DiamondView: 2009 New York Yankees
by Tommy Bennett on Dec 19, 2009 4:00 PM EST
in Visuals Above Replacement
Email
PrintI am very pleased to announce our newest series, in which we will give you a peek at the hitting performances from 2009, team-by-team, division-by-division, using Justin Bopp's excellent DiamondView. The honor of going first falls to the World Champion New York Yankees. But first, let me explain what our data expert (Jeff Zimmerman) and our graphics expert (Justin Bopp) have cooked up for you all.
There are four categories per player: defense (UZR/150), speed (EqBRR), power (ISO), and on-base (OBP). Each category is scaled as a percentile of the major league totals (for more details, see the original article linked above). Also included on each chart (in dark tones behind the image for the player) is the positional average at the player's dominant position. Justin then sprinkles his magic graphics dust (I really have no idea how he does it), and the results are like you see here.
Three caveats before we begin. First, we have excluded the DH because it's abhorrent to the spirit of baseball since there is no meaningful way to rate their defense. Second, the data used include only last year's performance. Finally, catcher defense remains problematic, so each catcher is given the same defensive mark. This is certainly not optimal, and we'll see if we can't come up with a better way to do things going forward. If you have a suggestion, please leave it in the comments.
Without further ado, here are the 2009 New York Yankees.
UZR didn't like Teixeira's 2009 season one bit, but Yankee fans tell a different tune. In all likelihood he'll bounce back with the glove next year. That'll make him a nearly all-around player, as you can see his hitting is phenomenal. I think he likes his new home park.
Cano put together a nice season with the bat for just about any position, but especially for the keystone. He's not a great baserunner and his fielding is below league average, but so long as the bat plays, Cano plays.
The fielding is the only thing that keeps ARod from retaining "complete player" on his list of titles (which now, thankfully, includes "The Man Who Destroyed the Myth of Clutch"). It makes me wonder about the accuracy on an individual basis of the fielding spectrum, at least insofar as players can move fluidly across it. Imagine a world where ARod stuck at short.
Posada versus the aging curve! Sometimes, the best way to win a race against time is to start late. Since Posada didn't become a regular until age 26 in 1998, he is able to put up an ISO that is in the 90th percentile of all players at age 37. He's death on the base-paths and may see considerable time at DH over the coming seasons, but there's still a lot of color on that chart.
The number one way to get a rise out of a Yankee fan is either to praise or disparage Melky Cabrera. I'm not sure why a player who is so close to league average causes so much disagreement, but it may be because it is sometimes hard to look at league average and appreciate it. He might have been over his head in '09, but he's earned an every day job.
LF Johnny Damon
He's flying that kite-shaped set of tools to free agency, and likely won't be Bomber in '10. His fielding, never world-beating, is approaching liability status. His bat has turned out to be quite the asset, and he still turns it up on the bases. He'll fall back on offense and bounce back on defense and will provide moderate value somewhere.
RF Nick Swisher
Not to be confused with the rap collective that frequently finds itself posted up in a parking lot, Swisher's house is built on a foundation of plate discipline and prodigious power. Tough to believe that he was traded for peanuts a year ago. Anyone heard from Jeff Marquez or Wilson Betemit lately?
SS Derek Jeter
Praising Jeter's tangible skills is the new hating on Jeter's intangibles and defense.
There are two camps in the sabermetric world: those who believe leadoff skills are unique and those who believe batting order is a Triassic relic. I happen to fall into the former group, and Patriot has demonstrated Jeter's excellence at the essential leadoff skills. His 96th percentile on-base skills are the best thing you could have getting five plate appearances a game, and his base running acts like a force multiplier.
Next up: Boston Red Sox
editor: Updated all images to include transparent position averages. - jbopp
46 comments
|
1 recs |
Do you like this story?
Comments
Awesome.
http://www.capitolavenueclub.com/
by PWHjort on Dec 19, 2009 4:02 PM EST reply actions 1 recs
I really like the transparency on Jeter's
it should be included with all of them.
We're working on it!
