The Cubs Have a #3 Starter
The way Jake Arrieta has thrown the ball this year — especially in the second half — has ended any discussion about who the Cubs #1 starter is. Arrieta has propelled himself into the discussion of true aces, and he’ll be the guy the Cubs put on the hill with their season on the line next Wednesday. With Jon Lester slotting in to the #2 spot, the Cubs top two starters should be able to hold their own against any other staff in baseball.
After that, though, things get a little more interesting. When asked who his third starter in the postseason might be, Joe Maddon stated simply “I don’t know.” Jason Hammel began the year as the team’s third starter, and his overall numbers are quite good for a middle-of-the-rotation guy, but those numbers are based on an excellent first half and a pretty lousy second half. Prior to the All-Star break, opposing hitters posted just a .261 wOBA against Hammel, but since the break, they’ve put up a .371 mark against him. The problems may be tied to a hamstring injury that landed him on the DL in July, and unless he really shows them something in the last week of the season, it’s not clear that the Cubs can trust that he’s healthy enough to be effective in high-leverage postseason innings.
Hammel’s recent ineffectiveness have given rise to the idea that the Cubs should be open to throwing bullpen games when Arrieta and Lester aren’t on the hill. With former starters Travis Wood and Trevor Cahill both providing some good performances in middle relief, the Cubs probably could simply tell Hammel to go max-effort to two or three innings, then bring in the left-handed Wood for a couple of innings, and go back to Cahill as the bridge to the late-inning relief corps. Such a plan would mostly limit the times through the order penalty, and would give the team multiple chances to pinch-hit for the pitcher’s spot, increasing their expected offensive output as well. Given the high-stakes nature of the postseason and the frequency of off-days, these bullpen games are almost certainly a better idea than asking a questionably-healthy starter to give you five or six innings in a crucial game.
But while that may very well be the best plan to get through the games started by Hammel, I don’t think the Cubs need to go that extreme with their third starter spot. Instead, I think they have a perfectly acceptable #3 starter, and they should give Kyle Hendricks a chance to prove it.
Given the sheer number of dominating power arms heading for October, it can be easy to overlook Hendricks as a weak link. From a stuff perspective, he’s never going to impress, with a fastball that sits around 89 and no big breaking ball as an obvious out-pitch. Hendricks is the classic soft-tossing command guy who has succeeded by throwing strikes and fooling hitters with an excellent change-up, and these pitchers are chronically underrated compared to guys who get the same results but throw harder.
But at the end of the day, results are what matter, and Hendricks’ results suggest that he’s an above-average big league pitcher. Here are his career numbers since getting to the majors last summer.
| IP | BB% | K% | GB% | HR/FB | BABIP | ERA- | FIP- | xFIP- |
| 254 | 6% | 20% | 50% | 10% | 0.290 | 92 | 89 | 92 |
That’s the classic good command-pitcher line; an average strikeout rate buoyed by very good walk and groundball rates. When you pound the strike zone but can still get groundballs and some strikeouts, you can succeed in the big leagues, even without premium stuff. Sonny Gray has basically ridden this brand of results — plus some BABIP prevention — to All-Star status, though his stuff is clearly several ticks better than Hendricks. But he’s a good example of what can be done with lots of strikes, groundballs, and a roughly average strikeout rate.
More realistically, a better comparison for Hendricks would be a guy like Mike Leake, only Hendricks has done a better job of keeping the ball in the yard during his time in the big leagues, and he’s missed more bats as well. And with some recent changes, he might be trending even more towards an increasing strikeout rate that could allow him to really establish himself as a quality mid-rotation starter.
When Hendricks got to the big leagues, he mixed five pitches; he mostly used his two-seam fastball and change-up, and then worked in a cut fastball as his third pitch, with a curveball and four-seam fastball as his less-often used secondary pitches. And because he pitched very well in his rookie season last year, he carried that same repertoire over to 2015, but some early struggles (by ERA, at least) have seemingly convinced Hendricks to adjust his pitch-mix. Most notably, he’s ditched the cutter.
