Now with mouse-over goodness

[ciphergoth]

the latest version of my chart, now with mouseovers:

http://hacks.ciphergoth.org/test-scenario/margin-bargraph/

Remember, this is an imaginary scenario half-way through election night. Here's the graph for the current polling situation, as per electoral-vote.com.

Comments

[purplerabbits.livejournal.com]

Yay funky graph!

Um, percentages of what?

[palmer1984.livejournal.com]

I assumed it's the percentage lead for the republicans or democrats!

[ciphergoth.livejournal.com]

The number of percentage points by which the Democratic vote exceeds the Republican vote, or vice versa...

[purplerabbits.livejournal.com]

Doh, I have five-o-clock brane!

[elfy.livejournal.com]

i talked with friends about the election a few days ago and someone said despite the good results for obama right now, mccain still might win. the reason he said was that when asked now, quite some people might answers they vote for obama, out of fear to seem rassistic and that they later will vote for mccain anyway. he said there even was a word for that, but couldn't remember it. did you hear about that? do you think this risk is there? i really hope not :-/
also, did i show you www.voteforpresident.com?

[simont]

he said there even was a word for that, but couldn't remember it.

The Bradley effect.

[vvalkyri.livejournal.com]

dang, now i'm trying to remember where to even look for a link, but someone i was reading recently was seeing the 'bradley effect' as a narrative to explain an otherwise unexplainable 'suddenly the gop gets all these votes nobody knew where they came from.' -- essentially a mask if there are irregularities.

thing is, there are all manner of reasons someone who doesn't want to vote for obama can give for not wanting to vote for obama. he is very different from mccain.

per this commentator, the obama folk aren't really trying to convince mccain voters to switch; they're working on trying to make sure as many obama folk vote as possible.

[ciphergoth.livejournal.com]

There's good evidence that the Bradley effect no longer exists, if it ever did.

[vvalkyri.livejournal.com]

ciphergoth found the reference before I did. I'd forgotten that I'd been reading on fivethirtyeight.com

[vvalkyri.livejournal.com]

actually that's not the thing I was reading, after all. Hm.

[babysimon]

Cool. Definitely the clearest yet.

[palmer1984.livejournal.com]

Awesome graph!

[sleepycinderell.livejournal.com]

You are a handy accompaniment to Jon Stewart!

[yoyoangel.livejournal.com]

You are so clever! Very clear and comprehensible even to me.

[palmer1984.livejournal.com]

Just to clarify a couple of things - If Obama gets past the finishing line he wins, right?

Also, what do the numbers (e.g, California - 92 and 37) mean? Is that polling data? The EVs are electoral votes, right? And I understand the percentages.

I understand everything else!

[lizw.livejournal.com]

what do the numbers (e.g, California - 92 and 37) mean?

I found that confusing at first, too, but I think it's a running total of electoral votes for Obama.

[palmer1984.livejournal.com]

That is indeed what it seems to be :)

[ciphergoth.livejournal.com]

It's distance to the finish line in electoral votes.

Note that the finish line is a tie - 269 votes each. To win, a candidate must get 270 votes, ie *past* the finish line.

So in this example, Oregon is projected to go for Obama by a margin of 14.0%, with its 7 electoral votes. Obama is 69 EVs from a draw with all the states above Oregon, and 63 from a draw with all those states plus Oregon. And it's blue, so it went for Kerry in 2004.

[palmer1984.livejournal.com]

Ok, thanks! :)

[lizw.livejournal.com]

I may be a lone dissenter here, but I liked the previous version much better. I preferred being able to see the whole graphic without having to scroll, and the white background makes my eyes hurt very quickly as I scroll up and down. I'm not that bothered about the winning margin in each state, so I'd rather forego the extra information in favour of overall usability.

Assuming you're keeping the mouseovers, though, I'd prefer it if they disappeared when I position my mouse pointer outside the frame of the graphic, which isn't happening at the moment. I find that distracting when trying to get an overview of what's going on. I also find the blue for called Obama states too pale to be easily distinguishable from the white.

[purplerabbits.livejournal.com]

The only thing that still needs tweaked is that the margin in DC, for instance, is 'off the chart' so I can't read it at all...

[alextiefling.livejournal.com]

That all looks good to me.

[mooism.livejournal.com]

I don't like this as much as the previous versions. It's not as immediately obvious from this graph that Obama is winning; I have to stop and think for a second.

I don't think I like the mouseoveriness. I'm not sure whether I don't like having anything pop up (not helped by some of the labels jumping up a pixel or two), or whether it's just that the numbers seem to me to be counting the wrong thing; I expect the numbers to be a running total of EVs, not the distance from the finish line.

I find myself questioning whether a swingometer style is the best way to go, and whether it might be better to show it as a horse race, with each candidate battling to go from left to right in their respective lanes.

[henry_the_cow]

I think it is helpful to hide some of the numbers, so the mouse-over is a good idea, although I agree that it would be nice if they disappeared when the mouse isn't over the actual graphic.

I still find it hard to distinguish the different background colours. If anything, I prefer the previous colour scheme.