Thanks for the comment
by Tommy Bennett on Dec 19, 2009 4:38 PM EST up reply actions
Replying to myself here to note that Boston’s DVC all have the transparent position average on top of the player-specific diamond. I think it turned out well and can’t wait to see your reactions.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:11 PM EST up reply actions
looks neat, but since obviously the real value of the 4 factors are not the same, the area impression is misleading. the area thing being the most important feature of this graph
This is an obvious and not-inaccurate criticism of this style of presentation. I don’t disagree, if one looks at it from your assumption that area is the most important part (which I would argue misses the point of creating a position-based expected “shape.”).
The best criticism of this visual is that it unfairly equates the ability to get on base with base-running ability. And the counter-response would be the same.
The key here, I think, is that we have the ablity to compare a player’s core tools across the literal spectrum of other players. We can overlook some flaws in the visual to understand the intention.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:17 PM EST up reply actions
The problem is that intuitively, one DOES tend to look at the area covered for a player as the key measure of value. That is what the human mind does with this information, whether that was the intent or not. A second problem, aside from the fact that the four components are not of equal value, is that the arrangements of the components dramatically affects the area of a player. Ellsbury, who currently has a very narrow shape and very little area covered, would look very different if baserunning were adjacent to on-base instead of opposite it.
Fantastic points, GreggBor.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 23, 2009 10:42 AM EST up reply actions
Catcher D
devil_fingers over at Driveline Mechanics did some work on Catcher Defense for 2009 back in October. It’s probably not a bad place to start, and is better than nothing. Reading through the comments, it sounds as if fangraphs is rolling out something this winter.
Now this will be exciting. The true usefulness will be if it can be translated to a UZR score that we can keep in the same DB, rather than pulling from several different resources for the “same” stat.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:19 PM EST up reply actions
First, we have excluded the DH becauseit’s abhorrent to the spirit of baseballsince there is no meaningful way to rate their defense.
I think you had it right the first time.
As for DiamondView, I love it! Loved it the first time it popped up here, and it’s still a very cool little tool. Excellent work :) .
"Admirably obsessive." - Uni Watch, March 24th, 2009
NA34 | HK | RMN
"Even our depth has depth." - Silverblood
Perhaps the best way to treat DHs would be to project their limited time in the field (LF or 1B) and just go with that. One of the things we ran into early on is that it is best to focus on a player’s best or most-played position and to throw out the idea of combinining and weighting his various UZR scores at all of them. (which is where we saw Adam Dunn at -27 or Zobrist at 112 or something).
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:22 PM EST up reply actions
He's a 2nd baseman. I'll give you two guesses.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
That's easy.
It’s Yuniesky Betancourt if he switched positions. Duh. Next question!
My old blog is Tigers By The Numbers.
Now I write at Bless You Boys.
Like music? See what I'm listening to at my Last.fm account.
I can always count on you guys to remind me of the burning hell we live in here in KC.
Thank you.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
Makes me feel better about the Tigers possibly turning down Scherzer for Edwin Jackson because it wasn’t a good enough deal for EJax!
My old blog is Tigers By The Numbers.
Now I write at Bless You Boys.
Like music? See what I'm listening to at my Last.fm account.
Is there plans to do this for the 2010 season using projections for each player? Also, you guys going to be running projected standings once the rosters become more finalized since there are Fan Projections at FG now that give you guys the playing time projections?
My old blog is Tigers By The Numbers.
Now I write at Bless You Boys.
Like music? See what I'm listening to at my Last.fm account.
Can't speak for others obviously
but I had given some thought to doing a playoff odds kind of projected standings using the same simulation code that I used for my post the other day. Basically do 100,000 runs per team and come up with percents based off of those.
by stevesommer05 on Dec 20, 2009 7:47 PM EST up reply actions
Great, you can take over BtB playoff odds next year!
I really wasn’t look forward to doing all of that work.
by vivaelpujols on Dec 21, 2009 5:29 PM EST up reply actions
Sweet!
My old blog is Tigers By The Numbers.
Now I write at Bless You Boys.
Like music? See what I'm listening to at my Last.fm account.
I absolutely would like to do a projected version of this. The best way I can think of would be to set up a DB with several years of data and then literally combine the visual output into a single chart. I think the eye could/would visually calculate the next year on its own.
I’d need some statistical support to come up with an acceptable projection for a single future-year, though.
