If You Lived Online, You'd Be Home By Now!
by Nico on Dec 20, 2009 8:00 AM PST
Email
Print
Well anyway...
If you can't get enough of all things Michael Taylor, you might want to check out this writeup that includes a couple quotes from the ol' Taylormeister. Other than that, I don't actually have a lot to say about Michael Taylor today. He's still huge, he's still an exciting addition, his stats are the same today as they were yesterday. However, we may have a Taylor exclusive coming up soon, so stay tuned. Meanwhile, it's onto other issues on this potpourri of a Sunday...
There has been much discussion (the nicest word for "argument" that Webster has to offer) about the virtues, and lack thereof, the ease, and lack thereof, of a certain stat called WAR. I think that one of WAR's greatest attributes is also one of its greatest liabilities, and that is the fact that it uses data from a lot of different pieces of a player's game in order to produce a single number.
This is good in that unlike batting average, which tells you about hits and at bats but nothing else, WAR considers hitting, slugging, getting on base, defense, shoe size and grit (maybe not the last two), and whirls it all up in an osterizer (undoubtedly located in someone's mom's basement) in order to draw a more holistic and comprehensive conclusion about the player's value or worth.
But this is also bad because when you see that a player's WAR is high or low, you cannot instantly separate out why. In a controlled experiment, you would take a non-allergic person and subject her to identical conditions except for the presence of a cat. When she started sneezing, you'd have a pretty good idea she was probably allergic to cats. However, if you take the same person and at the same time introduce a cat, a dog, freshly cut grass, mold, carbon monoxide, and pepper, then it's a little more difficult to assess the source of her sudden non-stop sneezing fits. She will also probably break up with you, which is, believe me, a lesson I won't have to learn twice.
In other words, a player who slugged .480 slugged better than a player who slugged .440, period. It's not as helpful to look only at slugging, but in terms of understanding better and worse sluggers, it's clear. But a player whose WAR was 3.2 was better than a player whose WAR was 3.0 because...? More complete stat, less clear at a glance what's going on.
So I propose using instead a stat called PEACE. PEACE looks at Pitches seen, Errors, Average, Called strikes, and Extra base hits. It is used primarily to assign a value to trainers, traveling secretaries, and 1B coaches, and is calculated on a scale from 1.0 to fish. If you're interested, based on 2009 data Steve Sayles comes out at a 1.8, Mickey Morabito rates 3/4 of a grouper, and Todd Steverson earns an impressive 4.2na.
I'm really not sure what the world of sabermetrics would do without me.
Finally, I want to say that I really don't agree with those who believe that if the A's aren't going to contend in 2010 it doesn't matter how good they are. There is good reason to add a piece that makes the team better, even if it doesn't get the A's "over the top" into contention range, even if it doesn't help the A's long-term. As long as it doesn't harm the A's long-term, there are good reasons for Oakland to try to improve to a 78-win team, or an 80-win team, or an 82-win team, if it can.
First of all, for a young team just winning more games than the year before is a concrete achievement that sets up success the following year. Anything over 76 wins would represent a high-water mark for the past 4 seasons. Anything over 81 wins would represent a "winning season" and would make the jump to 90+ wins seem like a hop and not a leap.
Second of all, not until June will it be clear how good the other 3 teams are, how good the A's are, who has been decimated by injuries and who is having a magical season, and so on. "Probably not contending" is different from "definitely not contending" and the first trick to hedging your bets is to "hang around" in the division. You can become "sellers" or "standers pat" at the trade deadline, but you don't need to become "sellers" in April. Try to get your team, if it's possible, to the point where it can at least "hang around" until you can size up the division and the season.
So if a Carlos Delgado (as one example of a high-upside player who will not impact the team long-term) is available on the cheap, there is every reason to take a flyer on him. Troy Glaus is another example. So he might take a few at bats away from some younger players for 2-3 months; April and May are the wrong months to concede a season. You don't build for the future and give up on the present -- you build for the future and in the meantime, you do the very best you can now. It's the right message to send to your young players and it's the right way to approach a season where at the moment your doesn't look great but it also doesn't look terrible.
Now if you'll excuse me, I'm off to Asia to search for someone with Teipei blood. Meanwhile, you can discuss, on this 5th anniversary of the Mark Mulder trade, anything you darn please -- from topics broached in this post to Ryan Sweeney's upside or backside.
Brandon Inge!!!!!!
25 comments
|
0 recs |
Do you like this story?
Comments
All I know
is that AN and bacon go together like, well, AN and bacon.
And I always believe the A’s will contend. Always.
I'm here to talk about the past.
Bacon is good.
It really goes with anything.
Bacon and 67M is delicious.
Sock puppets have never been able to successfully attack castles. -NM
by Leopold Bloom on Dec 20, 2009 9:12 AM PST up reply actions
Not True: Bacon Doesn't Go Well With Bobby Crosby
…since NOTHING goes well with Bobby Crosby.
Whatever else this off-season has done, it’s still brought us some much needed addition by subtraction!
There is no "i" in Teamocil. At least not where you'd think.
by GreenNGoldSooner on Dec 20, 2009 9:21 AM PST up reply actions
and, to be fair,
67M is delicious on his own.
Sock puppets have never been able to successfully attack castles. -NM
by Leopold Bloom on Dec 20, 2009 9:22 AM PST up reply actions
as well you should.
Sock puppets have never been able to successfully attack castles. -NM
by Leopold Bloom on Dec 20, 2009 2:01 PM PST up reply actions
Now see.....
I totally agree with this. A team should always strive to “hang around” even if it is in the midst of rebuilding. Whatever that means. You never know what a season can bring. I’ll agree that it’s best not to lose focus on long term goals, but there’s a lot to be said for incremental improvement. Losing is self replicating process, as is winning. Young guys will easily believe that making the next small step is well within their grasp, especially young guys who come from a culture of winning in the minors.
"You may glory in a team triumphant, but you fall in love with a team in defeat."--The Boys of Summer
WAR and Transparent Statistics
I know that a lot of this post is about da puns, but I think the point you raise about WAR is a good one.
I’ve always been puzzled by the NFL’s focus on Quarterback Rating, a statistic that is apparently so complicated that nobody ever even bothers to explain how it’s calculated (and I don’t honestly care enough about the minutia of pro football to figure it out for myself. Here it is on Wikipedia….I don’t even care enough to read the entry and figure it out). Yet the average football fan just trusts it as SOTA and uses it to evaluate QBs.
Traditional baseball stats, on the other hand, are all about the transparency. Any serious baseball fan can calculate ERA or BA. And that’s the main (arguably the only) advantage of these stats.
BA vs. QB Rating says a lot about the cultures of the two sports, I think. Football is seen a specialized game the finer details of which only professionals really understand “under the hood.” The fact that announcers and MSM writers never break down QB Rating (or, for that matter, BCS computer algorithms) reflects not just their complexity but the fans expectations of their own ability to comprehend the complexities of the sport. On the other hand, baseball is a game the statistical understanding of which serious fans have revolutionized. There may be football equivalents of Bill James, but they don’t have James’s cultural caché, let alone his team advisory gig. (And, of course, a lot of what the specialists who run football believe about their sport is the same sort of old, traditional non-wisdom that has long dictated baseball decision making, e.g., virtually automatically punting on most fourth downs when one is out of field goal range).
There is no "i" in Teamocil. At least not where you'd think.
by GreenNGoldSooner on Dec 20, 2009 9:30 AM PST reply actions
I should clarify that my point about WAR is a serious one
The paragraph about PEACE, not so much.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
that last table on a player's fangraphs page
is incredibly useful for breaking down a player’s WAR. Yeah, it’s only broken out into offense, defense, positional, and replacement components, so you can’t tell between a good hitter with high OBP and a good hitter with high SLG, but I think it is usually enough.
For examples, let’s take a look at Oakland offense 2009 vs 2010:
Player (year)BattingFieldingReplacementPositionalWAR
Magically, they had exactly the same total WAR last year. Cust was about 1 win better with the bat, but about one win worse with the glove. Cust gave away a lot on the positional adjustment, as a DH/corner outfielder, but he played way more games than Crisp (who was injured), so he does better in the Replacement category (because every inning Crisp is injured, you need to play someone in his place — who is presumably replacement level).
So it’s still not totally transparent, but there is some pretty good information there.
oh man
I really wish I had hit preview. I actually put a lot of work into what should have been a table above.
Ok, I just tried again, with preview this time, and writing out my table in html definitely doesn’t work (but it does for fANposts). Any help?
I tab it out in notepad, and then use < pre > and < /pre > before and after
Player (year) Batting Fielding Replacement Positional WAR Crisp (2009) -0.5 4.5 7.2 0.8 1.2 Cust (2009) 8.8 -5.9 20.4 -12.9 1.2
Remove the spaces from the bracketed pre’s.
Juan Carlos Perez, please start hitting.
