8 replies to this topic

[Luke]

When we look at the statistics for pitching prospects, we want to see evidence that they are able to control a game. We want to see signs that they have an out pitch, the pitches to set up that out pitch, that they can pitch out of trouble, and so forth.

Once upon a time the number we used to judge things of that nature was strikeouts per nine innings (K/9), or the K rate. If a pitcher has a high K/9, then that pitcher is clearly able to strike out a lot of hitters and therefor has an out pitch, can strike their way out of jams, and so forth.

But there's more to pitching than striking people out. In fact, the goal of pitching is simply to get outs without putting people on base (or letting runs score, but runs can't score without people getting on base). In that sense, a weak ground ball to short is as good as strikeout. If that ground ball can be turned into a double play, it's better than the strikeout in a lot of ways.

That suggests we should seriously consider GO/AO (Ground Outs / Air Outs) when considering pitching prospects... and I do (a lot) ... but that stat doesn't really cover everyone. Matt Garza, for example, wasn't much of a ground ball pitcher before he came to Chicago and he was still pretty good. Likewise, a ground ball pitcher with a Low-A level infield defense behind him is going to look worse than he is due to the flubbed plays and errors out of short and second.

And that brings me back to WHIP ( [Walks+Hits] / Innings Pitched). If a pitcher has a nice low WHIP, then runners are not consistently reaching base against that pitcher. That, in turn, implies that the pitcher has the pitches needed to put hitters away. That's not quite the same thing as having a great out pitch, but it is pretty close.

Lately I've been seeing a trend in the Cubs farm system towards low-WHIP, low-K/9 outings by a number of the Cubs pitching prospects. The old way of thinking would frown on this and be worried that it means the Cubs pitching prospects just aren't all that good based on the lower K/9. I'm not so sure. After all, if runners aren't getting on base by a walk or a hit, then the pitcher is doing his job pretty well. In fact, I'm wondering if the Cubs coaching is not telling the pitchers to just get outs by any means necessary. Strikeouts, pop flies, ground balls, alien abductions... it's all good. Just get outs. That puts a different spin on the statistical evaluation of pitchers and suggests quite strongly that WHIP is the better prospect indicator statistic.

I know we have quite the community of stat junkies here, so I thought I'd open the question up for debate. When evaluating pitching prospects based on nothing but numbers, which number would give you more weight: WHIP or K/9? Naturally, we're going to consider all of them in any evaluation, but which one do we look at with the most interest (no matter how slight)?

Currently I'm in the WHIP camp (when I don't live in GO/AO land), but I'm not committed to it.

[Fishin Phil]

WHIP makes more sense to me. I don't care how they get them out, as long as they get them out.

[Cubbie Blues]

I would rather see weak ground ball than strike-outs. The pitchers pitch count can stay lower than going deep into a count to get a K. Ground-outs aren't as exciting though. So, for the casual fan, they are alway going to look at the K/9.

[TWC]

But there's more to pitching than striking people out. In fact, the goal of pitching is simply to get outs without putting people on base (or letting runs score, but runs can't score without people getting on base). In that sense, a weak ground ball to short is as good as strikeout.

I would think that the statistical likelihood of a batter reaching base on a defensive miscue on a strikeout (PB/WP) is far, far, far lower than the statistical likelihood of a batter reaching base due to a defensive miscue on even the weakest of ground balls. Many more variables are introduced once the ball goes into the field of play. I'd much rather see the strikeout, every time.

For evaluating prospects, though, I'd look first to WHiP, and then to the K/9.

[Drew]

I know it seems like a cop-out, but I think you have to look at both evenly.

For instance, looking first at K/9 and seeing any number of our Hendry-signed prospects, one might get pretty excited seeing some high numbers. However, most of these guys have great "stuff", (Dolis comes to mind) but sometimes have a hard time throwing strikes.

Now, if look to WHiP 1st, the above problem is solved since it takes into account walks as well. The problem I have with WHiP is it's inability to distiguish an XBH from a seeing-eye single. Since we all know every hit is certainly not created equal, this causes a problem.

Solution? There's no sure-fire answer, but I'd have to say choice #3: K/BB Ratio. No one stat is perfect, but I feel this one is the best way to combine the two.

