Page 1 of 2

1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 4:39 pm UTC
by Kludgy
Image
Title text: I like the idea of cartograms (distorted population maps), but I feel like in practice they often end up being the worst of both worlds—not great for showing geography OR counting people. And on top of that, they have all the problems of a chloro... chorophl... chloropet... map with areas colored in.

Maps are where the the distortion is

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 4:45 pm UTC
by Velo Steve
This looks more evenly distributed than I would have guessed. Even though we seem to be divided to the extreme ends of the left/right spectrum these days, as least we're not completely isolated physically.

I agree about cartograms and chloropleths (I didn't look that up so my spelling could be like Randall's). Nice ideas, but sometimes hard to grasp.

A little odd seeing this 16 months after the fact.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 4:49 pm UTC
by cellocgw
Kludgy wrote:
Maps are where the the distortion is


Maps are where the Enemy's Gate is.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 5:03 pm UTC
by petercooperjr
Velo Steve wrote:This looks more evenly distributed than I would have guessed. Even though we seem to be divided to the extreme ends of the left/right spectrum these days, as least we're not completely isolated physically.


I think that this really does give a better picture of the distribution. There are certainly plenty of "red" areas on the populous coastal states, and plenty of "blue" areas within the sparse western states, it's just that nobody cares about them during a presidential election.

It's important to keep in mind that local politics (elections to whatever town/city/etc. government your locality has, state rep. elections, and here in New England the exciting tradition of Town Meeting) often have a much higher impact on one's life than the national ones that get all the media attention. There's not nearly enough interest/funding for good journalism of really local issues, and so one tends to see coverage of things that affect much larger populations since the larger base can sell enough advertising/"newspapers"/etc. to make that journalism possible. But it's the local issues that you can (1) most directly impact yourself, and (2) often most directly impact you.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 5:08 pm UTC
by Stargazer71
Once again, a map that only shows half the picture.

An interesting aspect of US elections is that the makeup of the electoral college is determined by total population, while the popular vote is based on voter turnout.

This is meaningful because voters from non-swing states have less incentive to vote than those in highly competitive states. With all else being equal, *both* candidates will receive more votes from a swing state than they will from a non-swing state of equal population.

All the more reason why saying, "Here's a graph of popular vote distributed by <whatever>" ... doesn't mean squat.

(EDIT: I said that it only shows half the picture, but upon second reading, perhaps this was the purpose of the comic)

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 5:15 pm UTC
by owenneil
I spent a while looking, but I give up. Where's Waldo?

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 5:21 pm UTC
by Soupspoon
Is it possibly intentional (and covered by approximate location within the state, either way, as a get-out) that Michigan UP has no voters marked down?

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 5:27 pm UTC
by MechaMaya
Stargazer71 wrote:Once again, a map that only shows half the picture

...

(EDIT: I said that it only shows half the picture, but upon second reading, perhaps this was the purpose of the comic)


Yeah, I think that is the point. Normally we see electoral college maps but that is only the other half of the picture. He's emphasizing that this helps see the remaining half. Neither map is complete without the other. Sure the electoral college maps show which states 'won', which is important. But when you're trying to get an idea of how many voters of each side are in each region, the electoral college maps really don't help much. If only there was a way we could combine the two, and get the best of both worlds. I will hope Randall will have that figured out by the 2020 election.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 5:56 pm UTC
by sardia
Velo Steve wrote:This looks more evenly distributed than I would have guessed. Even though we seem to be divided to the extreme ends of the left/right spectrum these days, as least we're not completely isolated physically.

I agree about cartograms and chloropleths (I didn't look that up so my spelling could be like Randall's). Nice ideas, but sometimes hard to grasp.

A little odd seeing this 16 months after the fact.

This map isn't good for determining the closeness of a race. Like if you won 60 to 40%, the map would look really similar. Instead it tells us that a lot fewer people live in a lot of states.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 6:18 pm UTC
by Pfhorrest
I think that's part of the point. Colored maps make it look like America is mostly filled with Republicans with a few small clusters of Democrats around the coasts. Cartograms try to compensate for that but aren't great at it. This map shows you that there are just lots more people near the coasts, slightly more of them being Democrats; and hardly anyone in the middle, slightly more of them Republican. It's pretty close proportions everywhere, there are just lots more people near the coasts, and a slight trend of more Democrats there too.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 6:20 pm UTC
by DanD
This does help answer the "but more states voted for" argument in a fairly visible way. At least to me, it manages it on a more visceral level than the ones that have the county sizes weighted by population or the like.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 7:31 pm UTC
by Velo Steve
Stargazer71 wrote:... This is meaningful because voters from non-swing states have less incentive to vote than those in highly competitive states. With all else being equal, *both* candidates will receive more votes from a swing state than they will from a non-swing state of equal population.
...


