The Most Surprising Starting Pitchers: Will They Keep It Up?
Last week, I discussed the most disappointing starting pitchers after comparing CBS’ projected versus actual ranking. Today, we look into the surprises and ask ourselves whether they could keep up the magic.
I limited the surprises to those ranked within the top 150.
| Player | ERA | WHIP | W | K | Projected CBS Rank | Actual CBS Rank | Diff |
|---|---|---|---|---|---|---|---|
| Ross Stripling | 2.43 | 1.1 | 8 | 114 | 631 | 71 | -560 |
| Tyler Skaggs | 2.68 | 1.22 | 7 | 106 | 626 | 133 | -493 |
| Marco Gonzales | 3.38 | 1.14 | 11 | 104 | 454 | 81 | -373 |
| Mike Foltynewicz | 2.85 | 1.16 | 7 | 125 | 448 | 92 | -356 |
| Miles Mikolas | 2.82 | 1.07 | 10 | 89 | 268 | 67 | -201 |
| Sean Manaea | 3.38 | 0.97 | 9 | 88 | 258 | 78 | -180 |
| Eduardo Rodriguez | 3.44 | 1.22 | 11 | 110 | 267 | 94 | -173 |
| Patrick Corbin | 3.13 | 1.04 | 7 | 158 | 198 | 51 | -147 |
| J.A. Happ | 4.18 | 1.18 | 10 | 130 | 240 | 109 | -131 |
| Rick Porcello | 3.93 | 1.24 | 12 | 121 | 217 | 95 | -122 |
| Blake Snell | 2.27 | 1.07 | 12 | 134 | 145 | 24 | -121 |
| Trevor Bauer | 2.44 | 1.15 | 8 | 182 | 128 | 26 | -102 |
Well duh, of course we all could have guessed who the most surprising starter has been. Obviously, Ross Stripling’s projected rank was so low because he was just your standard non-closing reliever to open the year. But after a month, injuries vaulted him into the Dodgers rotation, and he’s been brilliant since. However, this ain’t gonna last, even though his SIERA validates the breakout. You see, you need to first determine whether the underlying skills driving SIERA are actually sustainable before blindly pointing to it as reason for him to continue with a sub-3.00 ERA. His 11% SwStk% is barely above the league average, so his 28% strikeout rate seems seriously inflated. Could we really expect any starter to maintain a 3.5% walk rate? Since 2010, out of 731 qualified pitcher seasons, just 10 pitchers have finished with a 3.5% or better walk rate.
Then, there’s the innings concern. He has already pitched a bit more than 20 innings more than last year, and his professional career high is just 128, set all the way back in 2013. I don’t expect him to fall apart as his skills have lots of room to regress and still remain pretty good. But, I’d bet he finishes outside the top 100.
Staying healthy has been an issue for Tyler Skaggs, but this year he has ridden a SwStk% surge to a strikeout rate spike, to go along with better than average walk and ground ball rates. The jump in SwStk% and strikeout rate looks primarily fueled by his four-seamer, which has hit a double digit SwStk% for the first time. Since the pitch hasn’t gained any velocity, it’s hard to believe this is sustainable. His best pitch has only generated a 14.3% SwStk%, which isn’t exactly the mark of a pitcher with a dominant arsenal. As an owner, I worry about regression, not only toward his SIERA, but also in his skills which could mean a 4.00 ERA the rest of the way.
Who knew that Marco Gonzales, former Cardinals prospect, would enjoy his breakout in the American League? He’s displayed sterling control, but the SwStk% is below average and doesn’t give me much optimism that he can keep his strikeout rate above 20%. Fine in AL-Only leagues, but I wouldn’t want to count on him to earn value in shallow mixed the rest of the way.
Man, if Stripling’s strikeout rate looks suspect on the back of an 11% SwStk%, how does Mike Foltynewicz’s similar strikeout rate look with a worse 10.5% SwStk%?! Between his arm issues, an ERA nearly a full run lower than his SIERA (which is build on an unsustainable strikeout rate), he looks like a prime sell high as there’s sure to be an owner in your league who believes the hard throwing 26-year-old is in the midst of a true breakout.