[hansman1982]

I do love me some WHIP and I think it is a grossly underrated stat; however, for me, I like BB/K ratio more than just the straight K rate. I don't care if a guy is striking out 27 per 9, if he is also walking 27 per 9. Conversely, if he is only striking out 6 per 9 but walking 2, then yippie!

[Luke]

Conversely, if he is only striking out 6 per 9 but walking 2, then yippie!

Unless his opponents are batting .412 and slugging .670 off him. That's the downside of SO/BB. By itself (and yes, its unrealistic to consider any of these stats in a vacuum... but hey, it's the internet) SO/BB could indicate a guy with great stuff, or a batting practice pitcher.

[T C]

I much prefer K% or K/9.

When I read about pitchers in the low minors, I want to see them missing bats. I want to see pitching prospects clearly dominating hitters, especially at the lower levels. Plenty of guys can induce weak contact against shitty hitters (especially in low levels where any sort of mediocre offspeed pitch will fool 90% of batters), but its a rare and important skill to shut them down by striking them out. Walks and hits can be worrisome, but those are things that can be corrected. Raw strikeout stuff, though, cannot be taught. None of these low k/9 guys the Cubs farm is showcasing is going to turn into a dominating pitcher, and the lack of strikeout stuff means all of them will be below-average if they even make the majors.

EDIT:

"Strikeouts, pop flies, ground balls, alien abductions... it's all good. Just get outs. That puts a different spin on the statistical evaluation of pitchers and suggests quite strongly that WHIP is the better prospect indicator statistic."

The difference here is that WHIP is telling you how they're getting shitty A-ball and organizational filler type players out, while K/9 is telling you (hopefully) more about how a player will get players out when he reaches the majors, which is how we should look at prospects.

If Joe Pitcher is inducing a ton of weak contact, will major league hitters be fooled as much by that Joe's stuff? Almost certainly not. MLB'ers will be more patient, or barrel the ball significantly better. And Joe's good pitches wont induce nearly as much weak contact, his mistakes that he gets away with in the minors will be killed by major leaguers.

Compare him to Timmy Flamethrower, who's k'ing 13/9 and walking 5/9. Against MLB'ers, a lot of his k's will become weak contact, and if he gets his walks under control, he could be useful. The k's show that he will miss a lot of the barrel if he hits the big leagues, and that is how a lot of pitchers succeed.

[NormB]

I much prefer K% or K/9.

When I read about pitchers in the low minors, I want to see them missing bats. I want to see pitching prospects clearly dominating hitters, especially at the lower levels. Plenty of guys can induce weak contact against shitty hitters (especially in low levels where any sort of mediocre offspeed pitch will fool 90% of batters), but its a rare and important skill to shut them down by striking them out. Walks and hits can be worrisome, but those are things that can be corrected. Raw strikeout stuff, though, cannot be taught. None of these low k/9 guys the Cubs farm is showcasing is going to turn into a dominating pitcher, and the lack of strikeout stuff means all of them will be below-average if they even make the majors.

EDIT:

"Strikeouts, pop flies, ground balls, alien abductions... it's all good. Just get outs. That puts a different spin on the statistical evaluation of pitchers and suggests quite strongly that WHIP is the better prospect indicator statistic."

The difference here is that WHIP is telling you how they're getting shitty A-ball and organizational filler type players out, while K/9 is telling you (hopefully) more about how a player will get players out when he reaches the majors, which is how we should look at prospects.

If Joe Pitcher is inducing a ton of weak contact, will major league hitters be fooled as much by that Joe's stuff? Almost certainly not. MLB'ers will be more patient, or barrel the ball significantly better. And Joe's good pitches wont induce nearly as much weak contact, his mistakes that he gets away with in the minors will be killed by major leaguers.

Compare him to Timmy Flamethrower, who's k'ing 13/9 and walking 5/9. Against MLB'ers, a lot of his k's will become weak contact, and if he gets his walks under control, he could be useful. The k's show that he will miss a lot of the barrel if he hits the big leagues, and that is how a lot of pitchers succeed.

I'm with TC.
% is slightly better than 'per 9'.
I don't look at WHIP *at all* for prospects.
-
Not part of the question, but I look at:
BB's, K's (both per 9 and %)
HR/9 (since there isn't a HR/FB rate available, I stick with HR/9)
Groundball % (if available)
BABIP (only to see if it's extremely high or low)