And probably for good reason. On the season, his cut fastball was called a ball more often than any of his other pitches, generated fewer swings and swinging strikes than anything else he threw, and had the highest home run rate of any of his offerings. By pitch-type linear weights, his cutter was 3 runs below average per 100 pitches, the only one of his pitches that got below average results.
To replace the cutter, he’s upped his usage on both his change-up and his four-seam fastball, and the early results are promising. Here are Hendricks’ numbers since August 1st, since he began life without his cut fastball.
| IP | BB% | K% | GB% | HR/FB | BABIP | ERA- | FIP- | xFIP- |
| 58 | 8% | 25% | 54% | 18% | 0.301 | 118 | 90 | 78 |
And here are his numbers since the start of September, when he’s really pushed the changes a bit more aggressively, pushing his change-up usage to the highest rate he’s posted in any month since getting to the big leagues.
| IP | BB% | K% | GB% | HR/FB | BABIP | ERA- | FIP- | xFIP- |
| 27 | 6% | 30% | 49% | 20% | 0.258 | 94 | 78 | 63 |
The August ERA was a bit inflated due to the elevated home run rate, but there’s no real reason to think that Hendricks has suddenly become a guy who is going to struggle with the long ball; HR/FB rate over two months is essentially meaningless. But strikeout rate, on the other hand, is a bit more stable, and Hendricks’ strikeout rate has been spiking up over the last two months. He’s struck out an ace-like 30% of the batters he’s faced in September, and over his three most recent starts, it’s 37%. Last night, he struck out 41% of the batters he faced, made more impressive by the fact that he was facing the Royals, who are striking out at an historically low rate this season.
Certainly, you don’t want to overreact to recent data, and it’s always better to keep an eye on the bigger picture. But even in the big picture, Kyle Hendricks has shown himself to be an above average big league starter, and more recently, he’s started making guys swing and miss with some regularity. By dumping his cutter and focusing more on his very good change-up, Hendricks may be finding a way to go from a pitch-to-contact strike-thrower into being a guy who can really take over a game when need be.
Perhaps instead of Mike Leake, perhaps Hendricks is heading down the Doug Fister path. And despite never really throwing that hard, Fister was a premier postseason performer for the Tigers once they gave him a chance to show what he could do on the big stage. Hendricks won’t overwhelm anyone with his fastball, but if you get past velocity, there’s no real reason to worry about the Cubs #3 starter. Kyle Hendricks is worth trusting.
Dave is the Managing Editor of FanGraphs.
screw you guys!
Question about the times through the order penalty – is it due to diminished stuff, adjustments made by hitters, or a combination of both? Or do we not really know the reason behind the penalty?
All of the above, really.
It appears that the third-time-through-the-order penalty may be largely if not entirely a myth, created by the fact that starting pitchers often do not face the bottom 2-4 lineup slots (i.e., the weakest-hitting slots) when going through the order for the third time.
For a breakdown of the numbers, see the final remarks in the comment thread after Dave’s chat of last Wednesday.
This is false. The TTO penalty is well-documented by people like MGL who have done extensive research on it. That those conclusions don’t conform to your preconceived notions doesn’t make it a myth.
If you don’t believe Dave, I am happy to tell you why you are wrong. If you are buying the beer.
Have you ever noticed how Rick Sutcliffe NEVER says anything controversial or negative? He’s horrible.
I think it’s pretty safe to say he gets the start. It’s also safe to say he will have a short leash. If runners get on base and they need a ground ball, it’s Cahill or the lefty Richard. If they need a K it’s Grimm or the lefty Wood.
The back end is so much stronger with Neil Ramirez, man is he good. Rondon is also excellent. I don’t personally trust Strop or Rodney, but Joe Maddon does.