I know you're working for the CIA
Sock puppets have never been able to successfully attack castles. -NM
by Leopold Bloom on Dec 20, 2009 2:49 PM PST up reply actions
QB rating is a horrible, annoying stat which should crawl up in a corner and die
The proper analogy between QB rating and a baseball stat is “pitcher win/loss record.” Counterintuitive, confusing, silly, and useless.
O'Hara: Detective Lassiter is literally on fire.
Spencer: What kind of fire are we talking about-- "Michael Jackson in the Pepsi commercial" fire, or "misusing the word literally" fire?
After reading this post all I want to say is
give PEACE a chance. Now I have to get ready to watch the Raiders lose.
Let me beat the January rush.
This is all Nico’s fault.
Sock puppets have never been able to successfully attack castles. -NM
This is simply not so
when you see that a player’s WAR is high or low, you cannot instantly separate out why.
The stat is very helpfully broken down into components at Fangraphs. If you prefer a different metric for one of those components (eg you think you’re hot shit as a scout and want to replace their fielding metrics with your Mark One Eyeball version, or if you like statcorner’s hitting metric better, or whatever) then you just swap them out and recalculate.
I mean, there’s a level of hyper-detail which is abstracted, but that’s true of slugging percentage too. If I tell you a guy slugged .582 and another guy slugged .423, you’d think the first guy was a power hitter and the second guy is just an average dude, right?
Well, you’d be wrong. Guy 1 is Jacoby Ellsbury’s junior season of college. Guy 2 is Russell Branyan’s 2007.
O'Hara: Detective Lassiter is literally on fire.
Spencer: What kind of fire are we talking about-- "Michael Jackson in the Pepsi commercial" fire, or "misusing the word literally" fire?
My point is that even colin's table (of sorts!) above
only separates out “Batting,” “Fielding,” “Replacement,” “Positional.” The last two don’t have a lot of mainstream meaning to those who don’t study sabermetrics. The first two embody multiple considerations so it’s not immediately apparent what was better and what was worse in batting (OBP? SLG?) or fielding (range? errors? assists? scooping bad throws?).
There is an advantage to separating out these considerations one at a time — it’s less efficient but you also can better examine that aspect of a player’s game.
If I tell you that my health value is 12.3 while the average is 11.4, sounds like I’m a bit above average. If you find out that my rating is based on a well above average “doesn’t get sick” but also a below average “avoid getting cancer” you might see things differently — especially if your main concern at the moment revolves around my tendency to get sick or my likelihood of getting cancer.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
No one's suggesting that WAR is the only statistic that anyone should ever look at...
If you’re looking for the omni-stat, you will be disappointed forevermore.
That said, I think some degree of abstraction is actually a positive. If I’m a life insurance company, I actually don’t care what the odds are of you getting a particular kind of disease are— all I care about is when you’re likely to die. The individual components of your “health value” are little more than a distraction.
O'Hara: Detective Lassiter is literally on fire.
Spencer: What kind of fire are we talking about-- "Michael Jackson in the Pepsi commercial" fire, or "misusing the word literally" fire?
Actually, this applies to many people I know --
all I care about is when they’re likely to die.
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
I'm getting sick and tired of your Bobby Crosby bullshit.
He’s gone now. Let it go.
They call their best player "Kung Fu Panda" and they complain that people aren’t taking them or the game seriously enough? -Nick
Ha ha -- thanks. Literally the first time the cartoons have even been mentioned!
Yes, they are all my original creations, illustrated by 8th grade students, from a book called “Gullibles Travels” (briefly sold at Cody’s when it came out and kindly featured by Chip Johnson in his column when the book was released).
I like Cindi. A. She never pretends to know more than she does. B. She has unbridled enthusiasm for her "Hotties," and isn't afraid to show it. -IM4Oakgal
A's attendance tends to jump at the .500 level.
So it makes sense for them to spend money to get to .500 if they’re within spitting distance and I think they are. On the other hand their attendance hasn’t been much different if they’re a little under .500 or a lot. Once they know they’re not getting to .500, it makes sense for them to shed contracts.
It's not the results, it's how you look going about those results -- Tim McCarver
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
![](<a href="http://s516.photobucket.com/albums/u323/NicoPemantle/?action=view¤t=chessqueenjpeg.jpg" target="_blank"><img src=){kind=link}
![](<a href="http://s516.photobucket.com/albums/u323/NicoPemantle/?action=view¤t=typobloodjpeg.jpg" target="_blank"><img src=){kind=link}