One thing to remember is that all of these stats, being a percentile, are relative to every other player’s stats, rather than being able to stand on their own like a static score.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:26 PM EST up reply actions
I think
We’d only have to project EQBRR I think. Couldn’t we use one of the projection systems (i.e. CHONE) to get the offensive numbers and then either Jeff’s or my UZR projections?
by stevesommer05 on Dec 21, 2009 12:30 PM EST up reply actions
I was thinking in line with Steve. Couldn’t CHONE get used?
My old blog is Tigers By The Numbers.
Now I write at Bless You Boys.
Like music? See what I'm listening to at my Last.fm account.
Jeff / Justin / Tommy
Any thought on having a “summary Diamondview” that is based on team level percentiles?
This is a good idea as well, and I have thought of it. The question for me is what the best visual story we want to tell is?
I’m thinking about combining all important players onto one chart with every player set to the same transparency (5-15%?). That would create a really sweet visual effect where every shared area would get brighter and brighter based on the number of stacked players in that area.
Then an empty outline of that team’s average “score” across all categories? I’d have to ask my stat friends if that would be a valid thing to do (considering positional adjustments and whatnot) or if it would be better to stick with the idea above.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:30 PM EST up reply actions
I didn't mean
to get rid of the individuals, just have a summary at the beginning / end that has team level percentiles for the categories. You could just use the team value pages on Fangraphs I’d think.
by stevesommer05 on Dec 21, 2009 12:36 PM EST up reply actions
Awesome, I mean, Really Awesome
Arod came back in a hurry to save the season. Let’s see if his defense rebounds, or if winds up like Mike Lowell.
I’d love to see these (and future versions) with a 5-4-3 regression built in- that might help balance some of the flukiness of UZR.
"Have faith in the Yankees, my son. Think of the great DiMaggio."
Tommy / Jeff:
Can you reply to this or help me come up with a way to show this? I see the issue.
I read a great comment over at Pinstripe Alley that mentioned the problem of this being a single-year description, but as always, my strength is in visualization, not data projection. Perhaps we can come up with a really cool projected version of this (as noted in a comment above)?
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
by Justin Bopp on Dec 21, 2009 12:32 PM EST up reply actions
Position adjustments
I still think that the fielding number should be modified by position, i.e. UZR/150 + Position adjustment. That way, fielding is actually comparable across all players like the other axes are. A below-average fielding SS is probably better than most 1B’s.
Also, I second using the driveline mechanic catcher numbers as a better than nothing option. I think they’re actually fairly decent estimates.
-j
I write at:
Beyond the Boxscore | Red Reporter | Basement-Dwellers.com | Twitter: @jinazreds
Agreed on both counts, especially the first.
Solves the problem of DH fielding, too.
Beyond the Boxscore Not a member? Sign up.
Updated all images to include transparency.
In retrospect, it’s shocking I didn’t do it on all of them to begin with. I expect there to be some similar update after we do this a few more times.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
Your Photoshop-fu never ceases to amaze me.
by Tommy Bennett on Dec 21, 2009 8:32 PM EST up reply actions
aww thanks man.
If you or anybody knows anybody that needs my services, I’m cheap and easy!
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
Wow, look at that base running.
Very telling.
by Tommy Bennett on Dec 22, 2009 1:57 PM EST up reply actions
I actually meant to post this on the next article.
You know, for the Red Sox.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
Gardner
Any chance one of these could be added for Brett Gardner last season? I mean you can expect his defense and baserunning to be absolutely top notch, tolerable in on base and putrid in power but I’d be curious to see how it all looks put together… Thanks!
Team Composite
There was a team composite for the Red Sox. Will there be one for the Yankee’s?
I probably should make one, huh?
Lemme see if I can get one out soon. Things starting to ramp up with the holidays, so give me just a little bit of time.
See Data Differently.
beyondtheboxscore.com | Twitter: @ justinbopp
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recent FanPosts
Follow us on Twitter!
BTB: @BtBScore
Justin: @justinbopp
Satchel: @SatchelPrice
J-Doug: @RationalPastime
Daniel: @CamdenCrazies
Peter: @CapitolAvenue
Josh: @garik16
Julian: @splashinpumpkin
Matt: @devil_fingers
Bill: @BillPetti
JD: @JDSussman
Lucas: @DBITLefty
Jeff: @jeffwzimmerman
Lewis: @LewsOnFirst
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.
