Quality Starts

[brewtown82]

Why all the hate for the "Quality Starts" stat? I mean, six innings and three runs is going to keep you in a ballgame. What other metric should we use to judge this? Is it because it doesn't take into account the defense behind him or the number of walks issued?

[Invader3K]

I think it's a nice stat for tracking how the progress of a season is going, but not necessarily judging a pitcher's real value...maybe others have a different opinion.

[kramnoj]

QS is a stat in limbo. Nobody that is hardcore into sabrmetrics will use it, and it doesn't have much support among regular fans, because they think it means that a pitcher with quality starts will have a 4.5 era. Of course, in the higher offense era, a pitcher with 4.5 ERA isn't bad anyway.

 

I think QS does a good job with it's goal: how many times does a pitcher do a decent job of keeping his team in the game.

[pitchleague]

is it 3 runs or 3 earned runs? I'm having a mind blank right now. If it's 3 runs, I can see it as a quality start. If it's 3 earned runs, I can not, since the score could be 8-1 because you gave up 5 unearned runs.

[GormanHarvey]

3 ER

[TheCrew07]

Well you can give up 6 unearned runs like Suppan did last year and still get a quality start but your team isn't in the game. At some point you have to stop the bleeding if you're a pitcher. It's not bad, but it just doesn't mean anything to me because the better the pitcher the more quality starts you're going to get out of him, it doesn't tell anything we don't already know.

[Ennder]

I'd use something like game score to track how good starts are. You can go 6 IP, pitch terribly and only give up 1 run. I also have a problem with any stat that suggests 9 IP with 4 ER is worse than 6 IP and 3 ER or that 6 IP and 3 ER is better than 5.2 IP and 0 ER.

[rluzinski]

It's a quick, dirty way to see how consistent a pitcher has been. If he has a 5.0 ERA but 5 of his 6 starts has been a quality start, he's been a decent pitcher for his team. A 4.5 ERA starting pitcher is worth about $8 mil a year, so apparently teams value an average starting pitcher a heck of a lot.

[brewtown82]

I'd use something like game score to track how good starts are. You can go 6 IP, pitch terribly and only give up 1 run. I also have a problem with any stat that suggests 9 IP with 4 ER is worse than 6 IP and 3 ER or that 6 IP and 3 ER is better than 5.2 IP and 0 ER.

 

I like your reasoning here. Although, I think pitchers don't pitch far enough into games anymore, so I don't mind the six inning thing. But you're right, 4 runs in nine innings is better than 3 in 6. I guess you just can't take shortcuts when statistically evaluating pitchers. You have to analyze everything as a whole.

[bani1717]

How about this: Quality Start = 6+ IP and an ERA of no higher than 4.5 for the game by itself. 6-7.2 IP and 3 ER gets it done. So does 8+ IP and 4 ER.

 

If you wanted to get stricter about it, you could even lower the required ERA for that game to something like 4.25 or 4.00. (Lowering it to 4.00 would require 6-6.2 IP and 2 ER, 7-8.2 IP and 3 ER, or 9 IP and 4 ER.)

[Ennder]

I don't think ER should be the only thing you look at was part of my point. The other day Burnett pitched 6 IP with like 8 H and 6 BB and gave up 1 ER. I'm not someone who calls that a good start, that is a lucky start, it is bad hitting, it is a good result, but it is not a quality start. If he pitches like that every start he will have horrible results overall. QS is kind of like looking at AVG, it is sort of useful and very easy to figure out just by looking at a box score, but it doesn't tell you all that much.

 

Something like game score is too complicated to figure out on the fly which is a big downfall but it does tell you a whole lot more about a start than QS does, it is kind of like OPS (I assume most people can't just look at a box score and figure out OPS in their head at least!). Not saying game score is perfect, just pretty readily available.

[ILuvDaBush]

Isn't there a Quality+ Start? If not, there should be. 7IP/2ER. That is quality. Hand the ball to your team's top set-up man (while simultaneously preserving the best bullpen weapon, the LOOGy) and then, if all things go according to plan, the closer (because even if it's tied or close to being tied, you should have your closer out there then). An ERA of 4.50 doesn't appeal to me.

