Suffice to say that 2015 did not go as expected for Julio Teheran. In his first two seasons as a full-time starter for the Braves, he threw up a 3.03 ERA and 3.58 FIP (3.74 xFIP). Only about 30 starters outpitched him in those two seasons, and only about 24 starters had better overall run-prevention than him. (While hardly a fair comparison, his ERA over those two seasons just 0.01 above that of Max Scherzer.) With that said, expectations for him were not super-high, for a few reasons: 1) he continued to outpitch his FIP and his xFIP; 2) his potential reliance on strong outfield defense for run prevention; and 3) inconsistency with his fastball velocity. The latter is not something of huge concern to everyone, but a lot of people like to see young pitchers maintain or improve on their arm strength as they mature, rather than experiencing declines that often afflict older hurlers. In terms of projections, ZiPS discounted those concerns and pegged him for 2.8 FIP-based WAR, basically his career average to date, while both Steamer and myself figured he'd take a step back and post under 2 fWAR (1.8 for Steamer, 1.7 for me, on the back of an FIP close to 4.00). Honestly, I was hoping to be wrong, but I figured that the lack of a set of strong outfield defenders would rattle him a bit and force him to work in disadvantageous situations too much (Steamer was just really skeptical due to his velocity decline).
In any case, Julio Teheran ended up posting just 1.1 fWAR to go with a 4.04 ERA and 4.40 FIP. Needless to say, this performance took him from being a #1-ish pitcher in his first two seasons to a pretty poor starter as far as 2015 goes: over 100 starters were more valuable than Julio, and he was one of the worst full-time starters in baseball. Applying a postmortem to his subpar 2015 involves two interrelated components:
- What exactly went wrong?
- Did he really figure stuff out in his last few starts of the season, or are there danger signs to look for next season?
A Step Back
I had more than a few desires to write about Teheran's issues as he was in their throes, but I always hesitated because it was hard to say something profound and interesting. On the one hand, his issues were obvious; on the other hand, they were obvious in such a way where I couldn't find much more to say beyond the surface-level statistics that indicated he was not pitching very well. So, the following starts a very high and obvious level, and tries to drill down into some things that are potentially suggestive of why he struggled.
Big Picture Outcomes
Teheran's problems are evident in his 2015 stats: unfortunately for him, luck (or random variation, if you want to call it that) didn't do much to mask any of his problems.
Anyway, you can pretty much see that the things that are good when they go down went up, and all the things that are good when they go up went down. For pretty much anyone, fewer strikeouts, more walks, and more homers allowed are not a recipe for success, and the stuff in the second chart largely drive the first chart. (As a dark side note, one wonders how potentially nasty Teheran's 2015 numbers would have looked if he didn't increase his groundball percentage, as that would potentially led to even more balls hit into the stands.) So, what's behind some of these numbers? I don't have a definite answer, but here are some ideas.
(Side note: I know there was some stuff about mechanics and him moving around the pitching rubber and stuff. I'll be the first to tell you that I never played above a high school level and don't have a great understanding of this stuff, so I'm just explaining what I see in the numbers. If mechanics-related considerations led to his outcomes, that's not self-evident to me and is thus not directly addressed here.)
General Command and Control
Below are a series of three heatmaps (thanks, Baseball Savant!) showing where Julio Teheran's pitches have tended to go.
I think a lot can be read into the above, and I'm curious to see what other people draw from those charts, but here are my thoughts:
- From 2013 to 2014, Julio sharpened up the amount of pitches that went wide of the zone (horizontally). He's always missed occasionally to the third-base side of the plate, and while this got worse in 2014, he didn't miss as much to the first-base side. However, the amount of pitches wide of the zone is really noticeable in 2015.
- Additionally, there are a lot of pitches up and out over the plate. You don't really want to leave pitches up to major league hitters, especially if they can catch up to them. There were a lot more of those in 2015, and especially that area that would be up-and-away-ish (but still over the plate) against lefties, and up-and-in-ish to righties.
A lot of the peripherals and results might be explained by the above. (Or not, your mileage may vary.)
