Democratic Underground Latest Greatest Lobby Journals Search Options Help Login
Google

Sept. 18. Mark it down. The day Kerry hit rock bottom in the polls.

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 08:05 PM
Original message
Sept. 18. Mark it down. The day Kerry hit rock bottom in the polls.
Edited on Fri Sep-24-04 08:07 PM by TruthIsAll
Would someone please tell the mediawhores that their Bushit about Kerry being down for the count is getting a little old?




Printer Friendly | Permalink |  | Top
MissouriTeacher Donating Member (476 posts) Send PM | Profile | Ignore Fri Sep-24-04 08:22 PM
Response to Original message
1. So..
how is Kerry's win probability so high with the polls (the good ones) showing the race a tie?
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 08:30 PM
Response to Reply #1
3. Simple, the popular vote is tied, but he's ahead in the Electoral Vote.
Edited on Fri Sep-24-04 08:54 PM by TruthIsAll
Zogby has him winning the EV by 297-241.

In my Election Model, the Monte Carlo election simulation consists of 5000 election trials based on the latest state polls. I run the model for five cases (25,000 trials all together) in which I assume allocation of the undecided vote to Kerry at 50,55,60,67 and 75%.

With just 60% of the undecided, Kerry has a 92% probability of winning. Here's why: he wins 4600 out of 5000 simulation trials. His average EV in the 5000 trials is 304. He needs 270 to win.

If he gets 75% of the undecided, its a blowout. His probability would be 99.6%.

I know it sounds high, but if his popular vote margin is just 2% (51-49), he has a 92% probability of getting more than 270 Ev's and winning the election.

If Kerry wins the popular vote by a four point margin (52-48), then he has a 99.6% chance of winning. With 75% of the undecided, he will achieve a 4% victory margin, based on his current numbers.

Printer Friendly | Permalink |  | Top
 
MissouriTeacher Donating Member (476 posts) Send PM | Profile | Ignore Fri Sep-24-04 08:32 PM
Response to Reply #3
4. Well...
I'm a complete ignoramus when it comes to statistical analysis, but I sure hope you're right.

Is this your first time doing this with an election?
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 09:24 PM
Response to Reply #4
11. This is the first time I have done it for others.
I maintained a spread sheet for myself in 2000.
Printer Friendly | Permalink |  | Top
 
kevin881 Donating Member (429 posts) Send PM | Profile | Ignore Fri Sep-24-04 08:34 PM
Response to Reply #3
5. if you split the undecided vote down the middle, 50% to Kerry
what are his odds then?
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 08:48 PM
Response to Reply #5
8. With 50% of the undecided, he has a 70% prob. Check the numbers
Edited on Fri Sep-24-04 09:22 PM by TruthIsAll
on the first screen.
http://www.geocities.com/electionmodel/

I run the simulation to obtain win probabilities for the five undecided allocation cases: 50, 55, 60, 67, 75%.

That's 5000 election trials for each simulation (case).
Printer Friendly | Permalink |  | Top
 
Poiuyt Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 08:37 PM
Response to Reply #3
6. Question, TIA
The win probability is based on Zogby polls from today (or recently). If Zogby's polls change, then your win probability will change also, is that correct? You aren't predicting the future, just stating how things look at this point in time?
Printer Friendly | Permalink |  | Top
 
kevin881 Donating Member (429 posts) Send PM | Profile | Ignore Fri Sep-24-04 08:46 PM
Response to Reply #6
7. yes thats right. its why the probability changes by date
also, I think that he is using an average of all polls, but not sure.
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 09:00 PM
Response to Reply #7
10. I don't use an average of the state polls. The polls are used
to calculate the probability of winning each state. The simulation runs 5000 trial elections based on these probabilties. Check out the Monte Carlo report which shows the first 10 of 5000 simulation trials, with results for each state and a total EV for each trial.

http://www.geocities.com/electionmodel/
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 08:52 PM
Response to Reply #6
9. As state polls change, so will the probabilities, of course.
Edited on Fri Sep-24-04 08:56 PM by TruthIsAll
I use the state polls from electoral-vote.com, which right now have mostly Zogby and ARG numbers.

You can view the historic trends graphically for the popular and electoral votes and the win probabilties. They all move together.

http://www.geocities.com/electionmodel/

When Kerry gets to 52% in the popular vote, he will have a win probability of close to 100% with an average EV of 335.
Printer Friendly | Permalink |  | Top
 
David Zephyr Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 08:24 PM
Response to Original message
2. TruthIsAll, Thanks For All Your Poll Posts For Months Now.
Much gratitude! :)
Printer Friendly | Permalink |  | Top
 
xray s Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 09:31 PM
Response to Original message
12. the only thing you need to know about polls is....
...the trend is our friend! :)
Printer Friendly | Permalink |  | Top
 
NewYorkerfromMass Donating Member (1000+ posts) Send PM | Profile | Ignore Fri Sep-24-04 10:03 PM
Response to Original message
13. Pennsylvania, Wisconsin, and Florida are the Dem keys
Florida is looking solid , machines allowing. the other 2 I am less sure of.
Printer Friendly | Permalink |  | Top
 
bush equals idiot Donating Member (757 posts) Send PM | Profile | Ignore Fri Sep-24-04 10:17 PM
Response to Original message
14. Follow Zogby.
Zogby was the first one to ridicule the polls recently when the TV airheads were saying Bush had an 11% lead. He was proven right two days later and everbody had to report it was too close to say anybody was ahead.

The other pollsters don't like Zogy because he's got a completely different methadology. He's been the only one who was right in all the elections.
Printer Friendly | Permalink |  | Top
 
TruthIsAll Donating Member (1000+ posts) Send PM | Profile | Ignore Sat Sep-25-04 11:45 AM
Response to Original message
15. kick to put things in perspective..
tia
Printer Friendly | Permalink |  | Top
 
DU AdBot (1000+ posts) Click to send private message to this author Click to view 
this author's profile Click to add 
this author to your buddy list Click to add 
this author to your Ignore list Fri Apr 26th 2024, 09:21 AM
Response to Original message
Advertisements [?]
 Top

Home » Discuss » Archives » General Discussion: Presidential (Through Nov 2009) Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators


Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC