If he continues to strike out 5 and walk 4 per 9 innings then he is not an especially good bet to pitch well going forward. He might, but it’s unlikely.

Lee Young

I’m with Jeff…I wonder how many of these articles would’ve been written about Tommy Glavine had FIP, FAPs and FOPs been measured back then?

Plus, there’s nothing out there that says he can’t keep it up for a whole year.

I hate articles like this. Just enjoy the moment, dude!

Tim Williams

“I hate articles like this. Just enjoy the moment, dude!”

It’s not really anyone’s job on the site to “enjoy the moment”. We’ve got to point out and analyze what is happening. That’s true whether we’re saying that a good performance won’t be sustained, or a bad performance will turn around.

Obviously people don’t want to hear that Locke won’t sustain this level of performance. But it’s something we have to say. If we just ignore that, but then point out how guys who are struggling will turn things around, then it’s not objective coverage.

Jeremy J Stein

FYI in his last 6 starts, Locke has a 3.16 BB/9 and 6.32 K/9 so he has pitched slightly better than his overall numbers suggest.
That BABIP number stands out though, I’d expect that to return to the mean. Stranded runners percentage will probably drop a little too.

On the flip side, I’m not sure why his K/9 is so low. Perhaps his K/9 will continue to rise. I would guess that his ERA will regress to somewhere between 3.50 and 4.25 which isn’t bad for a 4th starter.


I feel like you guys were the ones drinking the KC juice when he was an all-star. FIP is real, don’t just randomly choose to ignore it.

Tim Williams

Are you saying “you guys” as in the site? Because we’ve pointed out the last two years anytime someone is due for a regression.


Oh by no means. I’m saying that the above comparisons to Tom Glavine are a bit outrageous. I like Locke, but just go to fangraphs and look at his player page…he almost never outperforms his FIP.

Jeremy J Stein

Someone once thought that Chuck Norris was due for a regression but then realized that Chuck Norris eats FIP for breakfast, so they reconsidered.

Lee Young

I’m not ignoring it, I am just saying that players HAVE outperformed their FIP, BABIP and all that other stuff for long periods of time, including, gasp, a whole year.

He has been a little wild and his K rate is down from what he did in the minors. If that K rate comes up….

As for Glavine, my point was, when he was posting his first 20 win season in ’91, I wonder how many FIP’ers, had they been around then, would’ve said Tommy is gonna regress.

Well….Tommy never did…so maybe Locke has figured out how to translate his minor league stuff to the majors. And wait until his K rate comes up.



Not to pick on you, Lee, but Glavine’s “breakout” 1991 saw a marked drop in his FIP as well as his ERA, mostly due to a jump and K rate and drop in BB rate.


LY: Being mentioned in the same company as Jeremy Hellickson and Al Leiter ain’t bad at all. I wonder if James was studying these indicators when Locke was in AAA last year and in the Top 5 all year. He was excellent, but only got to Pittsburgh at the end of 2011 and 2012 when he was cooked. He can throw 91 inside without fear and has one of the best change-ups going. He was good for about 150 innings the past two years and I expect he can do the same in the majors. I am with you – just enjoy the moment, and a 2.73 ERA from the No. 5 SP is a lot more than I expected and I have been praising this kid for the last 2 years.

Mark Ludwig

The thing is that when Locke was “in the Top 5” in Indy, he was doing it with better peripherals. That’s why his FIP (back to the indicator of future performance) was 3.24 or roughly .75 runs higher than his ERA. In Pittsburgh this season, Locke is walking an extra batter per 9 and his K rate is down by about 3 per 9. Those are big swings. Now he might not regress all the way to a 4.5 ERA but if he does, it’s going to be because he allows fewer balls in play (by striking out more hitters) and prevents more baserunners (by reducing his walks.

There’s nothing wrong with an article like this pointing out that Locke has things to work on if he wants to continue being effective. That’s true. It’s also not a bad thing. He’s a young guy getting his first extended look in a big league rotation and, thus far, he’s performed admirably. However, if he wants to continue to put up good numbers instead of being the guy pushed out by Morton, Karstens, McDonald, Gomez, Cole or whoever else is fighting him for innings he needs to improve his peripherals. Either that, or he could hope to succeed by sheer luck in a way that few (although not zero) pitchers previously have. That’s like saying that you don’t need to go to work tomorrow because you’re just gonna win the lottery instead.

Andrew Smalley

This. In. Its. Entirety.

Andrew Smalley

Actually, Tom Glavine’s ERA only outperformed his FIP by .4. If you actually look at his career stats, as he got older and less effective, his FIP showed as much.

James is ‘enjoying the moment’ like the rest of us who are interested in predictive metrics. But, he’s not enjoying the moment with his head in the sand. I presume you (and others who ‘hate these articles’) will be the first one to start second-guessing why Locke is in the rotation once hits start finding holes and fly-balls start leaving the yard. In this sense, James and others of his ilk, won’t be nearly as ‘pissed off’ because they actually expected some regression to happen.

At that point, James (and I) will probably say, “Enjoy the moment, Lee!”, because even with regression, Locke is still fun to watch and should be in the rotation.

Ignoring more accurate and more predictive metrics is akin to putting one’s head in the sand and hoping for the best. That’s fine if you’re a casual fan, or even if you’re not. But, in terms of what to expect going forward, you’ll be wrong more than right.

Great article, James.

Andrew Smalley

While I think some of the humor fell flat, this is a great ‘counter-point’ to Travis Sawchik’s piece in the Trib. James is using FIP properly, in terms of its predictive-ability for future performance. Travis, on the other hand, is saying that the ‘numbers aren’t always correct’. That’s erroneous; the numbers used to calculate FIP – BB/K/HR – are what they are – there’s no sense (or accuracy) in saying the #’s aren’t correct.

James is using historical examples to show why such a difference in FIP and ERA is unsustainable. Travis tries to say that the numbers aren’t picking up ‘non-quanitiable’ changes, like deception and pitch-sequencing. He’s right…..but, those changes that aren’t quantifiable usually just result in noise, having very little impact on the numbers long-term.

Well done, James. Very informative and even-handed. But, more importantly, great (and proper) use of the stats.

Kevin Anstrom

Late to the conversation. I usually don’t comment on stats articles.

How would you explain the large gaps between ERA and FIP for the Tigers and Pirates? Is it still an issue of “small sample size”?

Given that the Pirates will have Marte in left, etc.why wouldn’t we expect Locke’s ERA to be lower than his FIP?

Comments are closed.

Most Voted Comments