« No Bump For Kerry/Edwards | Main | Alabama Needs Dildos »

Lies, damned lies, statistics, and polls

Kevin posted about the most recent national poll for the upcoming election. Once again, I’m reminded just how bloody useless these things are.

Come November, we will NOT be having a national election, and we will NOT be electing a president. The 50 states (and the District of Columbia) will be having 51 separate elections to choose delegates to elect the president.

Let’s reduce this to the level of absurdity. November rolls around and Kerry gets every single vote in 38 states and DC, while Bush, Nader, etc. get none whatsoever. In the other 12 states, though, only a single voter shows up in the entire state and votes for Bush. If it is the right 12 states, though (a quick analysis shows them to be California, Florida, Georgia, Illinois, Massachusetts, Michigan, New Jersey, New York, Ohio, Pennsylvania, Texas, and Virginia), Bush will win 270 electoral votes to Kerry’s 268. Presuming the exact same number of people turn out to vote as did in 2000, George Bush would win re-election with 12 votes, and John Kerry would lose while receiving 45,424,585 votes.

Yes, it’s absurd. Yes, it’s completely improbable. But that’s the way the system works, and the sooner we stop deluding ourselves with the relevance of “national polls,” the better off we’ll all be.

J.

(Note: My thanks to these two websites for helping me crunch the above numbers.)


TrackBack

Listed below are links to weblogs that reference Lies, damned lies, statistics, and polls:

» Say Anything linked with Bush Jumps In Polls After DNC Convention

» This isn’t writing, it’s typing. linked with Pray for Dean Barnett

Comments (8)

Excellent points JT.... (Below threshold)

Excellent points JT.

Killjoy. ;-)... (Below threshold)

Killjoy. ;-)

Thanks for the example, Jay... (Below threshold)
Tim:

Thanks for the example, Jay. The electoral system has been a huge conceptual hurdle for me for years. It's not hard to understand how it works. That's not my problem. It's just that, as Jay pointed out, it seems so... inadequate. I wasn't sure how to mathematically illustrate it until that example.

And while the scenario that he pointed out is improbable, it effectively points out a glaring flaw in the process.

Sadly, I don't see the system ever changing. At least, not anytime soon. Whenever someone has the inspiration to change it, someone else (likely from the losing party of the past election) will protest that his/her party should have won the past election.

Tim: I don't think Jay's po... (Below threshold)

Tim: I don't think Jay's point was that the electoral college system is flawed. Instead, he seems to be saying that states elect the president (as the founders intended, for reasons still relevant), and so the polls we need to pay attention to are not the national ones, but those in individual states. From this electoral college polling we can get a more accurate and relevant understanding of the campaign.

Anyone looking to learn mor... (Below threshold)
Jim:

Anyone looking to learn more about the electoral college should consult two authors:

Pro EC: Judith Best
Against EC: Lawrence Longley

You're right, Professor Cha... (Below threshold)
Tim:

You're right, Professor Chaos. I just read Jay's post again and that's exactly what he was saying. (I was listening to Hoobastank at the time and misread his main point.)

Thanks for clarifying that (he says, a little embarrassed).

Hear, hear, professor.... (Below threshold)

Hear, hear, professor.

Even when the colonies and states numbered thirteen, the founders recognized that regional differences among the states would lead to tension. So they acknowledged through the creation of the Senate and the Electoral College that the people of various states should have an equal voice, or at least a proportionate voice, to the populous tribes of the Northeast.

So it goes today: certain populous states in the northeast (and the west) tend to think in common, and the residents in the majority of the country's land mass think differently, and the weight of the bicameral legislature and electoral college prevent the oligraches from running the country in their best interest.

So guess who wants to eliminate those checks on their power?

Nah.I disagree ent... (Below threshold)
Mike:

Nah.

I disagree entirely.

I DO like the the electoral system (love it after 2000). Anything that gives more rights to states and less to the Feds is fine by me. Congress works the same way with the senators from Montana having an equal vote as those from California.

BUT, I disagree with the comments about national polls being meaningless. Unless it's a REALLY close race, the electoral winner is gonna be the same as the national winner. Sure it didn't work that way last time, but how many times is the vote gonna be split by less than 1%? Does anyone know the odds of it happening TWICE IN A ROW?

The state polls follow the national polls. When Kerry was up a few points in the national polls earlier this month, all the battleground polls were leaning toward him. When Bush tightened up the race nationally, the battlegrounds started shifting back to him.

The battleground polls will only matter if it's a very close election (less than 1% spread). Sure you can theorize about more widely erratic behavior but it just ain't gonna happen. (In 2000, for example, if the national average had moved less than 5 hundredths of one percent toward Gore that would have pulled enough Florida voters to give him the win. If it had moved half a percent, for a total spread of 1%, then Gore would have won with over 300 electoral votes.)

And here's the problem, if the race really is that close, the state polls will be just as meaningless as the national ones. They're all taken at widely different times and by completely different methods. And they are sometimes not released for two weeks after they were taken. It's simply too imprecise to try to compile them and get any kind of feel for who's really ahead in a tight race. And if it ain't tight, then who cares who wins New Hampshire or West Virginia?

No, national polls (the honest ones) give an immediate indication of who's on top. State polls are just more fun!




Advertisements









rightads.gif

beltwaybloggers.gif

insiderslogo.jpg

mba_blue.gif

Follow Wizbang

Follow Wizbang on FacebookFollow Wizbang on TwitterSubscribe to Wizbang feedWizbang Mobile

Contact

Send e-mail tips to us:

[email protected]

Fresh Links

Credits

Section Editor: Maggie Whitton

Editors: Jay Tea, Lorie Byrd, Kim Priestap, DJ Drummond, Michael Laprarie, Baron Von Ottomatic, Shawn Mallow, Rick, Dan Karipides, Michael Avitablile, Charlie Quidnunc, Steve Schippert

Emeritus: Paul, Mary Katherine Ham, Jim Addison, Alexander K. McClure, Cassy Fiano, Bill Jempty, John Stansbury, Rob Port

In Memorium: HughS

All original content copyright © 2003-2010 by Wizbang®, LLC. All rights reserved. Wizbang® is a registered service mark.

Powered by Movable Type Pro 4.361

Hosting by ServInt

Ratings on this site are powered by the Ajax Ratings Pro plugin for Movable Type.

Search on this site is powered by the FastSearch plugin for Movable Type.

Blogrolls on this site are powered by the MT-Blogroll.

Temporary site design is based on Cutline and Cutline for MT. Graphics by Apothegm Designs.

Author Login



Terms Of Service

DCMA Compliance Notice

Privacy Policy