A t-test isn’t a bad idea, I’ll see if I can run one. It’s been awhile since I messed with them, my last stats class was also a couple of years ago!

@OGC

Thanks for the many recommendations. I’ll let you know as soon as I post my revisions so you can look over them.

On this:

I would also add, yes, small samples is involved, plus Tomko pitched in LA and SD after us, two of the most pitching-biased parks in the majors, so his numbers might be skewed by playing there, though the parks have not reduced walks as much as AT&T has in the past three seasons (according to Bill James Handbook’s stats), in fact, LA inflated by 10% vs. the road.

You’re that park effects can influence walks, but most of the reading I’ve done on the subject suggest that the effect is very minimal, it exists, but I’m not sure how much it would change the numbers. I’ll have to think about if I want to correct for PF’s or not.

Thanks all.

What he found was that when you have plus strikeout totals, you can have a horrible walk rate and still your ERA would be better. It makes sense. The less balls in play there is, the less hits there are, balanced by the walks. But walks don’t normally drive in runs, probably resulting in the lowered ERA relative to other good pitchers. That’s basically Cain’s thing thus far in his career, low hits balanced by extra walks.

So, what I would add to your graph would be K/BB, probably both ranking in the league so that it goes onto your nice graph, plus an array below showing what the actual number was each year. K/BB is a better measure of how good the pitching talent is; BB/9 is just part of it.

And I just noticed that you just ranked by total walks; for many of those years, the Giants were good and thus we had to pitch the 9th, but not the other team, meaning we might have more IP than other teams for some of those years. I would calc BB/9 for each team, then rank; it could still be the same as you have now, but it would improve your methodology. Same for K/9 (or BB% and K% if you prefer them) vs. total strikeouts.

I would also add, yes, small samples is involved, plus Tomko pitched in LA and SD after us, two of the most pitching-biased parks in the majors, so his numbers might be skewed by playing there, though the parks have not reduced walks as much as AT&T has in the past three seasons (according to Bill James Handbook’s stats), in fact, LA inflated by 10% vs. the road.

If anything, I think the Giants try to discourage the starters from nibbling. I don’t know how many times I’ve seen Giants managers say “just throw strikes” to the starters in print. Particularly with Matt Cain, he has always insisted on nibbling when his stuff is filthy and each year around now, he finds religion and starts throwing strike one with the first pitch, instead of fooling around, and suddenly he is untouchable.

Not a bad idea, but minor league stats are hard to come by sometimes and I’d prefer to keep this in the majors if I can.

@fireplacetv

Good call, that would help fill in some gaps. I’ll definitely do that and it would be interesting for guys like Estes in particular, who pitched several seasons before Rags and several after.

Great suggestion, I’ll add a update later tonight/early tomorrow.

Reason I ask is it could give a larger study group and use of other organizations pitchers as a control. Of coarse it could be too unwieldy or the data just not available.