And non-swing state voters may show up, but vote differently. In California, I voted for a hopeless candidate in a futile attempt to send a message to both parties that they could have done better. In a state where my vote might actually make a difference, I would have chosen between the two who had a chance.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 7:52 pm UTC
by Stargazer71
Velo Steve wrote:... In California, I voted for a hopeless candidate in a futile attempt to send a message to both parties that they could have done better. In a state where my vote might actually make a difference, I would have chosen between the two who had a chance.


I've heard that Mickey Mouse and Donald Duck get a substantial number of votes in every presidential election cycle. That *definitely* sends a message, but I'm not exactly sure what that message is.

As a voter in a kindof-sortof-sometimes-swing-state (Colorado), I have never had the motivation to try sending the animated mouse to the Oval Office, though I readily admit that the prospect of the White House becoming the new Mickey Mouse Clubhouse is ... appealing :D .

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 8:00 pm UTC
by da Doctah
You have produced a map showing the approximate distribution of where people who voted live in the US. Got anything more interesting to us colorblind folks?

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 8:10 pm UTC
by Keyman
cellocgw wrote:
Kludgy wrote:
Maps are where the the distortion is


Maps are where the Enemy's Gate is.

The Enemy's Gate is DOWN!

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 8:32 pm UTC
by Pfhorrest
And maps are also down, since they are obviously depicted from an aerial view.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 8:36 pm UTC
by Soupspoon
Stargazer71 wrote:I've heard that Mickey Mouse and Donald Duck get a substantial number of votes in every presidential election cycle. That *definitely* sends a message, but I'm not exactly sure what that message is.
That some people think that their vote… {exaggerates Scots accent} …Disney matter?

:) :arrow: 8-)

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 8:58 pm UTC
by rmsgrey
Velo Steve wrote:A little odd seeing this 16 months after the fact.


How long does it take for the raw data to hit the public domain?

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 9:51 pm UTC
by DLC43
Hey, there are a bunch of economists and other data visualization fans who love this map and are wondering whether it was created using software or by hand? If software we'd love to know what software!

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 10:17 pm UTC
by bantler
Velo Steve wrote:And non-swing state voters may show up, but vote differently. In California, I voted for a hopeless candidate in a futile attempt to send a message to both parties that they could have done better. In a state where my vote might actually make a difference, I would have chosen between the two who had a chance.


Dude, it's either Beef or Chicken on this flight. Ordering a Salad just annoys the stewardesses.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 10:48 pm UTC
by leeharveyosmond
Male red/green colourblindness.
It’s a thing.
Traffic signals are not a problem for me.
I find this map unnecessarily difficult to interpret.

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 11:00 pm UTC
by bheliker
[citation needed] !!
Anyone know where he gets his source data?

Best source I can find so far: https://data.opendatasoft.com/explore/d ... ic/export/

Re: 1939: "2016 Election Map"

Posted: Mon Jan 08, 2018 11:02 pm UTC
by Steve_The_Beard
Very interesting, but I'm disappointed.

So will be one man in twelve, and one woman in two hundred.

That's how many people share my imperfect colour vision. When I zoomed the image as far as possible I could finally see that the figures were actually in three different colours, but then I couldn't see the whole picture, or even a decent slice of it. Had you used three well-chosen different symbols, you'd have reached a larger audience.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 12:00 am UTC
by hailspork
Also colorblind (8% of men are.)

Any hints at what the key takeaways are?

Also, may I suggest "Lucky Charms" rules? Colors AND shapes. Or numbers. Or any non-color-coded display of information.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 1:18 am UTC
by ucim
bantler wrote:Dude, it's either Beef or Chicken on this flight. Ordering a Salad just annoys the stewardesses.
What if I have a beef with my chicken?

Jose

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 1:39 am UTC
by Steve the Pocket
I had no idea there were that many "other" votes. A whole quarter of Utah? Holy moly.

Out of curiosity I did some digging and also discovered there were a lot of dissenting electoral voters as well. This whole time I was thinking that "faithless electors" was just a hypothetical scenario fueled by wishful thinking.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 2:52 am UTC
by ivnja
Quick touch-up in Paint using the color palette here (namely, nos. 11, 6, and 14) to hopefully allow the colors to stand out slightly more:
2016_election_map.png


If that's still not working for some folks, I can try to put one together with shapes.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 4:50 am UTC
by hailspork
ivnja wrote:Quick touch-up in Paint using the color palette here (namely, nos. 11, 6, and 14) to hopefully allow the colors to stand out slightly more: 2016_election_map.png

If that's still not working for some folks, I can try to put one together with shapes.

