Making Sense of Brandon Morrow’s ERA
Brandon Morrow is set to sign a three-year, $20 million contract extension with the Toronto Blue Jays. The deal reportedly includes a $10 million option or a $1 million buyout in 2015. While we don’t have the particulars of the deal at this point — including how much Morrow could make in each of his three seasons — it’s sure to create some arguments among fans.
In Morrow’s two seasons with the Blue Jays, his 4.62 ERA has been below-average among American League starters. Yet Morrow’s FIP and xFIP during the same time indicate that he should be one of the league’s better ones. As a result, many stat analysts have predicted a major improvement from Morrow. But until his ERA matches up with his advanced stats, Morrow is going to remain a frustrating player.
During the past two seasons, the difference between Morrow’s ERA and FIP (E-F) has been extreme. In 2010, among pitchers with at least 140 innings, Morrow 1.33 difference was the highest in baseball. While the large difference meant Morrow was likely to improve, he didn’t improve by much. Morrow’s 1.08 E-F in 2011 was the fourth highest on the leader board among qualified pitchers. Though his peripherals indicate Morrow pitched far better than his ERA, things have yet to even themselves out.
A look at Morrow’s performance with men on base reveals some interesting findings:
| Brandon Morrow 2010 | IP | K% | BB% | FB% | BABIP |
|---|---|---|---|---|---|
| Bases Empty | 77.2 | 27.6 | 11.5 | 40.2 | .312 |
| Men on Base | 68.2 | 29.2 | 9.3 | 43.7 | .380 |
| Men in Scoring | 35.1 | 29.8 | 10.1 | 43.8 | .411 |
| Brandon Morrow 2011 | IP | K% | BB% | FB% | BABIP |
| Bases Empty | 104 | 25.8 | 8.7 | 37.6 | .273 |
| Men on Base | 75.1 | 26.5 | 9.2 | 47.4 | .335 |
| Men in Scoring | 40 | 26.3 | 10.2 | 49.5 | .355 |
Morrow’s peripherals — particularly his K and BB percentage — don’t change all that drastically when he has runners on base. In 2010, Morrow’s walk rate was actually at its lowest with men on base. While his walk rate increased with men on base in 2011, his strikeout rate jumped too.
Morrow tends to give up more fly balls with men on base, as well — perhaps as a result of pitching up in the zone to get more strikeouts. While we typically expect fly balls to lead to a lower BABIP, Morrow’s BABIP actually increased in these situations.
Morrow’s poor BABIP with men on base could be one explanation for his difference between ERA and FIP. Either that, or Morrow is terrible out of the stretch. That argument doesn’t seem likely since most of his peripherals stay fairly consistent with men on base.
For Morrow’s ERA to match his FIP, his BABIP in crucial situations needs to decrease. Thankfully for Morrow, there’s a lot of year-to-year variance in league leaders when it comes to BABIP and men on base. Over the past five seasons, only five pitchers have finished in the top 10 in BABIP with men on base more than once. Those five are Morrow, Derek Lowe, Livan Hernandez, Justin Verlander and Kevin Millwood. The fact that Lowe, Hernandez and Millwood made the list multiple times shouldn’t be as surprising, since they’ve routinely given up lots of hits each season.
On this list, perhaps Verlander is most comparable to Morrow. It’s not a perfect comparison — Morrow strikes out and walks more hitters than Verlander — but the other guys on the list are generally low-strikeout guys who give up a lot of hits. Like Morrow, Verlander’s BABIP with men on base ranked in the top 10 for two consecutive seasons. Following those two years, though, his BABIP with men on base dropped to a more reasonable level. In 2011, Verlander’s .249 BABIP with men on base was one among the league’s lowest.
So what’s next for Morrow? It seems that luck —actually, the lack of it — has played a big role in Morrow’s struggles. Even if Morrow is less effective pitching from the stretch, it’s incredibly unlikely he’ll end up with another top 10 BABIP with men on base for yet another season. This year, his ERA might finally match his FIP. If that happens, Morrow would go from one of the most frustrating starters in baseball to one of the best.
