The Interactive Hall of Fame Ballot
You didn't really expect me to sit back and watch Chris steal the show by introducing interactivity to our Visuals Above Replacement series, did you?
Ever since I first debuted the "bullseye" Wins Above Replacement/Wins Above Excellence/Wins Above MVP chart to honor Ron Santo, I though of how I could present the Hall of Fame ballot using a similar approach. Well, here it is.
So, click, drag, compare, contrast, play… let me know what you think! (All rWAR data lovingly extracted from Baseball-Reference.com.)
33 comments
|
5 recs |
Do you like this story?
Comments
I'm very proud right now.
See Data Differently: Beyond the Box Score | @justinbopp
Create. Coach. Conquer! Two Out Rally, Baseball MMORPG | Facebook
by Justin Bopp on Dec 16, 2010 10:39 AM EST up reply actions
Absolutely. Now I have to figure out
how I can make a chart where I can arrange the subjects into a giant smiley face.
See Data Differently: Beyond the Box Score | @justinbopp
Create. Coach. Conquer! Two Out Rally, Baseball MMORPG | Facebook
by Justin Bopp on Dec 16, 2010 10:41 AM EST up reply actions
Love the Brett Boone - Harold Baines comparison. compiler vs. peakvalue
This is so good, I’m not even asking for tweaks! Yet. ; )
Agreed. This view is a great way to see
guys with short, flashy careers vs. those that were just good for a long time. That Santiago one is fascinating.
See Data Differently: Beyond the Box Score | @justinbopp
Create. Coach. Conquer! Two Out Rally, Baseball MMORPG | Facebook
by Justin Bopp on Dec 16, 2010 10:43 AM EST up reply actions
I never studied statistics so forgive me if this is a stupid question
but is there a reason the HoF median is listed instead of the mean? Is there a major difference between those numbers?
The studs (Ruth, Mays) really move up the average compared to just the median.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Dec 16, 2010 10:37 AM EST up reply actions
Right.
It’s a way of accounting for outliers.
See Data Differently: Beyond the Box Score | @justinbopp
Create. Coach. Conquer! Two Out Rally, Baseball MMORPG | Facebook
by Justin Bopp on Dec 16, 2010 10:43 AM EST up reply actions
Hmmmmmmm
Given that this is the Hall of, ya know, Fame and all, wouldn’t it be right to include the studs in the mean than to exclude outliers via a median? If it was the Hall of Average Excellent Players then I’d say ok, but since there’s likely a compressed baseline on the low end (and thus fewer outliers in that direction) maybe average would be a better means of calculation and comparison?
If players are distributed on a curve from best to worst and you lop the curve off at some point near the positive extreme (at whatever is designated as the HOF cutoff point), then you’d have far more players near the lower edge of that scale due to the shape of distribution curves and of those players a certain portion would be the result of a mistake of judgment or reputation than the players near the high end (ignoring for the moment any and all arguments about roids, Pete Rose, players inducted for reasons other than play on the field, and the like). There’s no question about whether a guy like Willie Mays should get in, so the top end of the scale is without error and the undeserving outliers would only exist on the lower end. By excluding outliers on the upper end by looking to median, it looks to me like the math might be progressively lowering expectations over time, as every so often a Jim Rice is going to get in and be progressively applied to the median.
Just my $0.02. Is there anything I’m missing in my approach to the math here?
On a related note, I’m still lobbying for Mario Mendoza to make the HOF due to his contribution to arbitrary but amusing descriptions of mediocrity (and resulting fame), but despite my efforts I don’t think he or his cohorts will ever be in.
It's a triumph of number crunching over the human spirit...aaaaaand, it’s about time. -- Play-by-Play Announcer, The Simpsons.
by MookieTheCat on Dec 16, 2010 4:31 PM EST up reply actions
Suprise for me was Olerud
Case comes out a lot better than I would have thought.
Clear eyes, full hearts, can't lose.
One thing I'd like to see are some
standard deviation bullseyes bracketing our candidates, in something of a HoF-Zone for this method. I’d love to know what the boundaries look like.
See Data Differently: Beyond the Box Score | @justinbopp
Create. Coach. Conquer! Two Out Rally, Baseball MMORPG | Facebook
Sweet work.
