What Is Up With the Cubs Rotation?
After getting swept by the New York Yankees and losing two out of three to the Colorado Rockies, the Chicago Cubs are now sitting on a .500 record. By run differential or BaseRuns expected performance, tools that strip out sequencing, the answer is the same; the Cubs have played pretty much like a .500 team. The offense has been a little worse than average, the pitching a little better than average, but overall, the team has played roughly like an 81 win team so far.
Of course, this isn’t what recent Cubs teams have played like.
In 2015 and 2016, Cubs starters topped all of Major League Baseball with 36.3 WAR, just ahead of the Nationals, Mets, and Dodgers. The starters’ 3.50 FIP, without even considering the impact of defense or the potential of inducing weak contact, has been the best in the majors. The Cubs have returned almost all of last year’s staff intact, with Jon Lester, Jake Arrieta, Kyle Hendricks present for each of the past two seasons and John Lackey around last year.
The question we are trying to answer here is what is responsible for the downturn in results. We are just 34 games into the season, so we could chalk it up to luck. We could try to determine if the talent level has changed in any way, which might cause us to lower our expectations, and we could point to some outside factors that aren’t luck, but aren’t necessarily the responsibility of the pitchers, like defense. To frame our understanding of what is going on, it probably helps to create some expectations of what we would expect to see from a Cubs rotation this season. While Brett Anderson has made six starts, he’s only pitched 12% of the Cubs starter innings, and little was expected of him, so we will focus on the four returnees.
First, let’s take defense or inducing weak contact out of the question. Below we see the FIP for each pitcher from last season, this season, and our Depth Chart projections at the beginning of the season.
| 2016 FIP | 2017 Proj FIP | 2017 FIP | |
|---|---|---|---|
| Jake Arrieta | 3.52 | 3.38 | 3.78 |
| Kyle Hendricks | 3.20 | 3.57 | 4.24 |
| John Lackey | 3.81 | 3.87 | 4.17 |
| Jon Lester | 3.41 | 3.38 | 3.58 |
| AVERAGE | 3.49 | 3.55 | 3.94 |
Without accounting for defense, it looks like the Cubs are pitching a bit worse than they did last season. Part of the problem is the Cubs don’t have any pitchers over-performing at the moment, though when all of the pitchers have good projections, that can be difficult. Now let’s compare the above to ERA.
| 2016 ERA | 2017 Proj ERA | 2017 ERA | |
|---|---|---|---|
| Jake Arrieta | 3.10 | 3.21 | 5.35 |
| Kyle Hendricks | 2.15 | 3.40 | 3.40 |
| John Lackey | 3.35 | 3.74 | 4.29 |
| Jon Lester | 2.44 | 3.25 | 3.27 |
| AVERAGE | 2.76 | 3.40 | 4.08 |
So this table makes it look like the Cubs starters have gone way off the deep end. Here are some of the numbers underlying those changes with numbers from this season and last.
| Season | Name | K/9 | BB/9 | HR/9 | BABIP | LOB% | HR/FB |
|---|---|---|---|---|---|---|---|
| 2016 | Jon Lester | 8.8 | 2.3 | 0.9 | .256 | 84.9 % | 12.2 % |
| 2017 | Jon Lester | 8.7 | 3.5 | 0.9 | .316 | 75.9 % | 11.1 % |
| 2016 | Jake Arrieta | 8.7 | 3.5 | 0.7 | .241 | 74.9 % | 11.1 % |
| 2017 | Jake Arrieta | 10.2 | 2.8 | 1.4 | .364 | 58.5 % | 14.3 % |
| 2016 | John Lackey | 8.6 | 2.5 | 1.1 | .255 | 76.2 % | 12.9 % |
| 2017 | John Lackey | 9.9 | 2.4 | 1.7 | .310 | 73.7 % | 19.0 % |
| 2016 | Kyle Hendricks | 8.1 | 2.1 | 0.7 | .252 | 81.4 % | 9.6 % |
| 2017 | Kyle Hendricks | 7.7 | 3.6 | 1.1 | .255 | 78.6 % | 15.6 % |
Lester is about the same. Lackey seems to be pitching better except for those homers. Arrieta’s strikeout and walk numbers are better, but the homers have gone way up mainly because he has turned into a flyball pitcher in the early going this season. Hendricks’ walks have gone up as have his homers. Both Hendricks and Arrieta have seen fairly sizeable dips in their velocity in the early part of the season which could explain the increase in homers. One part of last season that has not carried over from last year to this point has been the historically low BABIP that the Cubs had as a staff last season. Only Hendricks is duplicating the BABIP from last season with every other member of the rotation seeing a 50-point or more change.
Last year, I posited that much of the Cubs’ low BABIP numbers were due to an incredible defense. Thanks to more searches on Statcast, we can come a little bit closer in that argument. Here are the Cubs’ pitchers xwOBA numbers compared to their actual wOBA from 2016 and their xBA compared to their actual BA in 2016.
| Player | xwOBA, wOBA | Difference | xBA – BA | Difference |
|---|---|---|---|---|
| John Lackey | 0.312 – 0.281 | 0.031 | 0.246 – 0.218 | 0.028 |
| Jake Arrieta | 0.293 – 0.264 | 0.029 | 0.224 – 0.194 | 0.030 |
| Jon Lester | 0.278 – 0.266 | 0.012 | 0.223 – 0.211 | 0.012 |
| Kyle Hendricks | 0.262 – 0.258 | 0.004 | 0.214 – 0.207 | 0.007 |
That xWOBA number accounts for any ability the Cubs pitchers might have to induce weak contact, and all but Hendricks still got results much better than expected. This is the impact of defense and luck. Here are the numbers this season.
| Player | xwOBA, wOBA | Difference | xBA – BA | Difference |
|---|---|---|---|---|
| Kyle Hendricks | 0.337 – 0.296 | 0.041 | 0.279 – 0.231 | 0.048 |
| John Lackey | 0.329 – 0.331 | -0.002 | 0.263 – 0.265 | -0.002 |
| Jon Lester | 0.280 – 0.325 | -0.045 | 0.207 – 0.255 | -0.048 |
| Jake Arrieta | 0.285 – 0.353 | -0.068 | 0.230 – 0.290 | -0.060 |
The Cubs are not getting that bump they received last season from their defense or some good fortune. In the outfield, the team switched out Dexter Fowler and a combination of players led by Kris Byant for Kyle Schwarber and a combo led by Albert Almora and Jon Jay. Mike Petriello took a look at those switches early on this season and they have not proved beneficial in the early going. Almora should prove to be better than Fowler over the course of the season, but Schwarber is a pretty big downgrade, and they might not get the same production from Jason Heyward, as he currently sits on the disabled list.
Last season created unreasonable expectations for the Cubs in terms of run prevention. The starters are pitching slightly worse on the whole, with the velocity drops of Hendricks and Arrieta the most concerning. The returning starters are pitching a little below expectations, but just one-fifth of the way through the season the numbers aren’t unreasonable. FanGraphs still has the Cubs as the sixth best rotation in baseball, and if Arrieta and Hendricks can figure things out, there’s no reason to think the Cubs won’t be very good going forward. Even with the changes in the outfield, the team should still be solid defensively, so there is room for improvement when it comes to BABIP and the related left-on-base numbers.
Yes, the Cubs’ rotation is underperforming, but it isn’t doing that poorly. Even an underperforming Cubs rotation with their lineup and defense should be enough to win 90 games, and 95 wins is still a reasonable possibility if the team pitches just a little bit better.
Craig Edwards can be found on twitter @craigjedwards.
