from Amazon
Friday, December 12, 2008
More updates at Fangraphs…
David rolls out all the Linear weights numbers I was discussing a few days ago: wRC, wRAA, wOBA. Each one is derivable from the other. wOBA you know, wRAA is runs above average, and wRC is runs above zero (i.e. Runs Created). To go from wOBA to wRAA, you subtract the league average wOBA, and apply a seasonal multiplier. To go from wRAA to wRC, you add in the league number of runs per PA times player PA.
There are a couple of issues to consider:
- park adjustments
- league adjustments
- pitcher/DH adjustments
At the moment, I think David just went the easy way, and did not apply any of these adjustments, and went with the overall MLB average for each year. (That’s how he was originally setup when he first launched Fangraphs.) There’s nothing wrong with that, as long as we are all aware that there are further adjustments that can be made. I’ll eventually get to it, but if someone wants to publish their own SQL to that effect, please do so, and we can discuss it.
He also changed “BRAA” to “RE24”. The calculation is the same, but the header changed. This makes it clear that this is runs above average, using the 24 base/out states. REW simply is the win version of RE24. I forget how David does the win multiplier, but looking at Raines’ page, it looks ok. Based on his previous posts, I think RE24 is Peter’s preferred metric.
In any case, it’s now all there. wRAA is the classic Linear weights (MGL’s preference). RE24 uses the 24 base/out states. WPA/LI uses the inning, score, base, out (deleveraged). And WPA is like WPA/LI, but leveraged.
Actually, one is missing: WPA/LI * boLI. This merges Peter’s favorite metric (RE24) with mine (WPA/LI), by giving greater weight to PA that actually have more impact. boLI is Leverage Index by base/out state. While I’m on the fence as to whether WPA/LI or WPA/LI * boLI is better for a hitter, there is no question that the latter is preferred for a pitcher.
WPA/LI treats each PA with a weight of exactly 1.0. This is fine for hitters, because we “want” each PA to count the same, because we don’t want to unfairly count a leadoff hitter’s PA less than a cleanup hitter’s, because it’s outside the hitter’s control. But, for pitchers, he positions his base/out states himself (for the most part… obviously, a bit of an issue with mid-inning relief changes).
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date