Chris is a blogger for CBSSports.com. He has also contributed to Sports on Earth, the 2013 Hard Ball Times Baseball Annual, ESPN, FanGraphs and RotoGraphs. He tries to be funny on twitter @Chris_Cwik.
I’d be curious to know how this compares with results of other Jays players.
I imagine you’re interested in examining how much the Jays inferior defense plays into this ERA/FIP discrepancy. While their defense was below average at -12 runs the Toronto staff as a whole had a BAbip of .293.
So while the defense may certainly be a contributor, it’s not the only problem. It’s a good bet that Morrow has other issues, i.e. pitching form the stretch.
I don’t think you can ever peg another pitcher as a ‘Ricky Nolasco type’ because he is so uniquely anti-FIP, but Morrow is approaching that territory.
For what it’s worth the Jays had one of the worst OF defences in the game. The infield defence was a bit better. Which would hurt Morrow more than the other Jays starters given his extreme flyball ways.
But I agree, I think Morrow’s more likely a Nolasco case than a victim of luck.
Good point.
Although that stretch of 173 IP without inducing a GIDP was certainly not an example of his Infield helping him any.
As a Jays fan, just from watching him in pretty much every start, he tends to give up alot more hard contact with men on base. The ground balls are sharp, the flyballs are deep to the wall and everything results in hard contact. He also falls behind alot in the count with runners on (he’ll miss outside by wide margins, heat maps should show this) and he’ll follow up those outside pitches with ones right down the middle because of his being down 3-1 and such in the count. His stuff is just so good that sometimes he gets away with pitching in the heart of the plate resulting in K’s.
I agree. It seemed like every time a runner would get on base, he would get behind, and try to blow a FB past the hitter, who would respond by crushing it to the gap. He may have had some bad luck in regards to BABIP last year, but he did not look good at all with runners on. Until this improves, I think his ERA will consistently be at least half a run higher than FIP.
The ALs Ricky Nolasco?
Kinda sorta….
If you look here, http://www.fangraphs.com/statsplits.aspx?playerid=3830&position=P&season=2011 , you will notice that Nolasco actually does pitch far worse with men on base and in scoring position unlike Morrow. There is reasonable evidence at this point that Ricky just does not pitch well from the stretch and his consistently low LOB% bears that out. Furthermore his K rates are rapidly declining over two seasons again unlike Morrow.
Sometimes it’s helpful to, you know, actually watch a game or two. If that was the case you would know that the Blue Jays OF defense sucked last year. Patterson in CF, Thames in LF and even Joey Bats in RF are below average to God awful defenders.
Stats are helpful most of the time but occasionally it’s a good idea to think outside them for the actual reasons why they don’t match up.
Word.
There is that too, I mean Thames Bautista Davis all in the OF at once was really disgusting, but the fact is he gives up so much hard contact when he was pitching from the stretch. I wish people would record MPH off the bad vs. pitchers per situation
who are you talking to, the author? I don’t recall him making claims to have solved the mystery. Rather, he’s doing a bit of research and offering possibilities. Can’t you contribute to the conversation without sounding like a douchebag? Furthermore, whether or not the OF defense was bad, I guarantee you can’t name an instance last season when the outfield defense directly caused Morrow’s ERA to rise. So why the need for “Sometimes it’s helpful to, you know, actually watch a game or two.”?
Parkes runs a fairly well-received Jays blog, but one of the criticisms levelled at him in the comments sections is occasionally over-banging the drum as to the usefulness of stats. My money’s on this comment not actually being him. Which is not at all to disagree with what you’re saying, Steve.
Well received? Parkes is an idiot and a lousy baseball writer.
So I understand the OF defense plays worse with men on base than with noone on?
Interesting theory…
Why not just ask if the Jays hate Morrow… or do all Jays pitchers exhibit similar splits?
Swing and a miss. Occasionally it’s idea to think about what you just wrote to see if it made any sense or not. Outfield defense would not particularly explain drastic RISP splits like these for one pitcher in an entire Jays staff.