I’m not sure where the defense is significantly worse–is it CF? Almora is a noted defensive wiz, Jay is serviceable, so the Fowler replacements seem like a wash. Is it all Schwarber in LF? Bryant logged quite a few LF innings last year, but Soler (poor fielder) logged the most. I don’t have an answer, but it is interesting (as you noted) that last year’s starters owned a .252 BABIP against (historically great), this year’s starters own a .324 (worst in MLB). That 70 point gap seems massive.
To note: Over the last three years, Arrieta’s groundball % was 49, 56, and 53. This year it is 39%. Like the chart shows, his LOB% is 59%!
I definitely think the answer to the Cubs defensive babib is mostly just regression to the mean. If they had kept every player from last years team it was unreasonable to expect they’d put up the same kind of numbers, so it’s not really surprising when a very similarly talented defense personnel wise doesn’t repeat the same results.
A babip that low historically just doesn’t carry over automatically, and that’s a big reason the computers saw the Cubs missing 100 wins, babip luck/regression. Good news for the Cubs of course is that they are still absolutely loaded, so they’ve got that going for them.
Right, I totally agree that one cannot expect BABIP to carry over, but a .324 team BABIP is way too high for a team this defensively sound.
Might be thwarted by crosstown rivalry politics, but I keep thinking a trade of Jose Quintana to the Cubs for something built around Happ or Candeleria make sense.
The deal could also be expanded to include Robertson if the Cubs need relief help and don’t mind taking on the salary (or the White Sox kick in some cash to make it work).
Yeah I don’t think Happ is enough for Quintana, it would take Eloy + Happ, I think. Quintana has 3.5 years left on that contract, so it seems like it would be a pretty big haul.
I agree it’s not enough, which is why I wrote “built around”…meaning the deal would include other stuff going to the White Sox.
Yup, I saw that, but I don’t think it’s an “or” situation, it is an “and” situation. Eloy and Happ, not Eloy or Happ.
Just looking at raw fielding numbers for the OF looks to me like a lot less range in both LF and CF. Roughly 1 play less every 2 games or so (attribute roughly 2/3 to LF and 1/3 to CF). Given that this is OF that means 1 extra ‘double’ (out turns into double or single into triple) every other game.
A little amusing that http://www.fangraphs.com/blogs/whats-going-on-with-the-cubs/, talking about how Cubs starters may be playing with dialing it down a notch, is followed about three weeks later with an article talking about how all the Cubs’ starters are struggling.
I was actually hoping this article was going to be an update of that one. Here, they only mention velocity in the context that Hendricks and Arrieta’s are down slightly. And if that’s the case, then I would think they’ve all rebounded from that original article, but it doesn’t say. Looking at each pitcher’s game log, their velocities haven’t seem to have increased at all since 3 weeks ago, so I would have thought that would come up here.
Re-posted from my post to Jeff Zimmerman’s article from May 8th. Numbers are now a few days old but probably not much different:
——————-
Here are the Cubs starters’ average velo by home/away per Fangraphs
NOTE: this is not a true average. I am simply averaging the game avg velo so if they threw more pitches in 1 outing than another, I wouldn’t get that with this simple method. I suspect it would have no bearing on the outcome.
NOTE 2: these are all 4 seam fastball numbers
Away Home
Jon Lester 91.13 91.25
Jake Arrieta 92.03 91.85
Kyle Hendricks 85.97 85.10
John Lackey 91.13 90.23
Not much difference but a little bit for Hendricks and Lackey. Decided to head over to Brooks Baseball and see what it says. Here it is at Brooks
Brooks
Away Home
Jon Lester 92.02 91.53
Jake Arrieta 92.80 91.88
Kyle Hendricks 86.99 85.23
John Lackey 91.91 90.14
Now, that’s a MUCH bigger difference. Lackey and Hendricks are both off by almost 2 MPH at home v road games.
Decided to pull every starter at Wrigley this year and compare Wrigley with everywhere else and see what happens
NOTE: This is at Brooks
NOTE: Except for Pineda, these are all 4 seam FB. Pineda throws a cutter per Brooks so I used that
NOTE: As with above, this is an avg of the game averages so not a true average.