Miles Mikolas reinvented himself in Japan and those improvements have carried over back to MLB. Unfortunately, most of it has been smoke and mirrors. Mikolas’ sub-20% strikeout rate is weak, and his SIERA is nearing 4.00, more than a full run higher than his ERA. Given the low strikeout rate, I’m not even confident that he’ll earn shallow mixed league value the rest of the way.
I was a big fan of Sean Manaea heading into the season, and while his ERA is making me look good, his poor underlying skills have me biting my nails. What happened to his strikeout rate?! His SwStk% is still in the double digits (which is funny because it’s just a notch lower than Folty’s, yet his strikeout rate is more than 10 percentage points lower), but for a second straight season, the whiffs aren’t translating into strikeouts. He’s obviously not going to maintain a .221 BABIP, so he’s going to need to start whiffing more batters to keep his ERA from skyrocketing.
Sadly, Eduardo Rodriguez’s ankle injury cut into what had been a nice season of skills growth, with improved control and spike in ground ball rate.
Patrick Corbin opened the season averaging 92-93 mph with his fastball, but since the beginning of May, he has only averaged above 91 with his fastball in a start once…his last one. And yet, that hasn’t hampered his performance in the least. The key here is that his already elite slider has become INSANE, generating an absurd 28.4% SwStk%, and the fact that he throws it nearly 40% of the time has made him absolutely dominant. I don’t know what’s behind the suddenly insane level slider, versus merely elite, and always get nervous about starters who lose velocity, but as long as he’s throwing that slider 40%, he should remain very, very good.
It figures that J.A. Happ outperforms his SIERA for three straight seasons, then when his skills surge and finally support a mid-3.00 ERA, his LOB% collapses and ERA jumps over 4.00. Happ isn’t doing a whole lot different from a pitch mix perspective, aside from swapping some sinkers for four-seamers, but three of his four primary pitches are generating low teen SwStk% marks. So while no pitch is a standout, everything in the low teen range (plus a sinker with an above average SwStk%) has led to a career best SwStk% and strikeout rate. I still don’t like repertoires that don’t feature an above average whiff pitch, so I remain bearish. I’m going to bet here that his skills regress and the SIERA rises to meet his ERA.
Rick Porcello has really alternated good seasons and bad seasons! His skills are more or less the same as they have always been, but his luck metrics continue to jump around, which is why sometimes you get a high 4.00 era and other times a mid-3.00 ERA. The 12 wins have also boosted his value thanks to the elite Red Sox offense. I see no reason to believe we won’t see more of the same the rest of the way.
It was going so well for Blake Snell until he hit the disabled list with shoulder fatigue. He’s only expected to miss one start, but it’s still a concern, albeit minor perhaps. He was enjoying a breakout year thanks to a strikeout rate surge driven by a 13.4% SwStk%, but his control remains an issue and that .243 BABIP and 86.3% LOB% aren’t going to stay at those levels. It might surprise you to hear that his SIERA sits at 3.71, giving him one of the larger gaps between ERA and SIERA. I’m still a fan as control can improve overnight, but I wouldn’t bet on a sub-3.00 ERA the rest of the way.
So I guess Trevor Bauer finally translated all of those pitches until an elite strikeout rate. He still is throwing six pitches, but he has featured his curve, slider, and cutter, with the latter two generating SwStk% marks above 20%. Even the curve has been excellent at a 15% SwStk%, given him a truly deep arsenal. When you have so many pitches to go to, it limits the downside risk in my opinion. He’s still going to struggle with his control here and there, but he’s near the top of the guys I feel most optimistic about from this list.
Mike Podhorzer is the founder of ProjectingX IQ, an advanced fantasy baseball analytics platform that transforms projection data and in-season performance signals into actionable intelligence. He is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year and three-time Tout Wars champion. He is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. Follow Mike on X@MikePodhorzer and contact him via email.
