How much should we really be looking at FIP and xFIP??

[Ender]

LD rates are nowhere included in SIERA right? I think it includes GB rate and FB rate and I'd love one stat that included the 3 of them. Something like FIP but where BABIP included these 3 things.

I think they leave out LD% because the year to year correlation is really low, it is 0.40 only.

[Skin Blues]

If using LD% made an improvement, I'm sure it would have already been included. A quick google search shows that LD% only has a 0.11 correlation from year to year, versus GB% and FB% which have Y2Y correlations of .85 and 0.81 respectively. That means incorporating LD% probably does more harm than good, and explains the omission.

[Ender]

Yeah what he said, LD was listed at 0.40, didn't notice it was missing the % so I gave the wrong number.

[jorgesca]

[quote="Skin Blues"]If using LD% made an improvement, I'm sure it would have already been included. A quick google search shows that [url=http://www.beyondtheboxscore.com/2012/1/9/2690405/what-starting-pitcher-metrics-correlate-year-to-year]LD% only has a 0.11 correlation from year to year[/url], versus GB% and FB% which have Y2Y correlations of .85 and 0.81 respectively. That means incorporating LD% probably does more harm than good, and explains the omission.[/quote]

Yeah I'm sure it would have already been included, and as a predictor of future ERA this might be better, but should it be included when analyzing past performance? a year worth of performance? I think a year worth of data on balls in play with a higher than usual LD% and a higher than usual BABIP would just indicate that his stuff was just not good and it was hit hard, not that he had bad luck. Just some random thoughts, I always feel something is missing in all of these statistics.

[Ender]

LD% fluctuates wildly year to year, you need a good 4 or 5 year sample for it to mean very much. That is why it shows such a small year to year correlation.