Merry Christmas from Fangraphs: Wins Above Replacement!
FanGraphs more fully asserts itself as increasingly the most dominanty stat-oriented baseball site with the introduction of their own WAR system, now available on the players pages.
I'm not sure how you could read the sentence and still be here instead of there. Unfortunatly, I need to reach a limit, so here's what they've got:
Batting - wRAA (Runs Above Average) with a park adjustment.
Fielding - The sum of a player’s UZR.
Replacement - The replacement level adjustment set at 20 Runs / 600 PA.
Positional - The positional adjustment set using Tangotiger’s values (see this link)
Value Runs - The sum of Batting, Fielding, Replacement, and Positional.
Value Wins - Value Runs converted to a wins scale.
Dollars - Value Wins converted to the following dollar scale:
The last is just completely win. I don't exactly know what pitchers' RAA will be based on, I want to look in to it. I have my own thoughts on that (I really like pRAA based on tRA), and I prefer plus/minus to UZR (and looking at both is even better), but these are minor quibbles, this is the best baseball news this off season.
Josh Hamilton 2008: 3.8 Wins Above Replacement
This stuff will also soon be available on leaderboards and team pages.
36 comments
|
0 recs |
Do you like this story?
Comments
Interesting
that site is freakin’ awesome.
Do you or anyone else have any idea how to project that value calc for 2009 using either the Bill James or Marcels projections there?
G G G E-flat_______ F F F D__________....
I just about have it figured out
Although their park adjustments to wRAA are throwing me for a loop, because I don’t know if they’re using one-year or three-year weighted adjustments or from what source they’re pulling those park factors.
by Joey Matschulat on Dec 24, 2008 2:56 PM CST up reply actions
You're reverse engineering it? lol
"A ~.650 OPS from a COF should get you deported, not traded for."
- The Huntressatron
by inactive lsb user on Dec 24, 2008 4:12 PM CST up reply actions
Well, figuring out how the fielding, replacement-level and positional adjustments are calculated is easy enough
Value Runs divided by 10 appears to be about as close as you can get to Value Wins without FanGraphs divulging the actual number being used (it’s apparently done on a season-to-season basis, so there’s no one single number you can employ) — a solid ballpark estimate, in any event. Attempting to differentiate between 4.1 wins and 4.3 wins is where you begin to go overboard, anyway.
One thing that is readily apparent is that the Rangers were heavily penalized for their home ballpark in ‘08 — Bradley’s wRAA was 39.8, for instance, but after being park-adjusted it’s knocked all the way down to 31.7 (about a 20% drop, which is also observable with guys like Young and Hamilton).
by Joey Matschulat on Dec 24, 2008 4:24 PM CST up reply actions
Value Wins without FanGraphs divulging the actual number being used (it’s apparently done on a season-to-season basis, so there’s no one single number you can employ)
I’ll bet you they’re using Tango’s chart, and I’ll bet I can find it. I read it not even a week ago.
I meant to quote all this:
Value Runs divided by 10 appears to be about as close as you can get to Value Wins without FanGraphs divulging the actual number being used (it’s apparently done on a season-to-season basis, so there’s no one single number you can employ)
Here you go: it was from Tango himself at FanGraphs itself.
10 runs is a win: why?
The basic idea is that if you look at all teams in baseball history that have scored 1 more run than they allowed, per game (+/- 0.1, to increase the sample size), you will find that they have a .600 win. And similarly if they allowed one more run than they score, they will have a .400 win. That means each additional run leads to 0.100 additional wins, above the .500 mark. And 1 divided by .1 is 10.
The lower the run environment, the more impact each run has. And the higher the run environment, the less impact each run has. So, the 10/1 ratio is not fixed, but dependent on the run environment.
Here is a chart that shows the various win%, for various run environments, at various run differentials:
http://www.tangotiger.net/winactuals.html
Excellent
Props for the saber linkage. You’re the man.
"A ~.650 OPS from a COF should get you deported, not traded for."
- The Huntressatron
by inactive lsb user on Dec 24, 2008 8:34 PM CST up reply actions
Am I correct in reading that Michael Young's page
shows him as (significantly) below replacement?
That’s pretty shocking, to me. If anyone can explain that, I’m all ears.
He wasn't good.
Certainly the broken finger didn’t help, but his defense sucks and his offense isn’t really enough to make up for it, even at shortstop.
I’m shocked that you’re shocked.
I should've rephrased
I was shocked that the replacement value was the same for all positions, and a replacement player’s worth 20 runs. That just seems a little weird to me. Maybe I’m over-thinking this, but I expected to see different replacement values.
There is a positional adjustment.
But it’s done on defense instead of offense. The result is still the same, except at DH.
To elaborate further
the tie-in to salary data also throws me for a loop here. Are they saying that replacement level players are worth ~$10M annually? That seems like an exorbitant sum considering the talent pool.
I admit though that I don’t fully understand the win to value correlation, so maybe that would help.
Silva, Carlos
"A ~.650 OPS from a COF should get you deported, not traded for."
- The Huntressatron
by inactive lsb user on Dec 24, 2008 2:50 PM CST up reply actions
They are talking about
market value I believe, right? Because if you take into account all the pre-FA players that teams control for cheap, there is no way that is accurate.
By 2028, Mark Teixeira will be in the HOF.
-The Outlaw
Yeah, it's just
I have a hard time believing that a franchise would pay $10M per year for a replacement level player.
I posted a question per philkid’s suggestion, so we’ll see what they say.
He wasn't quite a good deal above replacement level when he got the contract.
That contract is ridiculous, though, even still.
Yeah, even before I really got into baseball stats
I remember thinking, “Isn’t he a little old for a contract of that length and value?” I assumed at the time it was the Rangers’ way of saying, “You’ve been a good soldier for the franchise, so here’s an overvalued contract.”
Anyway, here’s hoping that the hand bothered him more than we know, and he bounces back next year so we can get some decent value out of him.
I think I would mind it less
If it didn’t include heavy no-trade protection.
"I´d like to apologize in advance for anything that I may say or do that could be construed as offensive as I slowly go NUTS."
Actually. . .
No, MY was not below replacement. He was worth 1.7 wins ABOVE replacement.
He was below replacement as a hitter and fielder, but with the positional adjustment, he’s above.
Okay
So you’re saying value wins are equivalent to WAR? I thought the article said that a replacement level player started at 20 value runs per 600 PAs, which converted to two value wins.
Hm.
I see now in re-reading it that they differentiate Replacement Level as 20 runs per 600 PAs, not value runs. This, and the overall obtuseness of their formula (particularly their Replacement column), irritate me.
Maybe I missed something, but
I don’t understand how they calculate the full value of their Replacement column.
Most of this vitriol is just a rant of general irritation at sports bloggers for not explaining their technical terms or having a readily available resource on their sites to explain their technical terms. Fangraphs, in particular, does this quite a lot.
That's why I spend a lot of time emailing guys like Tom Tango and places like Baseball Prospectus.
I don’t expect people to rehash the complexity of their calculations every single time they cite them.
This an incredibly belated response, but nevertheless...
This is why you create an extensive and verbose glossary of your technical terms and then link to the terms in the glossary in each of your articles! Anyone who’s spent a day in a technical writing class would know this.
Considering the size and complexity of the things done at baseball stat sites, it would be mere child’s play for them to write something that would automatically create links in their articles redirecting users to that term’s particular glossary entry.
There is never a reasonable excuse for obtuseness in technical explanations. It is either mere laziness on the part of the source or a desire to appear more intelligent.
Laziness
is a good explanation for your apparent inability to go to the glossary and look something up. If you want to continue reading things of this nature then it seems to me you should be willing to learn the lingo and get more familiar with some of them. Is it really that hard?
G G G E-flat_______ F F F D__________....
That's my point.
Their glossary is woefully incomplete. I wanted to know how FanGraphs came up with their particular statistics, asked questions, didn’t get answers, and couldn’t find the information in their glossary.
How exactly would you propose I figure this out, then?
Fangraphs is probably the worst with this
They have a glossary which has a small handful of their stats. The worst is that they have explained it all before and it wouldn’t be too hard for them to just have a link to those posts on their glossary page.
By 2028, Mark Teixeira will be in the HOF.
-The Outlaw

by 





