I'm severely colorblind, but I wouldn't bother for just me. I think I got the joke from other comments.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 6:22 am UTC
by Marximus
Great idea, but it doesn't really tell the story. The people who did not vote at all should be represented, as well.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 6:29 am UTC
by djagir
Steve the Pocket wrote:I had no idea there were that many "other" votes. A whole quarter of Utah? Holy moly.

Utah had the whole Evan McMullin thing going, so there were probably more "other"s there than would otherwise have been the case.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 10:22 am UTC
by Mikeski
djagir wrote:
Steve the Pocket wrote:I had no idea there were that many "other" votes. A whole quarter of Utah? Holy moly.

Utah had the whole Evan McMullin thing going, so there were probably more "other"s there than would otherwise have been the case.

Yup, home-state advantage. The same reason Reagan only won 49 states when he was reelected; Minnesota voted for the home-town boy Mondale.

OK, we have also voted for Jesse Ventura and Al Franken, and had Sanders and Rubio* winning the primaries last time, so maybe we're just really dumb and moderately nuts. We also gave McMullin more votes (53K) than Hillary's margin of victory (45K), so our Republican voters forgot we were a swing state and threw the election. (Never mind Johnson's 113K for the libertarian legalized-weed party). We take our dumb nuttiness seriously, though; we had the highest percentage of eligible voters show up to vote of any state in the USA. (We're the Puppy Bowl of voting... we're all happy to participate, but we really don't know how to play the game.)

...

There must be some "bleed" between the states on this map (the grey text implies there is), since the number of figures doesn't make sense on a per-state basis. Minnesota should have about 1.11 green people, 5.29 red people, and 5.47 blue people. It has 1, 5, and 6. Colorado should be 0.96, 4.80, and 5.35, and it has 1, 4, and 5. Obviously, that's not rounding off by state... "5.47 rounds up" while "4.80 rounds down" can't be what's happening, unless political pollsters are doing the rounding. :mrgreen:

That's not affecting Utah, though, which is 2.06 red, 1.24 blue, and 1.22 green. (Yes, "other" did almost beat "Democrat" there.)


* - an exception to hometown boys always winning; Rubio only won MN in the primaries. He came in second in his home state of Florida

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 1:57 pm UTC
by DavidSh
He did say he is trying to keep national totals intact. It would be surprising if using the same rounding point for all parties would achieve that. Can you determine if there is a consistent rounding point within each party?

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 3:27 pm UTC
by eidako
owenneil wrote:I spent a while looking, but I give up. Where's Waldo?

Northern Florida.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 3:54 pm UTC
by Bob Stein - VisiBone
Marximus wrote:Great idea, but it doesn't really tell the story. The people who did not vote at all should be represented, as well.


Great idea. I'd love to see ghost icons for eligible voters who didn't vote. And for the even more ghostly ineligible voters: minors, felons, illegal immigrants, aliens in area-51 cryovats.

Image
Source: https://pin.it/am5eldbud67hrj
Source: Washington Post / U.S. Election Project / Dave Wasserman, Census Bureau

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 4:57 pm UTC
by Stargazer71
Bob Stein - VisiBone wrote:
Marximus wrote:Great idea, but it doesn't really tell the story. The people who did not vote at all should be represented, as well.


Great idea. I'd love to see ghost icons for eligible voters who didn't vote. And for the even more ghostly ineligible voters: minors, felons, illegal immigrants, aliens in area-51 cryovats.

Image
Source: https://pin.it/am5eldbud67hrj
Source: Washington Post / U.S. Election Project / Dave Wasserman, Census Bureau


Why, oh dear God why, would you include minors in a pie chart of election results? Why don't you just include the entire population of Sweden inside the "ineligible to vote" slice while you're at it? It would make just as much sense.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 6:04 pm UTC
by Reka
Mikeski wrote:There must be some "bleed" between the states on this map (the grey text implies there is), since the number of figures doesn't make sense on a per-state basis. Minnesota should have about 1.11 green people, 5.29 red people, and 5.47 blue people. It has 1, 5, and 6. Colorado should be 0.96, 4.80, and 5.35, and it has 1, 4, and 5.