- .-. ..- … – / – …. . / .—. .-. - .. . … …
by Jeff Zimmerman on Dec 16, 2010 10:49 AM EST reply actions
I love the Lenny Harris reference. It's inexplicable as to how he ended up on the ballot.
Also, holy crap, John Olerud’s career was way better than I thought it was.
Mute the Sounders. Malamute the Huskies.
I laughed when I saw the note on Harris
If he gets a vote, I doubt he’s the worst player ever to get one though.
Bad Left Hook - The SB Nation boxing blog
"Baseball is played on the field, not on a calculator."
Why do I have the feeling Olerud won't even get 5%?
And maybe not Walker? Yet I fear that Tino does.
On Twitter: @baseballtwit
Actually, stupid me
Olerud got a pair in Toronto. Maybe Tino’s are more magical because he got them in New York.
Exchange rate's a bitch.
On Twitter: @baseballtwit
by adarowski on Dec 16, 2010 8:48 PM EST up reply actions 2 recs
Constructive Criticism
The human eye is really really bad at judging areas of different circles. We’re just not good at it. While the ability to move things around makes it a bit better, the simple fact remains that it is tough to compare sizes without cheating and looking at the numbers.
For a really really simple example. Take a look at the chart. Which is bigger? The inner circle for Brett Boone, or Larry Walker? Move them next to each other if you’d like. At first (and second, and third…) glance, Brett Boone has a bigger middle circle because we compare the size of the inner circle with the others, which makes us feel like it’s bigger. But they’re both the exact same size.
That’s the problem with any sort of circular chart — we are pretty bad at judging area, especially comparing two different areas in different contexts. If you set the median circle as an outline for each, then you’d have a better idea (equal reference for comparison for each graph), but then you run into a problem where not each ring shares the same baseline (they just pick up where the last one leaves off).
Here is an article about this phenomenon, and at least one solution:
http://www.perceptualedge.com/example18.php
If you feel really strongly about keeping the design (it does look nice), then can I suggest that you create two different color schemes for each wheel? You can use blue for each “ring” that is above the HOF median, and use a grey shade for each “ring” that is smaller than the median. That way, in a glance, you can see which areas each player excels in. If two of the rings are blue, you know they deserve a closer look for admittance into the Hall.
My Work: Henkakyuu
Thanks for the feedback
I think that’s why the dragability helps. On some graphics with just a couple “bullseyes”, I had some light gray guides show where things line up a bit better. It’s tricky with guys who are really close together, but zooming out to the whole ballot I think it does a good job of showing who belongs.
As usual, no one visual representation is perfect. I really like how these capture WAR, WAE, and WAM. Comparisons between two charts aren’t always perfectly clear, but in many cases I think that’s because the value of the players is so damn similar.
On Twitter: @baseballtwit
Any chance of making a version with the color switch I suggested?
I do like the charts, I just want to get more information at a glance than I currently am. I especially like the interactivity (how did you do it?). Any chance you could tweak the colors to make the “above median” and “below median” circles more clear?
My Work: Henkakyuu
It's all CSS, so certainly can try it.
I hope to do the pitchers soon, so I’ll try then.
On Twitter: @baseballtwit
No html 5
Some javascript and a lot of nice CSS
by Dan Turkenkopf on Dec 18, 2010 8:22 PM EST via mobile up reply actions
Well, it's all in HTML5, too.
But the real magic is the jQuery and CSS3.
Plus, all the CSS is built using Sass, which means I can make my CSS perform basic math functions. If you’ve used CSS but never used Sass, oh damn you gotta check it out. http://sass-lang.com/
On Twitter: @baseballtwit
Wow. This is unreal.
Can’t wait for Chris or Justin’s next graph.
Question of the day: Bret Boone or Bobby Higginson?
My Michigan State (and Big Ten) Baseball Blog.
Like music? See what I'm listening to at my Last.fm account.
One-upped
Props to Adam for taking interactivity to the next level. I love working with people who continually set the bar higher. Really, really great work.
by Chris Spurlock on Dec 16, 2010 11:16 PM EST reply actions

by 



























