The problem with Javier Vazquez: Bunching hits and walks?

Let’s start with some quotes…

Part of Vazquez’s inability to win at home so far can be attributed to hard luck, but part of it can also be explained by his tendency to be victimized by one bad inning.
MLB.com

Vazquez’s problem has been one bad inning, usually the fifth or sixth.
Chicago Tribune

He holds the patent on the Really Bad Pitch and is currently litigating for trademark rights to the term “One Bad Inning,”
ESPN (Keith Law)

This time, Javier Vazquez didn’t have reason to be frustrated about that one bad inning that doomed him courtesy of a number of soft singles.
MLB.com

Javy has three quality pitches, but the one thing that has got him into trouble this year is just one bad inning, a hiccup. … When he avoids that, he has been dominant.
Mark Grace

Javier Vazquez had that one bad inning syndrome thing we had heard so much about when he came over here, though, like Frenchy said, it wasn’t like they hit him hard or anything.
Talking Chop

Vazquez has a reputation as a “1 bad inning” guy. Now, I have yet to find anyone who has actually studied his game lines to see if he’s prone to clumping his hits and walks together (producing more runs than expected for that numbers of hits/walks), but it’s at least logically possible, and given his reputation, it’s worth investigating.
Commenter “Mark” on my Vazquez article from Monday

On Monday, I talked about why I believe Javier Vazquez will be one of the top pitchers in baseball this year. A couple commenters were less than convinced, saying that even with the improved peripherals predicted by CAPS (and which he is currently displaying), he still may not get to that elite level. I pasted an excerpt from commenter Mark above, essentially summarizing what so many sportswriters have been saying for years. Today, I’d like to examine whether or not this is actually true of Vazquez or if it is simply incorrect conventional wisdom that has developed into a sort of conformation bias each time it happens.

Methods

To test the validity of the claim, I used the ever-useful Retrosheet to examine Vazquez dating back to 2004. There were a couple different ways to tackle the problem, but I went with what Mark suggested—how often Vazquez bunches hits and walks (and HBP) together, “producing more runs than expected for that numbers of hits/walks.”

A Hardball Times Update
Goodbye for now.

To define “bunching,” I’ll say that it is any inning in which Vazquez allows more hits and walks than his WHIP would indicate. As almost every pitcher posts a WHIP between 1.00 and 2.00, every inning in which he allows two or more runners will be examined. In my calculations, I broke things down by the percentage of time Vazquez allowed at least two, three, four, five, six, and seven hits and walks in an inning.

[For those really interested, I made sure to use the number of instances in which a pitcher started an inning, not his total combined innings for the year (i.e. when a pitcher is taken out after recording just one out, this counts as a full inning for our purposes).]

In addition to testing Vazquez’s numbers, I also ran the numbers for league average. I wanted to test a group of pitchers with similar peripherals to Vazquez as well, but I couldn’t quite get it done in time. I may post those results in the future, though it’s entirely possible they don’t differ too terribly much from league average.

Results

image

If you’d like to see the results for each year individually, click here.

Overall, the results don’t lend too much weight to the arguments that Vazquez is prone to bunching his hits and walks together. He has been better than average in allowing two, five, and six H/BB innings and below average at three, four, and seven H/BB innings, but not by a whole lot (he also never allowed more than seven, while some pitchers allowed as many as 11). In addition, there doesn’t appear to be any recognizable year-to-year trend. He was almost exactly league average in 2004 and 2005, terrific in 2007, and poor in 2006 and 2008.

The fact that he is below average in the three and four H/BB innings might lead us to believe that this is what sportswriters are seeing, but what we’re really looking at is just 0.7% more three- and four-runner innings than league average. That comes out to 1.5 innings per season (assuming 210 innings pitched). Plus, in the really damaging five- and six-runner innings, he’s a bit better than league average.

The net result of his 2004-2008 work is actually the bunching of 1.6 fewer hits and walks than league average per 216 inning appearances (his average number pitched since 2004). If you want to exclude the innings with two hits and walks (which are much less likely to end in runs scoring), he would still only be bunching 5.4 hits and walks more than league average. Exclude the three H/BB innings? Drops to 5.1. Hardly seems condemning, and although it would be useful to see what similarly good pitchers are doing, I think it’s relatively safe to say that Vazquez isn’t some super-magnet for quick, sudden blow-ups.