[jeffy]

Isn't there a Quality+ Start? If not, there should be. 7IP/2ER. That is quality.

What if he gave up 5 "unearned" runs because of two out error followed by 3 hits? The problem with earned runs is it sometimes gives the pitcher a pass for some bad pitching just because an error occured in an inning. I think there should be some consideration given for things that happen due to errors, but "earned runs" sometimes gives too much of a break to the pitcher.

 

Why doesn't this work the other way? Using ERA logic, after 3 batters come to the plate in an inning any errors should not count against the fielder, because the pitcher should have gotten three outs and the inning should be over.

[brewerjamie15]

Gotta go with the flow here and call "quality start" a misnomer.

If a pitcher gives up a ton of walks, around 8 hits, trouble the entire time, and 5 runs (3 earnies, 2 on by walks or something) how is that "quality" pitching?

If it were 3 runs or less,and less than 4 walks ro something like that, I'd be better about the quality of a quality start.

[ILuvDaBush]

So does the argument hinge on the ease of his pitching or the fallibility of the "Earned Run"?

[brewerjamie15]

For me I guess perhaps both? I hadn't thought about the earned run loop holes. 2 outs: runners all over the place, an error occurs that should have ended the inning. Pitcher then can not get the third out until he allows half a dozen hits, and maybe 4 runs to score. All unearned, yet he's still eligible for a quality start if he gave up less than four earned runs but 7-8 runs scored in 6 IP?

[dadofandrew]

If a pitcher had all quality starts, his ERA would probably be close to 3 or under. It's certainly not the be all, end all of stats, but it does help show consistency and durability.

 

Suppan has 5 of 6 QS's in a row, as he missed one by one out, and his ERA is 2.92. Many let the "worst case" bother them, but the irony there is, a guy who throws 200 innings and has a 4.50 ERA makes about $10M a year. I wish I was that guy.

[1992casey]

Earned runs is definitely a fallible stat. The scenarios brewerjamie15 and jeffyscott describe illustrate issues involved. That's why saber people are interested in perfecting stats like FIP. By addressing how many runs would "normally" be scored given a certain number of Ks and BBs, a better picture can hopefully be painted.

 

The other issue with earned runs is that defense does affect it. If a player doesn't touch a ball, even though an average defender would have gotten to it, no error is charged. The runner reaches base due to poor defense rather than poor pitching, yet, the pitcher is charged with the earned run should he score. Of course, the competence of the official scorer comes into play, too.

 

Over large samples, however, flaws with the earned run stat will hopefully "even out" a bit, making it more accurate.

 

Adding to what Russ says, there's value to "quick and dirty" stats, provided they're not totally abused. Ease of use is nice. http://forum.brewerfan.net/images/smilies/smile.gif

 

Regarding the quality start stat itself, I think there might be less objection to it had it been given a different name.

[areacodes]

is it 3 runs or 3 earned runs? I'm having a mind blank right now. If it's 3 runs, I can see it as a quality start. If it's 3 earned runs, I can not, since the score could be 8-1 because you gave up 5 unearned runs.

3 ER

That is wrong. It is 3 runs' date=' not 3 earned runs.

 

From baseball-statistics.com:

QUALITY START: A game started in which a pitcher lasts for six innings or more and allows three runs or less.

 

From BP:

A "quality start" is defined as one where the starting pitcher goes 6 or more innings and allows three of fewer runs.

[endaround]

If a pitcher had all quality starts, his ERA would probably be close to 3 or under. It's certainly not the be all, end all of stats, but it does help show consistency and durability.

 

Suppan has 5 of 6 QS's in a row, as he missed one by one out, and his ERA is 2.92. Many let the "worst case" bother them, but the irony there is, a guy who throws 200 innings and has a 4.50 ERA makes about $10M a year. I wish I was that guy.