Pending further evidence, this looks like a case of “splits happen.”
From the stretch, his K% and BB% are about the same as from his full windup. but as the author notes, the FB% looks consistently higher. Wonder if heat maps confirm he does pitch higher in the zone from the stretch. If so, could just be an approach thing–either from what the author writes (trying to get Ks, although K rate isnt higher) or as Steve says, he gets behind in the count.
Chris, does data support either of these explanations?
I was hoping we could get thru one article discussing BABIP without mentioning the word ‘luck’. Alas.
Next time, I’ll use the term “good fortune.”
And then I will use the term “ferflingenflangen’
“Random chance” works well too, especially since it’s neutral (you can have bad fortune too, I guess, but it sounds weird)
I’d suggest fortuna next time. If Morrow was an astute reader of Machiavelli he’d wrestle Fortuna down and then he wouldn’t have to worry about RISP.
I’m genuinely curious to hear why you dislike having the word “luck” in an article discussing BABIP.
At $21M (20M + $1M buyout) for 3 years, he doesn’t even have to produce 2 WAR per year to be worth it. Even his FIP descended to his ERA, he would meet that goal.
This looks like an excellent deal for the Blue Jays.
Arbitration eligible players shouldn’t be evaluated using free agent $/win pricing – their prices are deflated by their inability to negotiate with anyone but one team.
True dat.
Yes, as all we know the goal of baseball is to achieve a surplus in $/WAR value, so I for one and feeling real good about the Jays’ chances of winning the World Process Series this year. What a deal, indeed.
Imagine if AA could take the Andrew Friedman Award for Excellence in Process from Andrew Friedman himself? Jays fans would party in the streets, to be sure.
Perhaps wishing for a surplus in $/WAR value AND wishing that AA could win the AFA is too much to ask, though. Getting ahead of myself!
This was supposed to be a reply to my friend Baltar above. I may have messed up the execution, but I promise you guys that my process was solid and I fully intended on clicking the proper reply button. If I had another chance at it I would get it right, so just regress the above comment if you dont mind.
AA strikes again!
Just a quick note:
“In 2010, Morrow’s walk rate was actually at its lowest when the bases were empty”. Looks like the opposite is true according to your table.
Yes, that should read “at it’s lowest with men on base.” I’ve fixed it now.
“At $21M (20M + $1M buyout) for 3 years, he doesn’t even have to produce 2 WAR per year to be worth it. Even his FIP descended to his ERA, he would meet that goal.”
If you consider the artificial salary constraints imposed by arbitration then I don’t think the bar is quite so low, but if you use fWAR at least he should comfortably exceed that. bWAR, which I believe uses runs and not FIP in it’s calculations, is another story.
Bah, humbug. I would have signed this chap three years ago for 8 yrs, 42.4mil. You can’t survive in this day and age if you try anything new, you young whippersnappers should know that by now!
I must say its possible that his stretch bb and k rates would remain the same because he is not losing stuff from the stretch. If he is simply losing command and not control it should show up, perhaps, as increased babip because guys are making much harder contact on more hittable pitches. Likewise if he has little confidence commanding his off speed stuff from the stretch it would explain the increased fb usage. This article badly needs a stretch vs windup. Heat map study. Still i admit Cain, morrow, nolasco are pretty much the most interesting thing in stats for me. Also righettis magic pixie dust.
What would you be looking for in a stretch vs windup heat map? location? to see if he’s missing corners?
I’d be interested in if he has a game theory problem. Throwing the wrong pitches in the wrong counts. I wonder if he shakes off his catchers more often than the other Blue Jays pitchers?
That or he’s tipping his pitches from the stretch.
Yeah good point on tipping from the stretch, very possible. That would also greatly affect that fb%. If the guy truly feels that from the stretch he is tipping his off speed stuff he naturally be hesitant to throw it. Need a follow up with heat maps and release points from different situations
Wouldn’t his K% w/RISP suffer a bit more if that was the issue?
