The Poblano Effect?

[mshan]

I was googling for "Obama April Fundraising" and articles about this Poblano Effect showed up: http://www.burntorangereport.c...wDiary.do?diaryId=5780

What was striking to me is that, in a supposed best case scenario, Obama could win the popular vote by only 5.5% and still get 80% (350) of the electoral votes.

Can anyone with real statistical expertise, and who is willing to put personal political views aside, analyze this study and tell me if there is any validity to the statistical analysis, and also whether the hypothetical increased turnout levels are concievable under any realistic scenario.

 

[Moonbeam]

Why? I could be a second behind somebody through an intersection and be the only one killed by somebody running a red light.

Cheese, I don't give Jack about the Poblano effect.

 

[dullard]

The entire point of the electoral college is to turn miniscule population vote differences into a landslide victory. So, no it wouldn't surprize me if a 5.5% popular win = 350 electoral votes. Also, you have a simple math error. 80% of 538 electoral votes is 430 not 350.

Remember, it is theoretically possible to have only a 60 vote lead (out of millions of votes) and get 100% of the electoral college votes. And it is theroretically possible to get 0 votes in 39 states/regions and still win if only you have 1 single additional vote than your opponent in any 11 of the 12 largest states. Thus, a candidate can completely ignore all 39 smallest areas and still easilly win (completely destroying any importance of the small states).

I am not going to spend the time to analyze each and every thing said in that link. But there is one common theme in all elections that is almost universally true: "if you need a strong youth turnout to win, then you will lose". That best case scenario seems to rely heavily on strong youth turnout. The key is that the youth repeatedly say that they will vote, giving hope to candidates, but then the youth never show up in large enough numbers to make a difference. Their bark is louder than their bite.

Currently:
General election: Democrats: 260, Republicans 240, toss-up 38.
Obama vs McCain: McCain wins big.
Clinton vs McCain: Clinton wins big.

 

[dphantom]

Sure it is valid. In fact, we can take it a step further. In most states, the electoral delegates are awarded based on a simple majority. So if Obama wins, say Michigan by 1 vote, then he gets all the delegates.

It is also possible to win the popular vote but lose the electoral vote and thus the presidency. It happened once before.

See this link for some interesting details on presidential elections.

Link

 

[Vic]

Bill Clinton won the 1992 election by only 5.6% of the popular vote, of which he only got 43% overall, and still got 370 electoral votes. Text

 

[mshan]

Sounds like he's gotta make sure he wins Pennsylvania, Ohio, and Michigan, then force McCain / RNC to burn money in states like North Carolina and Virginia, where he may be more competitive than he historically should be?

 

[dullard]

Originally posted by: mshan
Sounds like he's gotta make sure he wins Pennsylvania, Ohio, and Michigan, then force McCain / RNC to burn money in states like North Carolina and Virginia, where he may be more competitive than he historically should be?

Look at the Obama vs McCain link I gave. Obama currently is at 242 electoral college votes. But of those, almost all are fairly certain wins. Only two small states (Iowa and Colorado) are possible loss states for Obama in that 242 vote tally. Thus, Obama can afford to lose those two small states. That puts Obama at 226 solid votes out of the 270 needed.

Obama needs to pick up 44 votes somewhere. Ohio + Michigan (total 37) won't do it unless Obama also makes certain that he wins those smaller barely democrat states. That is a risky strategy.

Instead, if Obama wins Ohio and Florida, he'd be at 273 votes even if Obama loses every single other swing state. McCain could win Colorado, Iowa, Wisconsin, Michigan, Indiana, Virginia, North Carolina, and South Carolina and yet McCain would lose. Ohio and Florida have recently gone to both Democrats and to Republicans. Ohio and Florida both have McCain with only a 1% lead. Thus, Obama could very realistically pick up those states and have the win.

Pennsylvania isn't a place for Obama to spend much time or money. PA has gone Democrat in all of the last 4 presidential elections and Obama has a 7% lead there. Sure, Obama can't afford to lose PA, but as of right now, it is a fairly certain win.

 

[mshan]

I did click on the links, but when I see Clinton beats McCain Big, I become skeptical of the validity of those polls (e.g. how are electoral votes counted if lead is within margin of error?)