They don’t have the total shutdown lefty, but the pen is not a weakness.
I would imagine Maddon is one of the most likely managers to go with the bullpen game strategy, in addition to one of the most likely managers to not want to talk about his pitching schedule.
Question on up/down voting comments, are the numbers that actually appear only the votes from fangraphs authors or something? It doesn’t seem like other sites where you can see the effect your vote has on a comment.
When it gets to +/- 5, then it shows.
All of my comments have 4 upvotes
Not Anymore
Nitpicking here…but IRC, Hammel didn’t go on the DL in July…he SHOULD HAVE, but didn’t and that is probably why we’ve seen his 2nd half struggles…the game he tweaked his hammy is a pivot point in his season….I’d thank him for his contributions this season and shut him down – he can’t be trusted the rest of the way.
On the other hand, he was similarly excellent in the first half last year and then struggled in much the same way in the second half. Many believed it was related to the trade to the A’s, especially in light of Hammel’s comments that he was not comfortable making the transition, but this makes it look like he can’t handle the full season of work.
It’s better if it is just the injury and not fatigue, but the Cubs should still make some effort to limit his innings or even delay the start of his season next year. I’d say skip starts rather than limit pitches, since going deep into games when he’s on is really valuable.
There’s a meme that Hammel struggled in the second half of 2014: he really didn’t. He was bad in July, had a good results and bad advanced stats in August, and was absolutely lights out in September when the “full season of work” would theoretically be getting to him.
Check out the splits here.
So, you’re saying Hammel won’t pitch in the postseason. I read ya.
Perhaps I misread the article, but is this an article that quoted:
1) K% over 3 starts?
2) Followed by K% over 1 start?
And then slap in a disclaimer don’t overreact to recent data, to make this sort of practice OK?
I get the point of the article, but when crazy stuff like this is added in it reads a bit like someone trying too hard to prop up an argument. I’m not sure putting in small sample size disclaimers should enable the questionable use of some of the #’s quoted
I think the single start example was only to point out that he just earned a lot of strikeouts against a team very difficult to strike out, which would prevent a quality-of-competition counterargument. And the small sample size stuff is iffy but strikeout rate does stabilize quickly and recent history is important when discussing the effects of a recent change, since Hendricks is in theory a different pitcher of late.
“the effects of a recent change, since Hendricks is in theory a different pitcher of late”
Doesn’t it stand to reason that teams at least sometimes (mabye often) adapt to changes in pitch usage over time due to the effects of advanced scouting? I agree that positive initial success from a change in pitch usage is good, but I could also see a default assumption that the league will adapt to it.
I think I can get on board with this. After Arrieta wins the WC game, the NLDS rotation will be Lester, Kendricks, Arrieta… now the question is, should Cubs go Hammels, Lester for the last two or Lester, Arrieta on short rest?
If Cubs face the Cards in the NLDS, I think a good chance they sweep. Lester and Arrieta are head and shoulders above any starter the Cards have, and Kendricks is just as good as any starter the Cards have. Cubs have the better offense. Better Baseruns.
Really playing the Pirates in the wild card and the Cards in the NLDS is the best thing that could happen for the Cubs. Pirates have much better rotation and team for a long series, but they can’t compete with Arrieta in a one-game do or die. Cards can’t complete in a long series, so this works out well.
Man, should the Cubs even have to play the games or should they just get a free pass to the NLCS? I agree that the Cubs, on paper, are probably slightly better than the Cardinals, but let’s see how the rookies play in the postseason pressure before predicting a 4 game winning streak against Pit and STL. I think the NL Central NLDS will probably be a 5 game series no matter who the teams are. Anyone can beat anyone in the postseason, so predicting a sweep based off of supposed slight advantage is foolish.
Slightly better? Cubs better by BaseRuns and that’s before you factor that the Cardinals best pitcher this year just got shutdown for the rest of the season and Molina is shut down for the season. You really think the Tony Cruz led Cardinals with four starters, two of have been terrible for a while now and two of whom weren’t that good to begin with, stand a chance?