Having not seen him pitch frequently, my guess would be that he has poor mechanics out of the stretch. Tipping pitches for 2 full seasons seems unlikely – that can be recognized and corrected by a good pitching coach (e.g., Don Cooper with Jose Contreras and Edwin Jackson).
True. The real question is why the uptick in fb%. Is it simply going with what you feel is best? Is ge having more trouble commanding the breaking stuff. This is a SUPER important topic we are dealing with here and this was a nice start but Anders nothing. “Its probably luck” doesn’t answer the question. We need the heat maps and release point data from windup and the stretch. And of course the pitch value fats to go along. If he cant command his off speed stuff or it flattened out from the stretch its going to definately be in that data. If after examining that we still have no answer i will be totally ok with calling it simply an abnormally long unlucky babip stretch.
But what we have here isn’t nearly enough to make that conclusion when there are reasonable and testable hypotheses available that don’t involve the highly controversial babip luck trekking.
Yes, tipping pitches would be especially surprising since not only does Morrow have a pitching coach, but his manager was a good pitching coach very recently.
Damn autocorrect. Babip luck gremlin not trekking which looks nothing like gremlin and isn’t a word as far as i know. Unless its a verb to describe the act of fantasy toke playing star trek characters. Like ” oh i cant go to that game, me and some buddies will be trekking. I’m the guy in the red shirt who gets vaporized the second we transport down to the planet.”
Correct me if I’m wrong but doesn’t FIP and xFIP value K’s very highly? If so then wouldn’t pitchers who exhibit high K/9 rates would be valued highly, even if their other skills are more marginal? I think the same phenomena is seen in other high K/9 pitchers (J. Sanchez perhaps), if this is the case then the metrics FIP and xFIP should be taken with a grain of salt on certain types of pitchers.
Again, I’m not really sure I just have this vague notion that’s been rolling around in my head while being frustrated with Morrow the last two years.
I’d like to see heat maps of pitches (windup vs. stretch) and a spray chart, broken down winup vs. stretch. If he is leaving the ball up more out of the stretch, then more balls would be hit to RF from RH-bats. Morrows FIP-ERA just happens to align a lot closer when he had Ichiro catching everything for him in Seattle. If that explains the variance, then Morrow simply needs to bring the ball down against RH bats.
Interesting. Certainly going from Seattle to torontos of defense. Is going to hurt.
A simple question, how many putouts did Ichiro make for Morrow during his 200 innings playing for Seattle? How does that compare to the # of putouts made by Jays RF-ers for Morrow during a given 200 inning stretch for the Jays? How many of the putouts were with men on base?
It’s a problem with pitching out of the stretch. Until he can fix that issue, I don’t see this trend changing very much.
Why not just stop pitching out of the stretch?
There are other ways to influence a running game…more throws, change the timing on your delivery. If the option is giving up more stolen bases versus more home runs with runners on?
I dunno, is that a stoopid idea?
Even Adam Dunn would be able to steal on a guy pitching out of the windup.
I don’t get how anything he does can make him have higher babip out of the stretch. It would have to be particular to him. Other pitchers don’t seem to be consistently different in that regards says the author’s work here.
I know some people hate it, hell so do I, but luck seems to be part
of it. There are ~ 200 MLB SPs somebody’s gotta play the role of outlier.
Ok, I’m willing to say luck plays a role. But i don’t think its ok to assume that with three other viable alternate hypotheses available. Least likely is OF defense. That obviously shouldn’t be affected by the stretch. However if he uses Hugh fb mord from the stretch we should see an uptick in fb%. Possible.
Or the tipping pitches. I agree he has a good pitching coach and manager. Shouldn’t be the case but possible. Unfortunately i don’t think there is any way for us to test that.
Finally a serious issue with command from the stretch causing him to rely on more fastballs and causing hulis breaking stuff to flatten. Pitch fx. And release point data should allow us to test this. I agree he could just be an outlier but its been a couple years. These other hypotheses are no less likely than simple luck IMO.