 

I wish Suppan was that guy

[JohnBriggs12]

I don't think ER should be the only thing you look at was part of my point. The other day Burnett pitched 6 IP with like 8 H and 6 BB and gave up 1 ER. I'm not someone who calls that a good start, that is a lucky start, it is bad hitting, it is a good result, but it is not a quality start. If he pitches like that every start he will have horrible results overall. QS is kind of like looking at AVG, it is sort of useful and very easy to figure out just by looking at a box score, but it doesn't tell you all that much.

 

Something like game score is too complicated to figure out on the fly which is a big downfall but it does tell you a whole lot more about a start than QS does, it is kind of like OPS (I assume most people can't just look at a box score and figure out OPS in their head at least!). Not saying game score is perfect, just pretty readily available.

That's a bad example. Burnett isn't a typical starting pitcher. Guys in his salary range get longer leashes. The vast majority of starters aren't going to get anywhere near 6 innings having walked 8 guys. Conversely that he did implies he pitched out of several jams and that's a sign of a quality pitcher too.

 

I think the stat has value in the case of a guy who's ERA is high because of a few really bad starts. If that same guy records an average percentage of quality starts, then he's pretty much an average starter. It's not a perfect stat but no stats are perfect.

[Ennder]

Conversely that he did implies he pitched out of several jams and that's a sign of a quality pitcher too.

 

It can be, but in an individual game I don't really think it is. If Suppan has that same game it isn't a quality game either and he pitches out of those jams a whole lot less than a good pitcher.

 

I guess it depends on what you are using quality start to judge. If you just want to know how good a pitcher pitched in his career you can mostly use ERA which will tell you more than QS, over a season some sort of composite ERA is probably a better thing to look. The entire point of QS seems to be trying to judge a pitcher on a single game and I think it fails pretty miserably at that. As a measurement of how consistent a pitcher is it is pretty flawed as well because of all the little flaws people already pointed out.

 

QS is 3 ER as well for most sites at least, maybe a few attempted to change it and make it all runs but the standard definition is ER.

[Thurston Fluff]

I also have a problem with any stat that suggests 9 IP with 4 ER is worse than 6 IP and 3 ER or that 6 IP and 3 ER is better than 5.2 IP and 0 ER.

 

While true the quality start measures both how many runs and how deep in the game a pitcher goes. I think how deep into a game a pitcher goes is of some importance. I understand it is only 1/3 of an inning but this particular stat has to have a cutoff point and 6 is that point.

 

To me QS is not very useful in telling/predicting anything about a pitcher but can be informative about a rotation as a whole. The Brewers are a perfect example. They lead the league in quality starts and that is very telling IMHO of why they are currently leading the division. They don't have great pitching but they have consistent pitching. I don't think any other stat shows how consistent or how deep a rotation as a whole is. A couple great pitchers can skew the combined numbers of the staff and cover the fact that the team trots out two crappy pitchers on a regular basis. QS as a team stat measures something other stats don't.

Naturally having pitchers who have more QS leads to a staff with more of them, so in that small way, I think they can be useful if looking to build a staff. If a GM had a good offense but needed pitching he find QS is useful in finding a couple consistent bargain rate pitchers who will help more than finding a great one at the same price.

[TuesdaysWithRillo]

I think the Quality Start stat is a good gauge for pitchers in terms of eating innings and a pitcher keeping a team in the game, but other than that, I wouldn't measure a pitcher's value too much off of the statistic. Although a 4.50 ERA does have a lot of value in today's game. That could be the equivalent to a #3/4 type starter.

[CheezWizHed]

Pros:

1) Simple, easy to calculate stat that reporters and commentators can use for the common fan.

2) Shows some measure (albeit a weak one) of consistancy.

 

Cons:

1) Very basic and fixed measurement at 4.5 ERA.

2) Doesn't handle outliers well.

 

Basically, its good for newspapers and TV commentators to give a quick and dirty glance at how a pitcher is doing. Don't use it for anything more and its fine....