Mike, your old xKs and xbbs spits out a 26% k rate and 6% walk rate for Stripling (compared to 28% k and 3.5% bb in reality). Of course, you already knew this bc you are the human xstats calc…but anyway, looking strikes are a real strength for him…however, his strike % and looking strike % have never been this high so I agree he should regress a bit. Outside the top 100 seems harsh though — 26% ks and 6% walks with 47% gbs is still really good. Berrios and Nola are the closest comparables with these periphs…and he’s well past all the relevant stabilization points. Leave him top 50!
I agree with your takes on the other guys btw–corbin and Bauer are faves!
Thanks for calculating what I was too lazy to do myself! You got it right that a high looking strike% isn’t as sustainable as a SwStk%, especially when history shows it’s a career best.
Remember, those strikeout and walk rate expectations are built on what you agreed was unsustainable looking and overall strike % marks. If those regress, then you’re looking at maybe 24% and 7%, which is still solid, but nothing real special.
Anytime haha big fan of the old xks and xbbs equation. Fair point, but even at 24% and 7% with 47% grounders his comparables are all in the 3.5-3.9 siera range (price,Heaney, Anibal Sanchez, Cahill, E-Rod, Tanaka). It’s a pretty high floor…and it’s actually just the strike % that is uncharacteristically high. The looking strike percentage is only slightly up from last year, and similar to where it was two years ago…and he’s so for past his stabilization points for walks and ks! Also, the year to correlation for looking strikes is only slightly worse than it is for swinging strikes (citing your research again).
Plugged in the xks and xbbs again, this time using his three year looking strike and overall strike percentages. His xK is 25% and xBB is 8%–still good, and the bb number is well above his 3 year average of 5.6%. If he’s 25% ks and 6% walks he’s still really strong, albeit more of a mid 3s guy (perhaps like Nola and Berrios)
Happ’s already had a few bad outings in a row, Stripling is starting his regression…Corbin is holding steady but he’s also not showing his “Ace” form as he did in the early months.
I saw your take on Skaggs and feverishly tried to find something else backing up his gains. All I really came away with is that his GB% has spiked up nicely to 47% after trending FB heavy with a GB rate in the lower 40’s the last two years. Likely a few more grounders have led to some timely double plays fueling his elevated 80.8% strand rate that will surely come down. A 3.53 xFIP SP is a fine #2-3 fantasy SP but I have to agree with you that my enthusiasm for Skaggs should be tempered and the projections calling for a mid to high 3’s era are probably spot-on rest of the way. His heat maps with the 4-seamer also have a higher density of pitches up compared to 2017. In theory just throwing the exact same fastball higher on average could/should lead to more whiffs however it’s tough to know if the average height change of his 4-seamer accounts for the whole swSTR% surge; I doubt it.
Good stuff Mike!
This is a really lazy article. Main thrust of the 1st paragraph is Stripling’s SwStrk% isn’t high enough so he’ll definitely K less and BB walk more? Come on. He’s throwing 7% more balls in the zone this year and 5% more first strikes with extremely similar stuff(very similar O Swing, O Contact, SwStr), of course his ratio is going to improve:
2017: 8.96 Ks/9 2.30 BBs/9
2018: 10.26 Ks/9 1.26 BBs/9
2017: Zone 42%, F-Strike % 64.8
2018: Zone 45%, F-Strike % 68.4%
I also take issue with the SwStr% being the be-all, end-all of K/BB rate as implied in the Stripling paragraph as we all know this is the not the case. Here are some choice SwStr rates and their corresponding K rates this year:
In the Stripling(11%) range:
Francisco Liriano, 10.7%, 7.42 Ks/9
Clayton Kershaw: 11.1%, 9.15 Ks/9
JA Happ: 10.5%, 10.26 Ks/9
A tier higher(~13%) where the implication of the article would be that then Stripling’s #s would be ok if it was up here:
Chris Archer: 13.5%, 9.90 Ks/9
Dylan Bundy: 13.1%, 9.64 Ks/9
Jon Gray: 13.2% 11.09 Ks/9
It’s silly that I have to write this but the differences here are because O-Swing, O-Contact, Z-Swing, Z-Contact, Zone, F Strike % all makeup the profile that contributes to K/BB rate and simply picking just SwStrk rate and ignoring the rest is bogus.
