What's the Norm?

[Theoldmole]

i was wondering what's the standard for various pitchers is in terms of percentage of quality starts to total starts? What does one expect of an no. 1 starter, a 2, a 3, a 4, a 5?

What's the acceptable ratio for a starter in each of those slots, what would qualify the starter as very good?

[Benjamin Grimm]

http://www.sitcomsonline.com/photopost/data/1029/4328norm01.jpg

I was thinking I could churn out the 45-year numbers from the UMDB, but the problem with that is that it's often not clear who the Number 1, 2, 3, 4, and 5 pitchers are. Even this year, Glavine got the Opening Day start, but I think he'd be considered the Number 2, and Pedro would be the 1. After that, things are really fuzzy. Trachsel is 3, I guess. But what numbers do you give Soler, Hernandez, Bannister, Maine, and Gonzalez?

[Theoldmole]

I'm sorta figuring there's not much difference between what you'd expect from a 1 or 2, or from a 4 or 5. But there would likely be a diffence between the two pairings. If you're getting 3 quality starts out of 5 from a 1-2 guy, are you happy, or are you looking for a guy who'll give you 4 out 5? Is 3 of 5 OK for your numnber 3 guy?
Is 1 of 3 what you'd expect from the number 5 slot, or are there 5s who'll give you 50%?

[Benjamin Grimm]

The definition of Quality Start is 6 or more innings and 3 ER or fewer, right?

Nine innings and 4 ER isn't, even though it's a better ERA than 6 and 3.

That doesn't quite make sense, but I think that's the definition.

Anyone know otherwise?

[Elster88]

My own personal definition is:

6-6 2/3 IP, 2 or less ER
7-7 2/3 IP, 3 or less ER
8-8 2/3 IP, 3 or less ER
9 IP, 4 or less ER

[MFS62]

Yancy, 6 IP with 3 or fewer runs is what I've seen as the accepted definition for a quality start.

Given a 5 man rotation, that works out to about 32 starts each.
I'd want the nos. 1 and 2 to produce 25 or more, a decent number 3 to earn 20+ and would want 15 or more from the others. And I'd want more than just 6 innings from the 1 and 2.

Later

[Frayed Knot]

Overall I believe that around 25% of all starts qualify as quality starts (I believe Dickshot has the actual numbers on this. I suspect most fans think it's higher than is actually is)

Obviously the majority of those would come from front-of-rotation guys and much fewer from the back end - although I have no idea of the breakdown.

[Guest Johnny Dickshot]

Around 33% of all games started meet the specificications (6 IP, 3 ER or less). Teams whose starters get a QS win 67% of the time.

[Guest 86-Dreamer]

if someone throws six scoreless innings, then allows 4 runs in the seventh, was that a quality start? if not, it is a very weak statistic.

[Guest Edgy DC]

No. It's the final line that matters.

[Elster88]

86-Dreamer wrote: if someone throws six scoreless innings, then allows 4 runs in the seventh, was that a quality start? if not, it is a very weak statistic.

I disagree.

Final line is all that should count. How is 4 runs in 7 innings a quality start? Does it really matter if they were spread equally or not? The fact that 6 innings could ever be a quality start is pushing it as it is.

[Theoldmole]

So back to my initial question, which I think is a good one...

[Frayed Knot]

Well, as mentioned, I think the biggest problem in finding data on that is figuring out what's a #1, what's a #2, etc.

But, using Dickshot's 33% norm (NORM!!!!!) I'd guestimate that you should probably expect half or more from front of the rotation guys, while treating any that you get from the #4/5 dudes as if found money.

[Guest 86-Dreamer]

Elster88 wrote: ="86-Dreamer"]if someone throws six scoreless innings, then allows 4 runs in the seventh, was that a quality start? if not, it is a very weak statistic. I disagree. Final line is all that should count. How is 4 runs in 7 innings a quality start? Does it really matter if they were spread equally or not? The fact that 6 innings could ever be a quality start is pushing it as it is.

This "stat" is too dependent on manager's decision. why should a pitcher who gets pulled after allowing 3 runs in 6 innings be credited with delivering a performance better than a guy that pitches six shutout innings but stays in? It is a hokey stat as presently defined.

[Elster88]

I, for one, don't think 6 IP-3 ER should qualify.

But I think 6 IP-2 ER should and 7 IP-4 ER shouldn't, manager's decision on when the guy comes out nonwithstanding.

To me, the stat should mean that your team has a good chance to win the majority of games based on the innings pitched and runs given up by the starter. 7 IP 4 ER doesn't qualify in my book, no matter when the runs were given up or when the manager decided to give the starter the hook.

[Guest Edgy DC]

="Elster88"]To me, the stat should mean that your team has a good chance to win the majority of games based on the innings pitched and runs given up by the starter. 7 IP 4 ER doesn't qualify in my book, no matter when the runs were given up or when the manager decided to give the starter the hook.

Well, according to Dickshot (though I don't know his source), they do win 67% of the time.

[Guest 86-Dreamer]

This article has data from 1984-1991. It shows that for all its weaknesses, the QS has a very high correlation to wins and losses, and that very few QS's are the 6IP/3ER minimum standard. It also shows how good of a 1-2 Gooden and Darling were ...


http://www.diamond-mind.com/articles/qstart.htm

[Theoldmole]

The reason why I chose it as a stat is -- you have to choose something. And it seemed relevant to my question. A quality start means the guy pitched far enough into the game to get to your short relief men, and when he left, your team was still in the game.

And I know there's not much difference between a 1 and a 2 -- you should probably expect some consistency of quality from your top 2 guys. But there's presumably a difference between a 1/2 and a 4/5. So when you say, as everyone does, "We need more pitching," how does the more pitching you're likely to get compare with the pitching you have? If an Alay Soler, for example, gives you one quality start out of every three outings (just picking numbers out of thin air), should you figure that this is way below average, and that you can reasonably expect to go out and find a bottom-of-the-rotation guy who'll give you two out of five? And how much is it worth to you to give up, in order to boost the bottom of your rotation from 33% efficiency to 40% efficiency?

[Theoldmole]

Here are some numbers from Diamomd Mind that establish benchmarks for the top or the rotation...still looking for bottom-of-the-rotation benchmarks.[/url]

on edit: Sorry -- I see Dreamer gave us the same link.

[Elster88]

I like that article.

[Elster88]

="Edgy DC"] ="Elster88"]To me, the stat should mean that your team has a good chance to win the majority of games based on the innings pitched and runs given up by the starter. 7 IP 4 ER doesn't qualify in my book, no matter when the runs were given up or when the manager decided to give the starter the hook. Well, according to Dickshot (though I don't know his source), they do win 67% of the time.

I pretty much just chose numbers that "sound good" when I came up with my list of what "should" qualify for a QS. Not very scientific. I'm rethinking it after reading that Diamond Mind article.