Seattle Sports Insider

2010 Staff ERA - Park and Defense

Posted by

10/23/09

FLIP: In this post, Cool Papa follows Matty by noting that a big part of the M's #1 ERA was actually run prevention...

ERA is heavily affected by ballpark and defense. The M's play in a pitchers park and have a great defense, both of which should return (the loss of Beltre will be offset by having a competent fielder man shortstop the entire year). The actual performances of the pitchers weren't that great.

Matt, using his home-brewed metrics, pegs the 2009 staff as "the league's fifth best" and Cool Papa actually assesses it as "below average."

............

CHOP: Objectively speaking, the M's team pitching stats were indeed middle-of-the-pack.

They were 6th in fewest walks, 11th in most strikeouts, midrange on K/BB. The M's were 5th in fewest HR's, which is of course affected by Safeco. So, easy enough, right? Average pitching, great context. Next subject.

............

Well, let's look a little deeper, though. In this specific case, I would not brew too much soup off of this component-ERA oyster for these reasons:

....

1. The placings in the league are tightly packed. For example, the M's K/BB is 2.0, but the #3 team in the league (the White Sox) is at 2.2 and the 3rd-worst (Detroit) is at 1.9.

When the pack is as tight as that, you don't want to conclude too much. It's like saying Curtis Granderson was in the top 12-15 home run hitters in the league... fine, but there were another 20 hitters within about five of him, including Jose Lopez.

....

2. The M's K/BB numbers took a late tumble after Ian Snell and Brandon Morrow (along with French) spray-painted graffiti all over the team pitching stats.

Snell had what, a 35/30 control ratio with the M's, more BB's than K's, and in the tight pack that is the AL, he and Morrow were sufficient to drop them several places. But, it's not like Snell and Morrow were meatball pitchers. They were hydrochloric-acid spray cans that torched batters, fielders, stats, fans, and everything else around them.

....

3. You watched this team perform as well as I did. You know and I know that Felix, Bedard, Washburn, RRS and Co. did not comprise an "average" pitching staff.

Yes, the K, BB, HR, etc stats at the end of the year have the M's back in a tight pack coming around the corner. But we also know there was a long summer of outstanding pitching we saw out there.

=== Why Can't This Be Glove Dept. ===

It is true that the M's ERA was far lower than their FIP, and by a laughable margin:

ERA - FIP, AL

-0.52 - SEA

-0.18 - DET

-0.10 - TEX

-0.05 - CHI

But if you're going to take FIP as the gospel, why isn't UZR/150 consistent with FIP by this metric?

UZR/150, AL

8.9 - SEA

8.0 - TAM (-0.01 ERA-FIP)

6.0 - DET

In other words, FIP tells us that 0.5 runs' worth of the M's ERA was due to the park and defense, and that no other team's context was within 20% of that impact, except Detroit.

But the UZR/150 margin doesn't back that up, at least not as well as we'd like

What's that you say? The park saved the pitchers? But then why didn't it do that in 2008, and in 2007 -- and why does Baseball-Reference give Safeco park effects as 97 both for 2009 and for its life?

..........

As is often the case, SSI claims that the data is too inconsistent and contradictory to be overconfident about whether the gloves, park, or arms were responsible for the #1 ERA, and in what proportions.

In this specific case, human judgment should be brought to bear. You and I watched this ballclub. What did you see? What did you think of the pitching? ...

PART II

Comments

Minus Snell, Morrow, French

Umm... Snell, Morrow and French were part of the M's pitching staff. It's not really fair to take out the guys that don't push the data forward. It's like saying "The M's team defense would have been better without Yuniesky Betancourt." It's obviously true, but he was indeed the SS for a big chunk of time this year so there is no removing him.
If you're talking about next season, then that's fine. Snell and Morrow are expected to improve, and French will hopefully be no where near the big league rotation. But when talking about the team's ranking in 2009, I don't think it's fair to pick and choose.