Higher fly-ball rate leads to higher sacrifice flies? More runs and higher BABiP or is this effect too marginal and/or already adjusted for in the FIP and xFIP calcs?
Also, Morrow does exhibit higher HR tendencies when pitching from the stretch. Doesn’t necessarily explain higher BABiP, but larger variance b/w actual ERA and advanced figures.
Don’t most pitcher exhibit higher HR% from the stretch?
Strange that he has had the wiff success when all of his pitch values are negative (below average) over the last two years (except the slider). Does he not throw the slider with RISP?
His tERA last year was 4.14, so maybe when you look at other stuff like, how hard people hit him and don’t just sweep anything not a BB, K, or HR under the rug, he’s really nothing better than averagish? Verlander, the comp here, had a tERA of 3.09.
Can someone give me a good explanation why we use FIP instead of tERA? That’d be like someone being stupid enough to use ERA instead of FIP. It’s like saying “yea I know we have a measure that factors in more elements and looks at them more objectively, but let’s use this one instead”.
explain how tERA is a better estimator of future talent than FIP/xFIP/SIERA
future ERA rather
I think it’s a better talent evaluator. I don’t know of correlation to future statistics, some guys always outperform their FIP, some don’t. I just see FIP as an extremely flawed stat. For example, what if Morrow gave up 10 doubles that, had they been in a smaller park, had the wind been blowing out, had the wind not been blowing in, had it been hit a split second sooner and 10 degrees to the left or right, would have been a home run. Well, he now has 10 doubles that aren’t counted and doesn’t have 10 home runs because of luck. We don’t count these doubles because they were “lucky”. It’s a flawed stat. Looking at tERA, which factors in GB, FB, and LD, you see that the type of contact he gives up creates a stat that’s closer to his ERA. In other words, people hit the ball off him hard (well, harder than his FIP suggest).
FIP just bugs the hell out of me. HR can be lucky. Unless FIP only counts “no doubt” home runs, which I don’t think it does (correct me if I’m wrong), and only counts BBs and Ks where each called pitch has been correctly called a ball or strike (again, if it actually does adjust for that please tell me) then FIP isn’t taking all luck out, just counting different stats and still using luck. Why not discredit anything but singles? To me, most extra base hits are balls that are hit hard, thus, luck or not as to whether a fielder could have got to them, should be counted because solid, hard hit balls are a bad thing.
We can measure speed off the bat right? How about we group grounders, liners, and flyballs and then further group those 3 into how fast the ball comes off the bat? Then evaluate a pitcher that way. A “hard hit” grounder is worth X, a “average hit” grounder is worth Y, etc etc. It just seems so ridiculous to discredit everything but BB, K, and HR. I’ve read about DIPS theory, I think it’s a good starting point, but we shouldn’t stop. In 10 years everyone is going to make fun of anyone who is still using FIP because it’s obvious flaws will be fixed by then.
Not only that, but why don’t we evaluate hitters the same way? Hard hit, as defined by some finite MPH, average hit, and soft hit, then apply those to flyballs, liners (a liner can be softly hit, like a base hit up the middle that bounces just after second), and flyballs. Also count walks, infield flies, strikeouts, and home runs. find a value for each in a specific year, create your equation and bam, you have a statistic that’s concise and as objective as I can think of. It’s also something we really could do like, in 2012.
It’s also something we really could do like, in 2012.
Ok, get to it.
Between 30 hours of work, 10 hours at the gym, and 15 hours of in class time, plus like another 5-15 of homework/reading….no. Pay me and I will; maybe I’ll apply for an internship with the Athletics, although I can’t afford an unpaid internship so maybe not.
Morrow’s delivery is significantly different pitching out of the stretch versus when the bases are empty. It makes me think that his LOB% and BABIP with runners on isn’t going to regress to the mean anytime soon
I’m not sure if anyone mentioned, but if the rate of FBs goes from 37.6% to 49.5%, maybe the opposing managers told their hitters to look FB with men on base. It’s not quite tipping pitches, but it’s close.