Not to mention Piscotty out, Holliday and Grichuk not close to full strength, Peralta terrible now.
It will be a miracle if this team doesn’t get swept in the first round.
“It will be a miracle if this team doesn’t get swept in the first round.”
Are the Cards not at full strength? Undoubtedly. Are the Cubs (and Pirates, for that matter) a strong wild card team? Absolutely.
But when you consider the very best teams in baseball are generally playing around .620 ball over the course of a season, and the very worst are generally playing about .380 ball over the course of a season (and the Cards, even with their injuries, are far from a true talent .380 team), a .620 team would be expected to sweep three games from a .380 team only about 24% of the time.
Long way of saying I will place any amount of money that you’re comfortable with on whether the Cards will get swept in the first round.
The Cubs are a whopping 2 games better than the Cardinals by BaseRuns: 89-68 versus 87-70, so BaseRuns indicates that these teams are basically evenly matched.
Also, Molina might play in the playoffs. And to say that Garcia and Lackey “weren’t that good to begin with” is just plain dumb and wrong.
To be fair, he didn’t say the Cubs WILL sweep, he said there’s a good chance of it. And his logic isn’t terrible. Cardinals starters have struggled recently – particularly against the Cubs, with the exception of Martinez, who won’t be pitching. I definitely give advantage to the Cubs in that series but a sweep is not likely, of course.
“It will be a miracle if this team doesn’t get swept in the first round.”
Considering you clearly the biggest Cubs homer the world has to offer, you should really learn Jason Hammel’s actual name.
As a Cubs fan myself, I was totally thinking the same thing myself. And I totally think they’re going to lose a game or three on their way to the WS…
and I clearly like “myself”…
You totally do.
Hey Spencer, sorry I made a typeo. Did you have to make a post about it? Would like to address my actual arguments instead of making personal attacks? Ok, thanks.
And Hendricks’
You sound like you’re trying to convince yourself more than anything else.
Hammels? Kendricks? Know the players on the team you’re rooting for. I’m a huge Cubs fan too, but come on.
Sorry for the typeo’s you morons. Now try focusing on the actual arguments I made, ok?
See, here’s the thing, and this is true for whenever you’re writing anything, not just a FanGraphs comment: When your comment is littered with grammatical errors and typos, it makes it very difficult to take the actual content seriously. And considering that you made those typos repeatedly, it just tells me that you simply don’t know how to spell the names of the players on your own team. So why should we take you seriously?
You seem like a real jerk
Good thing the Cubs didn’t trade for Cole Hamels because BigDaddyCool would be eternally confused, especially if they could also add Kyle Kendrick from the Rockies.
Hey guys, that post wasn’t me it was an imposter. Sorry I made a couple mistakes in my original post. I was writing that one on the phone at the bar while drinking an extra dirty martini. My point still stands though – Cubs have an easy path to the NLCS. We’ll see how the teams matchup from there.
It’s a shame you get attacked for these comments, BDC. You’re just being a real and utterly typical Cubs fan, saying what we all know & believe & tell each other all the time, but ppl here pretend you’re some sort of outlier fanboy. Your critics have to act like your comments are unseemly bc at Fangraphs there’s a pretense of disinterest and objectivity, but deep down they agree. The fact is that we will probably only lose one game in the playoffs, to the Dodgers on one of Greinke’s or Kershaw’s starts, then we’ll sweep the Blue Jays in the WS and then our at least decade-long dynasty will have begun.
Can you please explain how Lester is on another level?
The Cubs “should” sweep? The odds of them sweeping is probably close to 5-8%, but they “should sweep.”
Honestly never met a more arrogant fanbase that has never won anything. It’s remarkable, but the good thing is it all ends once Cole shuts out that porous offense.