Conclusions

More likely, I’d wager we’re seeing at least some degree of confirmation bias. After all, a full 16 percent of Vazquez’s innings have resulted in three or more hits and walks. That raw percentage is pretty high. While this comes with the territory for all pitchers, because Vazquez has such a reputation for it, it gets noticed and pointed out much more often when it happens to him.

I may delve a little deeper in the future, but for now, I think this should definitely give us something to think about. At the very least, it means we shouldn’t rule out the possibility that he’ll pitch like an ace for the remainder of 2009. In fact, I think it makes it a little more likely.


18 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Bill
14 years ago

I think you’re making this much more complicated than it really is.  Javier Vazquez simply isn’t nearly as good of a pitcher when he has to pitch out of the stretch.  While his HR/9 stays steady, his strikeout rate plummets (8.8/9 to 7/9).  Coupled with an increased walk rate (2.1/9 to 2.8/9), there really shouldn’t be much confusion why Javier Vazquez underperforms his overall peripherals every year – when he is in a sticky situation he is a substantially worse pitcher than with no one on base.

Phil
14 years ago

One problem with the above analysis is that each hit and walk are weighted equally.  Maybe Javy isn’t prone to allowing more baserunners in bunches than league average, but maybe the types of baserunners he allows do more damage.  As an example, a pitcher like Daniel Cabrera might have a 4+ H/BB for an inning, but all baserunners could reach by walks.  This analysis treats that outcome the same as where another pitcher gives up a 2B, a walk, a HR, and another HR.  The first pitcher has only given up one run, while the second has yielded four.  Maybe a followup analysis could look at ISO as well?

Another problem is that here you are comparing Javy to the league average pitcher, but your previous article argued that Javy would be an elite pitcher this season.  Shouldn’t we expect him to beat the league average if he is an elite pitcher?  How do Javy’s Hits+Walks Bunching numbers compare to other elite pitchers?

David Gassko
14 years ago

Derek,

It looks to me like you’re proving the point that Vazquez suffers with runners on-base. With no one on, he’s one of the best pitchers in baseball, but with runners on-base, he’s merely average! No wonder his peripherals make him look better than he is.

With that said, Vazquez is still a very good pitcher.

GWR
14 years ago

I just can’t get past the fact it was suggested you should trade Roy Halladay to get this guy.

bpasinko
14 years ago

I took the trade for Vazquez stuff as you could swap a Halladay for him and not get hurt, but you likely can get something in addition to that, really making it worth it.

Vazquez is going to have 1 bad inning his next start and people are gonna say see I told you so!  And then not think anything when he pulls off 3 great starts in a row after.

This isn’t related but what would CAPS say about Porcello?

Barry
14 years ago

I agree with Phil’s suggest of comparing Javy to other elite pitchers…

Also, looking at your data another way, even though Javy gives up fewer 2+ H/BB innings than league average, he is more likely to turn those 2+ H/BB innings into 3+ or 4+ H/BB innings (i.e., once the fire gets started, he adds some gasoline… OK, maybe paper)… Although it’s presumably a small sample size, 80% of the innings that reach 6+ H/BB become 7+ H/BB, while the league average pitcher only “loses” about 30% of those 6+ H/BB innings…

Jim
14 years ago

If this was taken a step further to see how many runs were give up in each of these types of innings I wonder if that would show something. A comparison to of this to League average SP would be interesting as we may see he is similar to or worse than them at allowing runners to score when they reach base in ‘bunches’.

Dave
14 years ago

Excellent, thought-provoking article.  As a Yankee fan, I’m sure Javy gives up big innings.  He will look untouchable and then BOOM – a crooked number.  Perhaps you want look at correlation between HR’s or extra base hits with the walks and hits total for the inning.  Also, % of HR’s and XBH’s with runners on.  Both of these compared with league averages might show something.

Great work.

Andrew
14 years ago

I don’t mean to nag, but I’d still really appreciate a list of the LIPS ERA leaders, Derek. It doesn’t even have to be a full post. I just think it would be a great way to spot some Buy Low guys.

Dave
14 years ago

I think Bill makes an interesting point by comparing K/9, BB/9, and HR/9 with with and w/o baserunners, but those rate changes are likely to reflect situational pitching in addition to characteristics of the pitcher.  We need a control group to really determine if Vazquez is truly a worse pitcher out of the stretch.