Boom Whimpers the Dynamite
The below charts are similar, but just show the pitch locations where Teheran was taken deep. I don't think the conclusions here are clear-cut, but what's interesting to me is that in 2013 and 2014, Teheran occasionally just got beat on decent pitches,
Again, I take two main things away here:
- Before 2015, Teheran would occasionally get beat on pretty good pitches. In 2015, there was less of that, and he mostly just got beat on pitches around the heart of the zone.
- If you squint, you can kind of see a shape that looks kind of like \ in the heat map for 2013 and 2014. That basically means Teheran got taken deep on pitches down and in (but not down or in enough) to lefties, and to belt-to-letter high pitches that weren't on the inside edge to righties. (Of course, I could be mixing this up and lefties are homering on outer half pitches, getting arms extended. I could follow up on this if people are interested, but I don't want to add six more charts to this post.) In 2015, though, there's stuff in the bottom left-hand corner too. This is a pitch location where Teheran started getting beat in 2015 where he was fairly impervious in 2013-4. Maybe this is lefties getting their hands extended, or righties homering off of pitches that weren't buried enough down-and-inside, but either way, it's an additional weakness that hitters started exploiting.
Again with the Splits
Much has been made of Teheran's struggles on the road. It may or may not be surprising that in his worst season so far, those splits have been exaggerated as well.
In 2014, when Teheran was at his best (so far), he managed his road-home split into nothingness, and lefties on the road didn't torment him much more than in general. In 2015, though, lefties tormented him a bunch in general, as did pitching on the road (again, in general). He allowed a 7.22 FIP to lefties on the road, which is some kind of crazy video game slash Jo-Jo Reyes number. Meanwhile, he dominated righties at home, with a sub-2.00 FIP. Maybe Teheran and Shelby Miller were eating at the wrong places when away from Atlanta, or something.
I'm not sure what, specifically, causes Teheran to pitch worse on the road, or against lefties in a way that far exceeded his usual split against them. Whatever it was, though, it really shook up his outcomes and numbers for 2015.
Generating Fly Balls by Pitching Up in the Zone (or Not)
In June, Murphy Powell of SBN's Beyond the Box Score chronicled Teheran's newfound aversion to high pitches, preferring to go low-and-away to try to get whiffs (and/or groundballs). This is a complementary factoid to other things discussed above: it makes sense that if he's abandoned his high pitches but is still pitching worse, his command/control might be suffering, and he might be getting taken deep on not-so-high pitches as well.
Overall, Teheran's FB% went from 41% to 44% to 36%. The Braves had an average outfield defense in 2013, a great one (top five in baseball) in 2014, and a pretty poor one in 2015 (bottom 10). To the extent that Teheran was consciously altering his pitching strategy to try to pitch to his defense's strengths (and avoid its weaknesses), the movement away from high pitches makes sense.
Overall, here are some raw numbers (if you're interested, I suggest looking at the article linked above, which goes more in-depth on this issue, at least through the first three months of the season):
- In 2013, Teheran threw about 19% of his pitches at the letters, and about 8% above the letters (think eye-level).
- In 2014, these percentages were 20% and 10%, respectively. This is consistent with the increase in FB% over those years. The big change was really going up and away out of the zone against lefties, and a bit more up-and-in stuff to righties.
- In 2015, the percentages are 20% and 10%, so somewhat surprisingly, the raw proportion of pitches thrown fairly high from Teheran was similar to 2014. There were, however, fewer pitches up and away to lefties than in 2015, Against righties, Teheran switched from throwing more pitches up-and-in to up-and-away. I'm not sure how/whether these drive his FB%,
- The swing% is a bit of a complicated analysis. In short, lefties began to swing considerably less against his up-and-in offerings (in fact, a lefty never swung at a Teheran pitch that was inside and at the knees or above, though Teheran threw fewer than 50 such pitches altogether) as well as his low-and-away offerings. He did, however, get lefties to fish more down and in.
- For swing% against righties, the analysis is fairly muddled. Righties did swing more at pitches up and in, but also a lot more at pitches low and away. Overall, the individual swing percentages per part of the strike zone are very inconsistent. Overall, if Teheran were focusing on a different part of the zone to generate groundballs, it would probably be low-and-in pitches to lefties, as his FB% against lefties was considerably lower than against righties.