BDC is a special dude; he really needs to just go away. Cubs fans have just as many idiots as every other fan base. BDC just somehow made it to this site and has stuck around despite nobody ever appreciating anything he says, as it is always garbage.
I agree that the Cubs have just as many idiot fans as every other team, but he’s about 50 idiot fans wrapped up into 1.
It’s a shame, too, because the Cubs are a hell of a story but he makes me want to see the Cubs eliminated early just to spite him.
I said the following in April:
“Kris Bryant is a TWO TOOL player at best. How is his floor higher? My goodness. The hype surrounding this guy is atrocious.
A 5 tool player – Buxton – is going to have a higher floor than a 2 tool player. If Bryant doesn’t make contact enough to hit for average (which is entirely possible) then his only + tool will be power. He’s slow footed, with a poor glove and an average arm. I’ve never seen a two tool player have so many people drooling over him.”
I’m clearly very sour about the fact that I was horribly wrong.
See: http://www.fangraphs.com/blogs/kris-bryant-and-floor-versus-ceiling/
Don’t forget this:
“At best he’s a 45 with his legs, and that’s right now! When he’s young.
He’ll be a clunker and base clogger in no time.
As for his arm, it’s far from above average at his position. Whether he’s in right field or at 3rd base, his arm grades out towards the middle of the pack if not in the bottom third.”
One of these days, you’ll figure out that others post under this name.
For the record, Kris Bryant’s k-rate is still north of 30%. He will not have extended success as a “top level player” in this game if he doesn’t bring that down.
Excellent Straw Man you’ve built there.
This isn’t about Bryant specifically, but what’s wrong with getting excited over a two-tool player, if those two tools happen to be really, really good? Frank Thomas (to name one) was the quintessential two-tool player, and he went into the Hall of Fame this summer.
Honestly, I just think Cub fans are excited … and for once in recent years the Cardinals seem vulnerable.
The confidence in a team that not only hasn’t won anything major in the last 107 years and/or when it has had its chances loses in the most dramatic, self-destructive, and soul-crushing way is just not understandable.
If we know the Cubs like we do, it won;t be so much Cole blanking the offense … but someone making a mistake that leads to defeat and adding to the laughable history of the franchise. I don;t know whether it will be an opponent’s light-hitting pinch hitter going deep, a cubs pitcher allowing a multi-run homer at the worst time, a defender letting a ball go between their legs, or playing a double play ball off the heel of their glove or chest, or whatever … but it’s usually something … and it’s usually very entertaining.
But, being a fan of a team in the NLC, this playoffs has the intensity of the 04 and 05 series against the Astros.
I think with the Cubs winning so many extra inning games this year, something like 13-5, combined with the Cardinals being injured more than most, there’s that familiar feeling of “team of destiny” with the Cubbie fans … and that has almost always ended in a way that the non-Cub fans tend to enjoy and bring up in every conversation. ; )
As a Cardinal fan (living in Northern, IL), I’ve been hearing a lot of sentiments like this … along with the idea that the Cubs should dominate the division for the next 5 years (at least).
I admit that the injuries experienced by the Cardinals, especially Molina and Martinez, are great concerns. The Cardinals are “famous” for having the worst playoff team ever in 2006. The season win total was lower due to injuries to key players for extended periods of time, but that team got healthy for the post-season (and was the same talented team that had the most wins in 04 and went to the NLCS in 05).
2015 is different because as everyone has pointed out, the Cardinals are not healthy going into the playoffs. However, the team is likely to win 100 games in a year when the injuries have been piling up and on a regular basis all year. It’s either a miracle of luck or an expression of the Cardinals depth and ability to replace positions.
On the other hand, it’s the playoffs. There will be some non-key player that is going to over-produce in a small sample just like there always is whether it’s Tito Landrum, Kurt Bevaqua, Buddy Biancalana, David Eckstein, etc. That crap happens.