NOTE: The asterisks are guys whose season low velo was at Wrigley
NOTE: Yes, the spacing probably sucks. Deal with it.
Everywhere else Wrigley
Alex wood Omit as he started in the BP and this was his 1st start and he only went 3.2 IP
Brandon Mccarthy 94.10 93.72
Hyun Jin Ryu 90.41 88.66 *
Gerrit Cole 97.05 95.67 *
Tyler Glasnow 94.42 94.22
Jameson Taillon 95.63 94.89
Chase Anderson 93.03 92.57 *
Jimmy Nelson 94.99 94.08 *
Tommy Milone Omit – only made 3 abbreviated starts and then a couple relief appearances and then was put on waivers and claimed by the Mets where his DL stint will start any day now
Vincent Velasquez 94.76 94.64
Jeremy Hellickson 90.27 89.46
Jerad Eickhoff 91.59 90.80 *
Zach Eflin 93.58 92.26 *
Michael Pineda 94.96 91.21 *
Jordan Montgomery 92.08 91.75
Luis Severino 97.35 98.27
SO Severino is the only guy out of 14 that threw harder than his average everywhere else. At Wrigley, most were well below their season average everywhere else. A full 7 of the 14 had their season low velo at Wrigley.
I’m suspecting a calibration error in the Statcast system at Wrigley myself.
————————–
Since I posted that on May 8th, the Cubs played 3 games in Coors (velo per Brooks):
Arrieta 92.67
Lackey 91.75
Hendricks 86.72
I still say it looks Wrigley’s radar is off by a couple MPH
BTW, PIneda has been consistently 94-95 all year. Then he was 91 last start in Wrigley. He’s 94 again in the 1st inning tonight in the Bronx.
Tango has shown (you can check his Twitter feed) that the park factor for Wrigley Field wrt pitch speed is around average. So, no, it does not appear that the Statcast speed results at Wrigley are off much at all.
If you or anyone else wants, use Statcast data and look at similar pitch classifications (recommended to use fastballs), for pitchers at Wrigley and other parks. Use a “delta method” to make sure you have the same pitchers home and away. Weight by each pitcher’s # pitches, preferably using the smaller of the two values, home and away.
I’m confident that you will find the same thing as Tango – that there is NO significant (more than .3 mph) park effect at Wrigley or any other park (he did all the parks).
It is so hard to “undo” false information in this modern world of very fast and very broad information-sharing. Let’s please not spread the fiction that, “Wrigley’s radar” (Statcast) is off by a couple miles an hour” any more than it has already been spread.
Jeff: Hey Craig, you read my article?
Craig: What article?
You know you’ve entered The Darkest Timeline when you’re happy that the Cubs were swept by the Yankees.
That Cubs victory party had quite the hangover…
I remember after the 2002 WS, the Angels made almost no moves because, hey, why mess with success? They dropped from 99 to 75 wins, with their aging rotation collapsing.
The Cubs didn’t make many moves because their core position players are mostly young, and getting Schwarber back was almost like an acquisition. While some of their bats have disappointed, most of the damage has been done by their rotation aging out. I’m sure Lester and Arrieta still have stuff in the tank, but maybe no longer as #1s. What’s funny is that while Lackey was the youthful ray of hope on those Angels teams, he is now clearly the “past his prime” journeyman for the Cubs that Appier and Sele were playing. I wish he looked more like Gary Busey.
I’m going to go with a recurrence of fried chicken and beer syndrome by John Lackey and crew.
world series hangover + postseason innings
While I am a Cubs fan and must admit to a spot of worry, I do find this year’s NL to be way more interesting (as a baseball fan) than it was projected to be. With the surprise Rockies and Dbacks pushing the Dodgers in what was supposed to be a walkover NL West, combined with the Cards recovering from a poor start and the Reds and Brewers both looking much better than preseason in the NL Central, there are some actual division races! The NL East seems almost already decided, but it was supposed to be all three divisions decided by 10 games or something boringly one sided like that. Glad we get some fun races (so far).
