Baseball Hall of Fame ballot released

Not at all. I understand what you're saying. But I only partially agree.

But as far as measuring performance, ERA is still pretty interdependent on bullpen and defense. You can't just say Silva was great in 2005 without acknowledging that the Twins hoovered up balls behind him that season, and that the Twins pen stranded a TON of runners for him. I know. I watched it happen.

Just because the season is done and the ERA is in the books doesn't mean that his ACTUAL performance was what his ERA was. It was a chimera, to a large degree.

That's why some people prefer K/9 to just about anything. It's completely up to the pitcher.

I don't necessarily prefer either of those. I like to look at all of them. I'm just saying, a case can be made for K/9 or WHIP or what have you.

 

I would agree that in general, one would prefer a pitcher to have a load of strikeouts. Let's say you have one pitcher that has nine strikeouts in seven innings, and one pitcher who has five strikeouts in seven innings. As long as the second pitcher hasn't induced two double plays more than the strikeout pitcher, the strikeout pitcher is a better bet to have had a better day. Plus, a strikeout pitcher tends to demoralize hitters more.

But....in the big picture, if the first pitcher consistently can go only five or six innings, he may not be as big an asset as the second pitcher if the second pitcher consistently goes seven and at times eight. And if the second pitcher rarely gives up homers and the first pitcher gives up a lot, then the strikeout edge may be negated.

And...walks are a whole 'nother issue.

So strikeouts are very important, but often aren't the full picture.

 

OOP,

I think we can agree that a pitcher with low strikeout totals can be lucky for a stretch, but will almost certainly be unable to sustain it. I have no problem giving the Cy to a hit-lucky pitcher if it is truly the most-effective performance in the league that year. A guy who goes 19-8 with a 2.75 ERA did that even if he had a four leaf clover taking root in his colon. But would you agree that when choosing between two pitchers of roughly equal performance for the award, deciding that one performance was the product of luck and superior defense while the other was based on better overall performance and would be more repeatable is a fair tie-breaker?

 

Jeff Ballard

 

I have no problem with ERA as a measure of effectiveness.

At some point, you have to draw an arbitrary line between effectiveness and predictive value.

The player on the team that wins had the best day, but just winning doesn't mean the player did things that are likely to produce wins in the future.

Avoiding runs (i.e. ERA) is the best thing a pitcher can do to help his team win, but defense-independent stats (Ks, BBs, avoiding HRs) are more predictive of how well he'll do at avoiding runs in the future.

At some point, pitch f/x and similar data are going to start giving us virtual Ks (i.e. they'll say a pitch with certain velocity and movement "should" produce a K a certain percentage of the time) and credit the pitcher with those, and maybe those will be more predictive than actual strikeouts.

At some point, you just let the performance stand for itself.

 

Bill Plaschke is trying to figure out why steam just started coming from his ears.

 

I got no problem with anything Dools or Pope just posted.

 

I'm not sure I would say that judging whether or not the performance is repeatable is the best tie-breaker, but if everything else is even, I would certainly look at the guy with more strikeouts.

Take Jonathan Sanchez as an example. Overall, he was pretty darn solid in 2009 and he certainly has the strikeout rate to indicate he can be even better, but do you really trust him to do that? I know I don't, because his control comes and goes. And when it goes, it REALLY goes.

I absolutely believe strikeouts are an important part of evaluating a pitcher, I just differ on which stat is most important.

 

That fits in with what I am saying. If you are evaluating past performance, I believe ERA is more useful. If you are projecting future performance, strikeouts are at least as valuable as ERA and probably moreso.

That virtual strikeout that you describe sounds like a horrible idea. Let's stick to what guys actually do rather than making educated guesses at what should have happened. Please.

 

But what if educated guesses are more predictive?

Bill James is the first to say that if statistical analysis does not first show the plainly obvious (i.e. the system shows Glenn Hubbard is a better player than Sandy Koufax or it doesn't show the Babe as one of the greatest players ever) then it is useless. But if potential strikeouts prove to be more predictive, why not follow the data where it leads?

 

You're a fucking charmer.

 

Pythagorean strikeouts....hahahahaha