• Enter your email address to follow this blog and receive notifications of new posts by email.

    Join 11 other followers

  • April 2014
    S M T W T F S
    « Mar    
  • Recent Bookmarks:

  • Archives

  • Categories

  • Pages

Personal Update

I’ll be taking a bit of a break after the election. My pWP stat did great, it was over 94% accurate in predicting the races where I was keeping track, and that’s a higher percentage than the model itself would expect. The model only failed to be predictive in races where there wasn’t regular polling (North Dakota). I’m hoping to write a long and detailed analysis of the 2012 election and codify all the lessons I learned in using pWP to create a guide for future elections. But I’m going to start a new job here very soon, so I may not get to any of that at all. There will be a few posts on here, from time to time, but for now I have some other big projects to get to. I’m still working on the final draft of a novel, which should keep me plenty busy. Thanks everyone for your interest in my pWP stat, I hope my Republican friends learn to embrace polling as another tool in winning future elections.

Obama pWP: 52%; Ohio pWP: 73%

Anything in the 40-60 range is a tossup as far as pWP goes, so I can’t predict who will win the popular vote. However, Ohio is strongly leaning in Obama’s direction. Romney has to win Ohio and one other swing state, so I would say there is a one in eight chace Romney wins. Again, this is just based on polling data. I have a gut feeling Romney is going to do better than what the polls are suggesting, but I can’t quantify that.

Here are a bunch of graphs and tables:

Current Obama pWP: 46%

Here’s Obama’s pWP for the last 11 polls taken over the previous week (the data overlaps so you don’t see all the dots), with a linear trendline:

And here’s the table of rolling averages:

I don’t feel comfortable predicting who will win the popular vote, but I would feel especially uncomfortable with the way the race is shaping up if I were Mitt Romney.

Current Obama pWP: 43%

The pWP spread continues to shrink:

And Obama’s moving averages are slowly moving up:

As a general rule, I consider any pWP between 40 and 60 to be a coin flip. And, barring a surprise, it looks like predicting who will win the popular vote in this race will require a fair coin toss, not a collection of polls.

Current Obama pWP: 42%

Obama’s numbers are consistently ticking up, bit by bit:

Also, I’ll be keeping track of all the nationwide polls as they’re posted, and I will be giving two and three-day rolling pWP averages:

I think I’ve approached the limit of my OpenOffice spreadsheet abilities, as I won’t be able to graph the rolling averages they way I’d like to. However, this should be fine, I’ll just have to publish a bar graph every day.

Who’s Ahead?

Polls are being posted everyday, and the data is coming hard and fast. As my previous two posts have suggested, it’s not entirely clear who, between Romney and Obama, is going to win the popular vote. So let’s take a look at the data, first Obama’s:

The numbers are showing quite a bit of spread, but it’s getting better. Over the last week or so, Obama looks to be polling between 45% and 48%.

Here’s Romney’s graph:

As you can see, there is a much greater range when it comes to the GOP nominee. The graph shows a hammerhead-like formation as we enter the final week of the campaign. Pollsters are having a [more] difficult time finding the electorate’s propensity to support Governor Romney. Maybe it’s because of Romney’s chameleonesque political liturgy or just because he’s not the incumbent with four years of apocalypse-free stewardship.

Still, there’s a strong level of support at the 49% mark. If Obama is at the top of his range, the race is 49-48.

In terms of pWP, Obama has 32% chance of winning the popular vote, assuming undecideds break proportionally. But, this is not normally the case. There are a few general rules when it comes to predicting how a race plays out in terms of turnout: 1) Undecideds break away from the incumbent, unless the economy is really rocking; 2) Democrats do better in POTUS election years; 3) The base of the party out of power has a stronger turnout.In this election, the economy question is open, but most people are unhappy with Obama in this category. Point 2) favors the Dems, and point 3) favors the GOP. Things look pretty good for Romney again. Of course, Ohio is where all this really matters. And Obama is still in control in Ohio.

No Signal, Just Noise

The first few polls with reactions to the third debate are out, and when combined with the other recent polls… Well… I don’t want to talk about it:

That’s a lot of noise. Obama’s pWP average over the last week is 55%. That doesn’t mean much when the data is so scattered about. Here is a graph of just Obama’s pWP numbers, without his level of support included:

Unlike in previous weeks where there was a clear central tendency in the polling numbers, the pWP numbers since the VP debate have been ambiguous and to a surprisingly strong degree. I have never seen data so stubbornly and amazingly ambiguous.

On the state level, where the election will actually be decided, the numbers are much less confusing. Obama has comfortable leads in Ohio, Iowa, Colorado and Nevada. While, just based on popular vote, this is looking like a coin flip, the actual distribution shows Obama is in a very strong position for re-election. I’ll be taking a closer look at Ohio in later posts.

If you’re looking for a big tip, might I suggest picking up the Obama to win contract at Intrade.


Get every new post delivered to your Inbox.