I'm having a problem with the speed-only guys mucking up my per-unit auction values. There's an old thread I always go back to that Elijah, LBJackal, Bloody Nips, myself, and other posted in, but it does not offer a good way to account for this problem.
Whenever a Figgins/Pierre/Bourn/whoever slips into the "replacement player" range, their SB stat wildly skews the avg replacement player's SB number up, causing starters' values to get whacked b/c they are viewed as having garbage SB contributions relative to the skewed replacement. If I try to finesse the range of players used to calculate the replacement, I have to re-sort and often the SB player will fall back into the replacement range. I've considered dropping these types of players from my draft universe entirely since I typically don't draft them anyways (6x6 w/OPS), but I'm afraid that will cause someone like Crawford to be way overvalued and 30-SB guys to be moderately overvalued. Maybe this is a good thing since I'm totally knocking the SB-only guys out of consideration, but then power players would be relatively undervalued and I won't bid them up enough.
You can weight the categories to account for the skew. The recommended split that I've seen work is bumping HR up by about 35% and dropping SB by 25%. BA was also dropped by 20%, and RBI bumped up 10%. These are arbitrary weights that were concluded by people far smarter than me, so I can't fully explain the way they were derived. Using a 6x6 scoring system would obviously affect things, but it's a starting point at least.
StlSluggers wrote:Why'd you dump the Jackal account, man?
Not really sure. Felt like forever since I'd been here when I came back and figured I'd get a fresh start. Can't believe it's almost been 8 years since I first joined... I was a teenager! Went and dusted off the avatar for old time's sake.
StlSluggers wrote:Why'd you dump the Jackal account, man?
Not really sure. Felt like forever since I'd been here when I came back and figured I'd get a fresh start. Can't believe it's almost been 8 years since I first joined... I was a teenager! Went and dusted off the avatar for old time's sake.
This doesn't matter if the guys you select are truly replacement level. Somebody like Bourn skews the replacement level for SB up, but also skews HR & RBI way down.
mblax10 wrote:This doesn't matter if the guys you select are truly replacement level. Somebody like Bourn skews the replacement level for SB up, but also skews HR & RBI way down.
True, but its the magnitude difference between the two that causes the problem. Depending on the dispersion sometimes I get an SB value of over $2 each.
Skin Blues wrote:You can weight the categories to account for the skew. The recommended split that I've seen work is bumping HR up by about 35% and dropping SB by 25%. BA was also dropped by 20%, and RBI bumped up 10%. These are arbitrary weights that were concluded by people far smarter than me, so I can't fully explain the way they were derived. Using a 6x6 scoring system would obviously affect things, but it's a starting point at least.
Sincerely,
Skin Blues aka LBJackal
Good to have you back. I do skew my weights some, which helps but depending on the dispersion it can still get out of whack. Here's what I've been using:
R - 16% HR - 21% RBI - 16% SB - 10% BA - 16% OPS - 21%