CF and great E-F

In the past decade, great E-F shows up on average on only one team per year or so in MLB. What is common among those, as we have discussed, are great M's CFs Cammy and Gut. The rest of the list is similar. Cammy made his third appearance on the list with the Brewers in 2008, Edmonds for the Cards in 2001/2004/2005. Other single appearances were Upton, Granderson, Rowand, AJones, and Erstad. It seems having a great CF is no guarantee of a great E-F, but you probably won't make it without one.
So how does it look for the M's chances for great E-F in 2010? If history is a lesson, the years we had a hot CF (and Ichiro in RF), we made the list three of four years. Without a great CF, well, not ever. I'll take my chances for success with our OF defense, but I'll see what Jack does with the rotation between now and spring training.

You missed the point Jon

n/m
And just out of curiosity, are you only going to post when you disagree with something? :- ) Is that what I should do on your blog?

Agree with that Watcher

The thing that has me scratching my head, is that I haven't seen it repeat the next year.
......
Without even looking, I'll bet dollars to donuts that the 2009 Brewers didn't have a great E-F.
And as you note, the Upton-Granderson-etc appearances were single appearances.
......
Which raises questions about whether the CF's are *causing* the E-F's, because if they do, why don't they cause it the following year?
But like you say, we'll take our chances, because given Zduriencik's apparent bias towards defense, the LF will probably be a CF too, and if there's any way to help pitching, it's 3 CF's in Safeco.
Problem is, there might not be any way to reliably help pitching, to the degree ours was in 2009...

not all CFs

Having 3 CFs may increase your chances for great E-F, but some of those teams had more like DH types at the corners. None had it over ichiro and Gut. For his bat, I'd still take Dunn in LF, if we couldn't get him off the field at DH or at 1B, and take the chance that Gut could cover for him.

Doc

I didn't say that there weren't a few pitchers who performed well for stretches of time. Of course there were. What I pointed out was that those guys didn't pitch MUCH. Yeah, Bedard was great, for 83 innings. RRS looked good, but only threw 96 frames. Washburn is a big regression candidate, but he was ditched with a full third of the season left so he contributed only 133 innings. As I said in my original comment (in the portion you left out) this team had to turn to a whole bunch of junk in order to get through games. That means bringing in even decent pitchers can make a big improvement.
And as Matt made clear, pointing out that Morrow and Snell were terrible isn't a counter to my arguement that the staff as a whole didn't pitch well. That's exactly the sort of thing that I had in mind. Those guys have a great chance to do better. So I don't see how my arguement is refuted. This team is still perfectly poised to get better performances from it's pitchers which would offset any decline in defense or synergy or whatever. It is perfectly reasonable to expect another league leading ERA.

Great E-F not repeating year to year

The reason that great E-F doesn't repeat year after year may be that a team's pitching may change more than its defense. For an individual pitcher, what gives the best E-F? It's a K rate of ~4-6/9, walk rate of ~2-3/9, ~1 HR/9, and an ERA

Wrong conclusion to draw, I think.

You forgot to check STRAND RATE, Doc. E-F is not JUST park and defense...it's also sometimes strand rate luck and bullpen quality. Tampa Bay (0.01 E-F, pretty shoddy bullpen, below average success stranding runners, well above average defense) demostrates my point here.
So does Seattle (which likely pitched into some good luck on bullpen strand rates this year and that may revert some unless we add some legit relief weapons).
I don't think Seattle's park and defense deducted half a run per game. I do think Seattle's park deducted about 0.2 runs to 81 of their games and their defense deducted another 0.25 runs per game (all 162 of them) and they had some good fortune in their reliever appearances helping them out too.

22 more RA on the road

Matt,
You seem to be in the ballpark, with The Safe saving ~0.2 runs per 81 home dates, puns intended. We gave up 22 more runs on the road than at home. Did we play worse defense on the road?

Not really, no.

And my -0.2 R/G estimate comes from the Fiato/Souders matrix for seasons through 2007. I co-created a system that tries to more accurately identify what parts of run scoring were caused by the abilities of each team's defense, starting pitchers and offense, the parks, the home plate umpires and the leagues all at the same time (because doing park factors or league factors or umpire factors separately and adjusting statistical analyses one factor at a time is mathematically bad (because you're using data tainted by context to try to figure out the contextual adjustments at all times). The F/S Matrix is much more stable for park factor calculation. The Safe has been preventing about -0.2 R/G/Side since 2003 when the black hole batter's eye was installed, thus moving it from a deadly pitcher's haven (due to glare) to merely a fair pitcher's park.