I think he must be assigning the number of people first, and then deciding what colors they should be based on proportions. So Minnesota needs to have 12 people: 1 is obviously green, but how to distribute 11 people almost-evenly between red and blue? Answer is, you can't, so you do 5 red and 6 blue. This falls down for Colorado, which according to your numbers should have 11 stick figures, but only has 10. Maybe Randall is using slightly different numbers than you?

I still don't understand the timing of this comic.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 7:06 pm UTC
by Keyman
Stargazer71 wrote:
Bob Stein - VisiBone wrote:
Marximus wrote:Great idea, but it doesn't really tell the story. The people who did not vote at all should be represented, as well.


Great idea. I'd love to see ghost icons for eligible voters who didn't vote. And for the even more ghostly ineligible voters: minors, felons, illegal immigrants, aliens in area-51 cryovats.

Image
Source: https://pin.it/am5eldbud67hrj
Source: Washington Post / U.S. Election Project / Dave Wasserman, Census Bureau


Why, oh dear God why, would you include minors in a pie chart of election results? Why don't you just include the entire population of Sweden inside the "ineligible to vote" slice while you're at it? It would make just as much sense.

This is sarcasm, right? In Minnesota, we have a large population of Swedes. (Whether they fit in Mikeski's "Puppy Bowl" voting theory, I don't know.) But there are many ways one might 'ineligible to vote'. Assuming Convicted Felons are not 28% of the populace and/or non compos mentis ,we must also be including those not registered to vote....

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 7:28 pm UTC
by Leovan
Keyman wrote:
Stargazer71 wrote:
Bob Stein - VisiBone wrote:
Marximus wrote:Great idea, but it doesn't really tell the story. The people who did not vote at all should be represented, as well.


Great idea. I'd love to see ghost icons for eligible voters who didn't vote. And for the even more ghostly ineligible voters: minors, felons, illegal immigrants, aliens in area-51 cryovats.

Image
Source: https://pin.it/am5eldbud67hrj
Source: Washington Post / U.S. Election Project / Dave Wasserman, Census Bureau


Why, oh dear God why, would you include minors in a pie chart of election results? Why don't you just include the entire population of Sweden inside the "ineligible to vote" slice while you're at it? It would make just as much sense.

This is sarcasm, right? In Minnesota, we have a large population of Swedes. (Whether they fit in Mikeski's "Puppy Bowl" voting theory, I don't know.) But there are many ways one might 'ineligible to vote'. Assuming Convicted Felons are not 28% of the populace and/or non compos mentis ,we must also be including those not registered to vote....


If minors and immigrants are included then 28% is actually pretty easy to explain. I wouldn't be surprised if minors alone make more than 20%. Where I live, we have more than 20% immigrants as well, so it would be around 40%.
Not registered I'd place under "didn't vote".

Edit: looked it up. Minors make 24% of the US population and immigrants 13%. Implies immigrants were not included in "How America participated"

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 7:53 pm UTC
by mathmannix
Leovan wrote:Edit: looked it up. Minors make 24% of the US population and immigrants 13%. Implies immigrants were not included in "How America participated"

That 13% of the total U.S. population is ALL immigrants, most of whom become legalized citizens, and can (and many of whom do) vote.
Nobody knows for certain how many illegal immigrants there are, but it's probably close to 3.5 percent.

Re: 1939: "2016 Election Map"

Posted: Tue Jan 09, 2018 8:14 pm UTC
by Soupspoon
(Several people came to post whilst I was rushing back home regretting having not taken my second emergency USB power pack with me. I expect ninjas and other treatments of the issue, once I read the interim.)

Those not yet eligible to vote due to age (but not through being Scandiwegian rather than Leftpondian) are still affected directly by the result.

There'll be some future Swedish immigrants, certainly, but that's a choice that native-born children already have not been given and are already affected by (and child Dreamers, but let's not muddle things with that, right now), so one can legitimately exclude the former disenfranchisees from the stats but show the latter ones, especially if you're including Won't Voters and Disqualified Myself From Voting Because I Got Unpaid Parking Fines people, who (by different measures) self-disenfranchised themselves.


Oh, it's all arguable. Like the original Brexit voters dying off quicker than the Remainer ones, due to age demographics, and then trying to work out what the votes of the newly come-of-voting-age voters would have done for that vote, by the time the whole Exit approaches reality (before even looking at protest-vote retraction rate vs "we voted for it, even if I didn't, so just get it done already" adjustments, and all kinds of other new factors).

But I definitely see the point. That it's never been a fully open and level playing field at any time the past, doesn't stop one pointing out where there's blind-spots this last time as well.