The AL has had a couple of surprises, but the West and Central look already decided. Only the AL East looks like it will feature a race.
I, for one, hope that all six divisions are decided by a game 163… maybe even a 164 with tie breaker scenarios!
Don’t you worry about the NL Central–they’ll deliver. The Reds-Brewers divisional championship battle is going right down to the wire.
I hardly dare to say it but could age be an issue? Cubs hitters are really young but lester is 33, lackey is 38, arrieta is 31, only Hendricks is still young.
I’m not saying they are done (especially lester and arrieta) but they might be slowing down a little.
I think the rotation will be solid but maybe not great anymore. But the cubs hitting will come around, I think they will be even a little better than last year with schwarber, a more mature russell and baez and contreras instead of ross.
That should offset some of the pitching loss and make them a mid 90s team again which will easily win that still weak division.
But next year the cubs will make one or two big rotation moves (top guys not back end fillers like anderson).
Bunch of fluky ass contact. Keep running K9 of 10+ and bb9 under 3 (like Arrieta and Lackey have) and at the end of the season your numbers should look pretty good. Truth be told, they didn’t get trounced by the Yankees. The first game they blew the save and lost by a run, the second game Brett Anderson pitched but if you started the game after he pitched, the Cubs outscored the Yankees (lol, kidding but true), and the third game was 2 games in 1 in which the Cubs held the Yankees scoreless for like 10 innings. So, 2 one-run losses, and another in which your garbage #5 couldn’t get out of the 1st inning.
Cubs have a bigger problem and that’s that 4 of their everyday players are hitting like complete crap. And they can’t hit for shit with runners in scoring position lately. The series in Colorado was a total joke. No hit by German freakin Marquez for 6 innings. They looked completely unprepared against him as if they hadn’t watched a single bit of video. Threw away the game and series.
Just a reminder, Jason Vargas has been the best pitcher in baseball lol. Its still early.
False! Carrasco is the best pitcher followed closely by Sale.
I upvoted you before I realized you were saying Carrasco was *better* than Sale. I want my vote back.
I meant through the season so far.
It’s still not true though. Carrasco has a slightly smaller walk rate and a miniscule 0.06 lead in ERA but Sale wins everywhere else.
“…though when all of the pitchers have good projections, that can be difficult.”
If the projections are reasonably accurate, then that is mostly false and is an important point to make. It should be (nearly) equally likely for a pitcher or group of pitchers to exceed or fall short of a reasonably good projection, whether that projection is good or bad. That’s the whole point of a projection. All of the regressions toward the mean have already been done.
There are some selective sampling, survivorship, or even injury (e.g., a pitcher can get injured and pitch worse than his projection, but it is not clear whether that is included in the projection) issues which make this a little complex, and mean projections are not necessarily the same as median projections, but that doesn’t change much the point I am making.
The lack of defensive flexibility shown by Joe Maddon this year is curious, considering it was the Cubs’ strength last season. Ben Zobrist is just a bad outfielder at this point in his career, yet Maddon’s “primary” lineup is when Bryant, Zobrist and Baez are all in the lineup with Zobrist in RF. I’d argue RF is the most important/difficult defensive outfield position in Wrigley due to the wind, sun and awkward dimensions.
Zobrist has already played 94.1 innings in RF after 158.2 all of last year (and just 34 the year prior). He cannot play right field. He just doesn’t have the athleticism or instincts to do it anymore, not to mention his complete lack of arm strength in the OF spot that requires the most arm strength.
Recognizing that, would it not make more sense to play Bryant in the OF, Baez at 3B and Zobrist at 2B if you really want all three in the lineup together? Bryant is far better than Zobrist in the OF, Baez is likely far better than Bryant at third and Zobrist is less bad at 2B than he is in the OF.