Putting some things together, it looks like Teheran may have been redoubling his efforts to neutralize lefty hitters' platoon advantage against him by getting them to roll over pitches down and in. However, this only worked to some extent, as he was successful in the down-and-in corner of the strike zone, but got hammered more whenever the pitch was either not down or not inside enough (see here, for example). Unfortunately, there's no way to know how much success he would have had if he continued to go up-and-away-ish to lefties with greater regularity in 2015, even with the spottier outfield defense. In any case, the stratagem did not directly work out, especially not on the road (for some reason), and combined with worse command/control overall, led to a disappointing season.
What about a Turnaround?
Over his first 14 starts, Julio Teheran struggled mightily. Through mid-June, he was throwing up a 5.07 ERA and 5.08 FIP (4.35 xFIP). Most glaringly, his strikeout rate was dwindling to 7.67 per nine innings, and he was allowing 1.54 homers per nine due to a HR/FB% of about 17%.
The remaining 19 starts were a correction of sorts. The HR/FB% went to a normal-ish 10% and the strikeout rate crept back up to over eight punchouts per nine. His ERA went to 3.33 over this period, with a 3.94 FIP to boot. (Note that this FIP is pretty close to what I thought he'd pitch to for the season.)
It may be tempting to say that his last 12 starts, covering August, September, and that one game in October, represent even more of a corner-turning for him. However, his success there was almost mirage-y: his ERA of 2.95 was a whole run below his FIP of 3.99 (similar to the broader set where the 7 starts before that were incorporated), and his HR/FB was unchanged, with a lower strikeout rate (7.66 per nine) to boot. The real difference between the sets of his last 12 and his last 19 starts was an uptick in his strand rate and a further decline in his BABIP-against, which brings me to...
Overall, I think Teheran figured out some things after mid-June, but I'm not sure he figured out enough, or applied it consistently enough, to warrant a blanket proclamation that he fixed whatever ineffectiveness ailed him throughout the season. Specifically, see the following:
Specifically, the first 14 starts of Teheran's 2015 are just nasty all over the place. Fewer strikeouts, way more walks, more homers allowed (partly due to a crazy-ish HR/FB%), harder hit balls in general, and fewer runners stranded, all this despite an uptick in grounders. The next 19 starts, constituting a bit more than the back half of his season, however, look more like the Teheran of yore. Pretty much everything is in line with either league averages and/or his past performance, with the exception of an elevated walk rate. Cutting it down to his last two months of 2015 makes the ERA look nicer, but the decrease in walks has a corresponding decrease in punchouts, and is pushed downward by lower BABIP and higher LOB% over that stretch of starts.
In other words - did Teheran rebound? Yes. But even as he did, he didn't rebound to the type of pitcher he was earlier, and definitely not to a sub-3.00 ERA level of expected performance. Instead, he mostly got everything in line except for his walks, and went back to being the kind of average-y pitcher that results from from solid stuff and occasional command/control issues. He showed he could be better than the April-June pitcher he was, but the walk rate remains a concern.
So What Now?
Julio Teheran is a bit of a nightmare to forecast for 2016 because he's followed up two strong years with a poor one. If he can pitch like he did since the latter half of June with walks reined in, he can go back to being that effective pitcher; if he can't, it may be some rough times ahead for him and the Braves (or at least mediocre times, if not outright scary). With a bit of optimism but not expecting a turn-on-the-dime improvement from his 2015,
I would feel comfortable saying he's somewhere marginally above a 2 fWAR pitcher next year, with an FIP a tick below 4.00. Basically, I think he'll be something like the middle column of the table above, with the ERA catching up a bit to the FIP (think closer to 3.5 than 3.3) due to a slight uptick in his BABIP. I do think he'll rein in his walks a bit, but this may require an even more contact-oriented approach that will cost him some strikeouts as well, so I'd guess he puts up something like a 7% walk rate and a 21% strikeout rate.
In any case, he's signed longterm by the Braves, and it remains to be seen whether any team will feel comfortable enough with his 2015 backslide to obtain him for what would likely be a pricy package coming back the other way. Here's hoping he improves beyond my expectations, and that we never see silly things like this ever again.