John Franco solves pitching questions

[scorekeeper]

This should show all you SABR guys that you ain’t such hot stuff. Heck, even ex-ML pitcher John Franco can come up with meaningful metrics! ;)

http://www.sportsblurb.com/baseball/stats.asp

[SABR Matt]

LOL

Moron.

[honus14]

Other than that, vulture wins appear to be a crapshoot.
[snip]
It is also worth noting that starting pitchers with high win-efficiency numbers tend to find their performances unsustainable.


Uh, John? You've just shown your metric to be almost totally due to chance. Thanks for playing though. You can pick up your lovely parting gifts just offstage.

[scorekeeper]

I have to apoligize! I didn't know that he changed his article.

The one I was referring to was the following.



Statistics 101
By John Franco
04/12/07
Wasting Pitches
With all the focus on pitch counts in our era of statistical overload, it is important to consider where those pitches actually go. The impact of pitch counts is well known: excessive pitch counts lead to injuries (Kerry Wood, meet Mark Prior), while pitchers who rack up high pitch counts in few innings are less likely to stick around for a win.

Four balls equal a walk, and three strikes equal a strikeout (unless A.J. Pierzynski is batting, in which case any combination of balls and strikes can equal any result he wants), so those numbers establish a baseline for the number of pitches a pitcher “must” throw. Another pitch is required for each out (more or less) or for each hit. So, in a typical outing where Daisuke Matsuzaka records 11 outs on balls in play, 10 strikeouts, 1 walk and 6 hits, he must have thrown at least 51 pitches to achieve those results (11 + 30 + 4 + 6 = 51). The rest of the pitches can be considered “extra” pitches that did not contribute directly to the results of each at-bat. Given that Matsuzaka threw 108 pitches in that game, he threw 51 necessary pitches and 57 unnecessary pitches, giving him a “meaningful pitch percentage” of 47.2 percent. Perhaps a better way to think of this number is a measurement of “action pitches” where something happens. In this case, something happened on 47 percent of Matsuzaka’s pitches: a ball in play (hit or out), a walk, or a strikeout.

Isn’t this just another meaningless statistic in a world that already has far too many, you might ask? Perhaps, but consider that meaningful pitches have a moderate positive correlation with both ERA and WHIP. In other words, a pitcher who makes things happen is likely to have a better WHIP and ERA than a pitcher who throws a lot of pitches without generating a lot of offense. (For those of you inclined to try the math yourself, you will notice that the correlation between ERA/WHIP and meaningful pitches has a negative sign. This means that they tend to move in opposite directions, but remember that a lower ERA is better, so ERA moving in the opposite direction of another indicator is a good thing). Unfortunately, there is a weak negative correlation with wins, meaning that pitchers who generate a lot of walks and strikeouts are less likely to stick around to get a win despite their chances for a better ERA and WHIP.

Naturally, now that I’ve argued the benefits of pitchers who rack up high walk and strikeout totals, it is time to point out that the highest ranked pitcher in baseball, in terms of making things happen during the 2006 season, was probably the man known for high pitch counts: Carlos Zambrano. Zambrano led all pitchers with more than 28 starts by using 43 percent of his pitches towards accomplishing his end results. As the only pitcher in baseball with over 100 walks and 200 strikeouts, Zambrano may not have been efficient in retiring hitters, but he made his pitches count towards his final numbers. At the opposite end of the spectrum, Paul Byrd walked just 38 hitters and fanned just 88, so a mere 32 percent of his pitches counted towards his numbers despite allowing 232 hits in 179 innings. Apparently most of those hits were surrendered after Byrd had already thrown a few pitches to the batter, and the result was less than a one-third chance that something would happen on a given pitch.

The rest of the list of pitchers who were efficient in making things happen is a who’s who of exciting power pitchers, especially if your definition of exciting includes the occasional seven-walk, eight-run performances. Numbers two through ten include Scott Olsen, Jeremy Bonderman, Johan Santana, Ian Snell, Felix Hernandez, Matt Cain, Josh Beckett, John Smoltz, and Jake Peavy. The list of pitchers who need a lot of pitches to make something happen includes the usual soft-tossers like Byrd, but also includes interesting names like Brad Penny (35%), Curt Schilling (36%) and Bronson Arroyo (36%). The bottom line: it may be painful to watch your pitchers rack up the walks that go with their strikeouts, but it is much better than the alternative of watching them get to a 3-2 count on every hitter before giving up a single.

[Tango Tiger]

You should delete your cut/paste. The article is here:
http://www.sportsblurb.com/baseball/stats/stats70.asp

[digglahhh]

Save this link for the next time some pseudo-jock makes some comment like, "When it comes to determining who the best pitchers of all time I trust John Franco over Bill James."

[yank]

Four balls equal a walk, and three strikes equal a strikeout (unless A.J. Pierzynski is batting, in which case any combination of balls and strikes can equal any result he wants) At least he has a humor about it. And about a paragraph after that I decided I was wasting my time. Then I decided to read the last sentence...and thought to myself, I regret doing that.

[Ubiquitous]

I would argue that at times it is better to go 4-0 then then to try and work back from a 3-0 count. Today the Cubs had righty reliever who was pretty strong except for a line shot down the line for a triple to open up the inning. They then brought in Cotts the lefty with two outs to get the platoon advantage. So of course Tony brings in the righty and Cotts has to pitch to him. At the time I thought Ronnie had said that the count was 3-0 and I'm thinking I hoped to god he throws a ball low and away and he starts over on the next batter instead trying to throw one over for a strike. Then you can play around with your next matchup. It actually was a 3-1 count and So did in fact let it go by for a called second strike and So eventually bounced out. But to me the point is still valid. A lefty on the mound, a righty at bat and the pitcher behind 3-0 I'd rather have the next pitch be a ball and have a new match up then to have this battle continue with the pitcher trying to throw strikes.