Here’s a quick-and-dirty comparison of the career rates of Vazquez to Roy Halladay, Bases Empty/Runners On/RISP (+/-% compared to all situations)

Halladay:
K/9:  6.81 (+5%)  / 6.06 (-7%)  / 6.89 (+6%)
BB/9:  1.83 (-11%) / 2.33 (+14%) / 2.79 (+36%)
HR/9:  0.80 (+8%)  / 0.66 (-11%) / 0.55 (-26%)

Vazquez:
K/9:  8.80 (+9%)  / 7.05 (-13%) / 7.48 (-7%)
BB/9:  2.10 (-12%) / 2.78 (+17%) / 3.58 (+50%)
HR/9:  1.19 (+1%)  / 1.16 (-2%)  / 0.99 (-16%)

One man does not a control group make, but this seems to support the hypothesis that Vazquez is a worse pitcher across the board when pitching with runners on (out of the stretch) rather than being merely hit-unlucky.  I won’t mine too deeply here, but the fact that HR-stingy Halladay was able to depress his HR rate further with ROB while Vazquez wasn’t is pretty interesting.

Mark
14 years ago

Interesting article and discussion…regarding the Halladay comp, Dave, did you notice the spike in BB/9 for both pitchers with RISP?

I wonder if intentional walks (and intentional/unintentional walks) contribute much to this figure, given that we might expect some of those circumstances to be with 1st base open and 0 or 1 out.

Derek Carty
14 years ago

Thanks for all the comments, guys.  Before I address some of them, let me say that I would like to do a follow-up and look at some more things.  This was done as more of a preliminary thing to look at a specific issue that one commenter brought up.  There is certainly more to look at, as I noted a few times in the article.  Maybe I’ll try for Friday.

Bill: You’re right in that he seems to pitch a little worse with runners on base, but a career 70% LOB% isn’t awful by any means.  A pitcher can still be very good with this, although I think a deeper look like Dave begins might be insightful.

Phil: This is one way in which I was thinking about following up.  I’ll likely look at the quality of hits in these situations and see if there’s anything we’re currently missing.

David Gassko: Maybe.  This was actually one of my concerns when presenting this.  If every pitcher averages between 1 and 2 hits+walks per inning, though, I’m not sure we can assume that the distribution of, say, 4+ H/BB innings is going to drastically differ among good and average pitchers.  It very well could be, but if the averages are roughly the same, I don’t think it’s preposterous that the distribution will be roughly the same.  We’d need to prove that actually *avoiding* bunching is a skill that a pitcher can possess, which to my knowledge hasn’t been done yet.  It would be very interesting to look at, though.

bpasinko: My point exactly in regard to Vazquez/Halladay.

Unfortunately, I don’t have CAPS for 2009 yet.  Because so much goes into it, and because some of the data sources I use aren’t available mid-season (i.e. Retrosheet), I don’t yet have in-season CAPS available.  This is definitely something I’d like to hook up at some point in the future, though.

Barry: Yeah, comparing to elite pitchers is something I wanted to do but ran out of time (I think I mentioned this in the article a couple times).  That would be something to look at in a follow-up.  Also, I hear what you’re saying about things spinning out of control, although the actual totals produced here are pretty small, as noted/alluded to in the article.

Jim, Dave: Yup, looking at stuff like this would be nice.  I’ll have to figure out exactly how I am going to run the follow-up.  I said initially that there were a lot of different directions I could go, so I’ll have to think about things.  Thanks!

Andrew: Check the Fantasy home page for it smile

Ameer
14 years ago

Eric Seidman did a Vazquez analysis on BP for anyone interested.  It boiled down to Javy pitching significantly worse out of the stretch.  I won’t post any numbers here since it was a subscription article, but it’s worth checking out.

bpasinko
14 years ago

Thanks Derek.  Here’s to wishing you can get midseason CAPS in the future.

Bill
14 years ago

Ah, thanks Ameer – I knew someone must have written an article about him, he always has been a perplexing and frustrating pitcher.  For those with a BP subscription, here is a link: http://www.baseballprospectus.com/article.php?articleid=8872

For those without, the general upshot is that in general, pitchers don’t strike out as many guys from the stretch, but Vazquez is worse than league average and also allows more walks.

David Gassko
14 years ago

Derek,

See the Sean Smith article I referenced earlier in a comment on your previous article.

Dave
14 years ago

A “one bad inning” guy should have a higher variance for runs/inning.  So just find the variance for the data set of number of runs he’s allowed every inning he’s pitched and compare to some other pitchers with the same ERA.

Mike
14 years ago

Very cool – this shouldn’t be just confined to the fantasy blog though, it’s an interesting question about baseball perceptions in general.