Tavish wrote:

It doesn't change much how pitchers are valued amongst each other, although it could make the top MR more valuable than the top Closer. What I said was it changed the value between hitters and pitchers which it abolutely does.

I don't see how that can be true, Tavish, but would be interested in hearing the evidence. Adding a pitching category almost by necessity must change relative pitcher values.

But with respect to pitchers and hitters your statement, I think, was that it reduced pitcher values because pitchers were now only 4 category contributors. So, I'd like to see evidence not just that it changes pitcher/hitter values, but that it reduces pitcher values and does so because of the fact that almost no pitchers contribute to all 6 categories.

My point (and I have no hard evidence either) is that the valuation of a player depends both on the number of categories and the distribution of outcomes within that category. Assume a really silly example to see this---let every player with 100% certainty have the exact same result for all categories., including two categories where pitchers make no contribution. Every player will have the exact same value because they all contribute equally to the outcome.

So, it's not just the number of categories, it's how outcomes are distributed both across and within categories. I'd agree with your point if the distribution of results was the same within each category. But, each category's distribution leaves each player's statistical contribution that is different at the margin. The addition of holds, changes the distribution of within each pitching category, and I don't think you can say for sure how it impacts pitcher versus hitter value.