The Cardinals really could have used a couple of recent wins by Chicago over the Pirates, and didn’t get it. So, given that the Pirates are better than the Cubs (standings) and it will be their 3rd consecutive year of playoffs, I’m not sure I would take the Arietta vs. Pirates game as a Cubs win. That seems to be a trap Cub fans continually fall into. While it’s entertaining for fans of other teams, it really should serve as reminders to stay conservative in your expectations.
A team that K’s a lot at the plate (Cubs) is going to face Cole who strikes out enough batters and whose season-long performance isn;t boosted by otherworldly LOB%, BABIP, etc. If I were a Cubs fan, based on team history, I’d be worried that the good luck on LOB%, BABIP, etc for Arietta isn’t going to continue. Anyway, the Pirates are good. StL, even with their current injuries is good. I don;t feel there are any “easy” or “dominating” series to be had with any 2 of the 3 NLC teams. StL, even with injuries, just beat the Pirates and we’ll see how it goes today.
There should be cautious optimism from all playoff teams. With the injuries sustained by the Cardinals, the 5 teams left are very close in terms of talent. Any series could go either way, and not just due to randomness or small sample baseball, but because the teams are close in talent and ability.
But, to assume an easy series would be akin to Cardinals fans pointing out that the cardinals won 100 games without Wainwright, so next year we should expect 105 wins with him.
My bet would be that the Cubs easily lose the play in game
no one cares
yet here you are, commenting on his comment.
Thank you for making this piece. I was one of the fans who were really rooting for the Cubs to give Hendricks a serious chance in the Majors last year, and it’s been great to see the results he’s put up while getting a steady diet of innings; even if he, and his velocity, flies under the radar.
Really? That’s how you’re going to phrase it? By abusing percentages and rounding?
Last night, Hendricks faced 22 batters; thus, each batter represented ~4.5% of all the batters that stepped in against him. So there’s no effective difference between 37% and 45%. You already rounded to 41% and that last digit is barely significant anyway, so you could just round to 40% — but when the total is so small why use percentages at all? He struck out 9 of the 22 batters he faced. That’s both exactly accurate and more informative.
Wait, you’re upset that he rounded to the correct number ? You would have preferred that he rounded to something that is clearly less correct? Welp.
I’m complaining that he’s using “41%” when that last digit implies greater precision than is actually warranted (or available). When I saw that percentage, my first thought was “Wait, he faced over a hundred batters? That can’t be right”
So…you’re worried that he said 41% (correct) and not 40% (less correct)? I’m still struggling to understand. He could have just said “roughly half” and been even less precise, if that would help?
I’m embarrassed, enough to tell you, that i thought the entire article was about Kyle Kendrick. And i couldn’t believe how far out of touch i had fallen… kyle kendrick plays for the Cubs? And is good enough to start a playoff game?!!
Oy vey. Baseball is slipping through my old hands (shakes fists)
Yugoslavia doesn’t exist anymore
Sadly, you’re not the only one here who made that same mistake. I’ll bet that you’re not a Cubs fan, though, so at least there are others here more out of touch than you.
http://www.fangraphs.com/statsd.aspx?playerid=12049&position=P&type=2&gds=2015-08-01&gde=2015-09-28&season=
This is the game log since Aug 1. Should be pretty clear that the increased K’s is just the last 3 appearances; one of which was 3 IP, but since the author is cleverly using %’s all semblance of sample size is getting lost. The first 8 games in that sample are ~21% K rate which is right in line with his career #’s.
Not sure if it was intentional or accidental, but this claim of increased K’s since Aug 1 is ridiculously misleading. There was basically no change until his last 3 games. But by aggregating that 15IP (yes FIFTEEN innings) over a 2 month sample, it sure does look like a much bigger sample size to the reader and the strikeout data can be believed.
Graphing things or using confidence levels would avoid this mess. Unfortunately, if you throw stuff in a table you can start getting some misleading aggregates.