Your point would have more strength if you used K% and BB%, instead of K/9 and BB/9. Why use the latter, when the former is much more descriptive of actual performance, and not influenced by BABIP?
My point doesn’t need “more strength”, you made the argument that Stripling’s numbers aren’t sustainable purely based on his SwStr rate only being 11%. Banking an entire K%/BB% projection on SwStr rate is ludicrous. Additionally, I showed you that he is throwing considerably more strikes this year with very similar stuff, therefore his K% and BB% will stay improved(unless you have a reason for his *control* to regress).
Stripling’s K% is down by about 10% from earlier in the season.
Agreed. Awful lot of weight in the SwStrk%, as if that’s the one Ring to rule them all. See: “Moyer, Jamie”, never had a SwStrk% over 9.8% his entire career. Herein lies the limit of FIP, xFIP and SIERA, because Moyer’s were atrocious every year.
Stripling’s problem recently has been that he’s given up a lot more fly balls than he was prior. Up in last 8 starts 9% from his first 6 real starts(not counting that 1st spot start). You give up more fly balls, you’re going to give up more homers(which he has done).
With how he’s done recently and the fact the Dodgers have 6 starters now(not to mention Ryu is close to being back)- the bigger problem could be that he’s no lock to continue to be starting regularly. I mean Stripling could lose a start as soon as this Sunday when his turn is scheduled to be up- Dave Roberts didn’t commit to him starting that yesterday when discussing the Braves 4 game series.
Will these very surprising performances on the far right of the bell curve continue?
Short answer… no…
I pretty much agree with the assessments. The point I’d like to make is regarding the batted ball profile. The average P has roughly a 22% K rate and a 8% BB rate. That leaves 70% to Batted Ball Events (BBE’s). As far as I can tell, all of the xERA models regress this 70% to the mean, which is a lot of data to just regress to the 50th percentile. I know that BBE’s are more volatile for P’s than hitters, but the kind of BBE’s that P’s are giving up should play a factor. Are there any current xERA models that incorporate BBE’s for the individual P’s, rather than just regressing all BBE’s to the 50th percentile (I realize FB’s play a role)? I’ve done a deep dive on this and developed one. Some interesting stuff has emerged. For instance, the time of year and which year play a huge role. For example, last year Barrels resulted in a HR about 62% of the time over the whole year. But, it was, off the top of my head, like 55% in April, 58% in May, 66% in June, 64% in July, 68% in August and 61% in September. This year, as a whole, it’s been more like 50%, and there’s only been a slight uptick since June 1, about 52%. This year has been the most forgiving for P’s while 2017 was the least forgiving. This is likely why some of the guys like Mikolas and Porcello and Gibons, etc., are outperforming their xERA’s, especially if those models are based on the last 3 years (’15, ’16 & ’17) outcomes. With the knowledge that the pitching landscape is more forgiving in the first two months of the year I went with a 7SP, 2RP and a deployment of almost all 2 start SP’s in March, April and May strategy and then dealt for a 4th RP in late May and have gone with a 5 SP, 4RP and deployment of only good match-ups since early June. It’s worked well so far. Just some food for thought I guess.
Where did you get the idea that xERA metrics regress batted ball events? FIP uses homers allowed and completely excludes other batted balls, xFIP does normalize homers to a league average HR/FB rate but excludes all other batted balls like FIP, and SIERA reflects actual batted ball rates and doesn’t regress anything.
A podcast I had listened to in the last couple of weeks. Though, I did “misspeak”. It was HR’s that the pod mentioned regressing. My bad. However, shouldn’t a xERA model incorporate the BBE’s? If we know what the xR’s are for a BB, a 1B, 2B, 3B, HR and we have the data to project the xBBE’s based on launch and exit velocity, it seems like there would be a xERA model based on the BBE’s that the individual P is giving up. I’ve put one together and it matches up very well year to year, I just didn’t know if there was one out there currently being used
SIERA does use batted ball rates, but since it was developed years ago, doesn’t incorporate any Statcast metrics.
Thanks! I wasn’t trying to troll. Just curious. Thanks for the work
Uhhh Kyle Gibson??!