I said after last season that if the Cubs let Fowler go, the most obvious replacement for him in CF is actually Kris Bryant and I stand by that. If the Cubs can make Fowler into an above average CF, they can make a similarly sized and equally athletic Kris Bryant into an above average CF, which allows Heyward to play the tougher field in Wrigley with his elite arm strength and Zobrist to play his “least awful” position. Obviously with Heyward out, or on days with Baez and/or Zobrist out of the lineup, everyone goes back to their native position and Jay/Almora play the OF.
Even if you didn’t want to go *that* crazy (and I admit it is crazy even for Joe Maddon), at the very least putting Bryant in RF, Baez at 3B and Zobrist at 2B solves a lot of problems with the current defense while letting you keep your top 8 hitters in the lineup. Moving guys around to squeeze the most out of them as a unit is why the Cubs were so good defensively, it wasn’t because they had 8 great defenders on the field every day. So far, that flexibility isn’t showing up from Maddon.
The problem is they aren’t ready to bench Schwarber. I agree, that is the best alignment they can put out there but they’re looking ready to sink or swim with Schwarber in LF.
Zobrist absolutely should not see any more time at RF. Total liability.
Schwarber is a bad defender obviously, but his defense alone isn’t killing them. Schwarber being a bad defender, with Ben Zobrist also being a bad defender in the other corner and Jason Heyward playing out of position but still very good defense in CF certainly is. Heyward or Almora can only do so much, then if you put Jay out there, yikes.
The Cubs can maximize Schwarber by pulling him late in games for a defensive sub, that’s the best they can do there. You can’t bench him. Now, Schwarber playing awful defense while also hitting poorly in the leadoff position with the pitcher hitting 8th is another problem all together and one that needs to be addressed.
The best way the Cubs could align their preferred 8 is:
Zobrist 2B
Bryant RF
Rizzo 1B
Schwarber LF
Contreras C
Baez 3B
Heyward CF
Russell SS
Pitcher
I’d personally try Bryant in CF and Heyward in RF but it isn’t vital. In certain parks like PNC, swap Bryant and Schwarber so Bryant has the tougher defensive field of the two. Or even just sit Schwarber in a park like that and put your best defensive team on the field for a guy like Lester or Hendricks who really needs the defense.
It feels like Maddon has just been feeling things out for a bit to see what works and obviously this really isn’t working. The Schwarber at leadoff experiment has probably run its course, the trickle down effect on the lineup has been ugly. Hopefully he goes back to some of the mad scientist experimenting that was the calling card of the 16 team.
Perhaps the biggest factor not mentioned yet is added pressure, from various sources:
— Decreased run support. Last year the Cubs bashed away out of the gate, averaging 6.0 runs in their first 34 games, vs. 4.9 so far this year. It’s easier to pitch with a big lead.
— A clearly diminished Arrieta. The Cubs won 23 straight Arrieta starts from 2015 into 2016, but since late June last year he’s been a below average SP. In his last 28 starts (including postseason), Arrieta has a 13.6% SO-BB% (vs. 21.6% in 2015), and a 4.95 RA/9, serving 22 HRs in 165 IP. The Cubs split those 28 games. Losing that “automatic win” adds pressure on everyone else.
— Widespread expectations of a dynasty. Not saying the players bought into that, but there’s a sizable chunk of the fan base that seemed to start counting up future titles before last year’s was fully savored, and I can’t help but think that’s affected the Wrigley atmosphere. The Cubs are on roughly the same road pace as last year, but they’re 7-9 at home, vs. 57-24 last year.
538 delved into this issue today as well: https://fivethirtyeight.com/features/have-the-cubs-forgotten-how-to-field/
I think the issue is more to do with the Cubs’ fielding (which was otherworldly last year and helped suppress their team .BABIP allowed to historically low levels) and is regressing back to more normalized levels this year, some due to trades (Heyward) or different players playing (Schwarber) and some due to the same players taking a step back from their amazing defensive performances last season.