Did click on that Polano's 538 website, and it actually seems to give McCain a slight edge over Obama right now (seems to be consistent with snapshot sentiment right now).

Do appreciate your hypothetical electoral math above, and guess I got to start paying more attention when Tim Russert or Chuck Todd pull out their electoral count chalkboards and really start disregarding macro-national popular vote percentages. :thumbsup:

 

[ProfJohn]

It is highly unlikely that Obama wins Florida. The demographics don't line up for him.

Lots of old people, military and Hispanics. Three groups that Obama struggles with.

 

[dullard]

Originally posted by: mshan
I did click on the links, but when I see Clinton beats McCain, I become suspicious of the validity of those polls (e.g. how are electoral votes counted if lead is within margin of error?)

Right now, that website gives a win as a win, even if it is in the statistical margin of error. That is why I went back and took out the narrow wins (going from 242 to 226) in my examples.

But, things will change. It is quite early. Only after the Hillary voters split among their three choices (Obama, McCain, or stay home), then we can seriously look at the picture.

Obama has to climb out of a hole (he probably won't even win the democrat party popular vote), so it is going to be harder for him to win. Remeber, like ProfJohn said, Obama does poorest in the swing states (ones that Hillary would have an easier time winning). And it is only the larger swing states that really matter. But it is certainly possible.

 

[theeedude]

Ohio is going to decide it again, mark my words.

 

[mshan]

"Unlike some of the other stuff I do here, this is not necessarily meant to be predictive. I'm not necessarily saying there will be big increases in turnout among these groups. Instead, it's meant to be illustrative of how the map would change if Obama did get higher turnout from these groups. The whole thing is kind of in the conditional tense."
http://www.fivethirtyeight.com...ity-turnout-model.html

Sounds like statistical analysis is valid, but his hypothetical increased turnout percentages may have been chosen completely arbitrarily.

 

[miketheidiot]

Originally posted by: ProfJohn
It is highly unlikely that Obama wins Florida. The demographics don't line up for him.

Lots of old people, military and Hispanics. Three groups that Obama struggles with.

*against hillary

 

[b0mbrman]

Originally posted by: mshan
I was googling for "Obama April Fundraising" and articles about this Poblano Effect showed up: http://www.burntorangereport.c...wDiary.do?diaryId=5780

What was striking to me is that, in a supposed best case scenario, Obama could win the popular vote by only 5.5% and still get 80% (350) of the electoral votes.

Can anyone with real statistical expertise, and who is willing to put personal political views aside, analyze this study and tell me if there is any validity to the statistical analysis, and also whether the hypothetical increased turnout levels are concievable under any realistic scenario.

He could lose 100%-0% in some states and win 51%-49% in others.

McCain could do the same thing

 

[Aegeon]

Originally posted by: senseamp
Ohio is going to decide it again, mark my words.

This is by no means certain.

All Obama needs to do is win all the states Al Gore won in 2000, plus New Hampshire and Colorado and he wins the election even if he loses both Ohio and Florida. Alternately, just winning Virginia without winning Colorado and New Hampshire would do the trick. While I certainly expect Obama to compete in both Ohio and Florida, he doesn't have to win in those two states to win the election.

Edit: Actually I'm recalculating it and it looks like just winning Colorado would do the trick for Obama and New Hampshire is not even required.

 

[b0mbrman]

Originally posted by: Aegeon

Originally posted by: senseamp
Ohio is going to decide it again, mark my words.

This is by no means certain.

All Obama needs to do is win all the states Al Gore won in 2000, plus New Hampshire and Colorado and he wins the election even if he loses both Ohio and Florida. Alternately, just winning Virginia without winning Colorado and New Hampshire would do the trick. While I certainly expect Obama to compete in both Ohio and Florida, he doesn't have to win in those two states to win the election.

Edit: Actually I'm recalculating it and it looks like just winning Colorado would do the trick for Obama and New Hampshire is not even required.

I'd keep your eye on the upper Midwest. Whoever takes 2 out of 3 in MI, MN, and WI, will get the presidency.