Called Strikes, Swiging Strikes and K%

[ssmallz]

How do you guys integrate called strikes into your player evaluation. They certainly have value as the article from pitcherlist below shows but my question is how much do you weight it as opposed to SwStk rate to better encapsulate a pitchers true effectiveness? I was thinking something like a 3:1 weight based on the low correlation between called strikes and K% and the difficulty of year to year consistancy. Any thoughts from the junkies?

https://www.pitcherlist.com/csw-rate-an-intro-to-an-important-new-metric/

[revo]

That's a good question that's above my paygrade. I hope someone here can answer this.

[Gregg]

Thanks for posting this. It is interesting on many levels.

[chancellor]

For fantasy baseball purposes, I play in traditional 5x5 leagues, so K's are important. I'd see this as a tool to help me predict if I think a pitcher will increase or decrease their K/9 rate; and then take that with the potential for more/less innings and have a good degree of confidence that a pitcher will improve, regress or stay about the same on total K's.

It's another handy tool in the toolbox for evaluating luck versus skill on something like BABIP. But if I'm looking at ratios, I'm not convinced this provides better feedback than hard hit rate, BABIP, barrel%, strand rate, walk rate, Ball%, and a couple others.

And I'll echo Gregg - thanks for posting this. Great food for thought.

And did anyone else notice the last name of the author?

[ssmallz]

I use these kinds of things mostly predraft when looking for players with K upside that other sites may have missed. It also helps me understand why players such as Aaron Nola have consistantly put up higher strikeout rates than would otherwise have been expected

[cavebird]

Fascinating article. The bit about foul balls was cool, too. The bit about Tyler Skaggs was sort of sad. My personal gut feeling would be to consider called strikes like swinging ones for pitchers if it has been consistent over a few seasons for the pitcher or if there seems to be a good reason it isn't statistical noise on a smaller sample size, like a very good curve ball.