Status of the Electoral College

While there aren’t many at the moment (only one that I know of right now, the Muhlenberg College Pennsylvania daily tracking poll), my policy on such tracking polls is to enter them every time the tracking poll has a new sample.  For example, the Muhlenberg tracking poll is a 5-day tracking poll.  The first sample was taken from September 21st through September 25th.  That means a “new” sample would be the sample from September 26th to September 30th, and then October 1st through October 5th, and so on.  If there is a three day tracking poll, then I’ll add it every three days, and so on.

I just though I would post that now since were starting to see some state-level tracking polls pop up.

One of the things that I’ve found out about my current methodology is that it is very resistant to change once they have a lot of state polls at once.

In a way, that’s intentional, but the methodology was set up to accommodate 1 or 2 polls in any particular state a week and not 6 or 7.  As a result, polls from a week or two ago, which otherwise I wanted to keep propped up due to the lack of polling, are now making it harder for a slew of recent polls that all say the same thing to change the status of a state.

Take North Carolina for example.  That trio of bad polls for Obama from 2 weeks ago are still dragging him down in that state.  Yes, those polls are only counted at 1/3 strength right now, but a poll with a 20% lead counted a 1/3 strength can still pull a lot of weight around.  Yes, North Carolina is now finally a Too Close To Call state, but it probably should have occurred sooner than it did, and the calculated margin in the state may still be larger than what it is in reality as well.

While I don’t think it’s particularly proper to change my methodology at this point in the race, this is yet another thing that I may wish to consider when and if I do this in 2012.

Having said all of that, Obama may finally be looking at breaking through this week, or maybe next, despite the resistance of states to shift based on the above methodological bugs (or features, depending on your point of view).  Five lean Obama states, holding 67 Electoral Votes, are closer to becoming “Weak Obama” states than they are lean McCain states.  The question is whether Obama can get enough state polls which show him far enough ahead in any, some, most, or all of those states to actually make them light blue on the map.

If all five states were to switch to Weak Obama in the next couple of weeks, that would push Obama’s official total up to 269, which would, of course, mean that any other state would give him an Electoral College win, and at worst he ties which, the way it’s looking, would mean that the House would make him President if all House Democrats vote for him (which they should).

But even if Obama doesn’t get them all, he can still start adding to his total.  If both states which are currently within 1% or less of becoming Weak Obama states actually get there, Obama would jump to 229, which would give Obama a 66 Electoral Vote head start on McCain.

Also good for Obama is that virtually all of McCain’s lean states are closer to switching to lean Obama than they are to switching to Weak McCain.  McCain’s states where his lead is under 2.5% account for 76 Electoral Votes.  If they were all to switch to lean Obama, the count with leaners would put Obama just shy of 350 Electoral Votes.

This is a very dangerous time for McCain, where the election could really start to get away from him.  As Nate over at 538 notes, polling wise, this is the strongest position Obama has been in since the start of the election, which is exactly where he wants to be with only 5 weeks to go.

New polling has pushed North Carolina from Weak McCain back into the Too Close To Call category, pushing down McCain’s official number down to 158, and increasing the Too Close To Call number up to a record 178 electoral votes, a full third of all Electoral Votes available.

Meanwhile, new polls in Virginia has finally pushed that state from lean McCain to lean Obama, and increasing Obama’s Electoral Count with leaners to 286-252.  However, Obama’s lead in the state stands at 0.1%, so any poll showing McCain ahead should push it back into lean McCain.

Besides the beforementioned New Hampshire, almost all swing states have been either inching towards Obama, or at least inching away from McCain.

The top 4 closest battleground states are now all lean McCain, with New Hampshire with a margin of only 0.5%, Virginia with a margin of 0.7%, Nevada with a margin of 1.6%, and Ohio with a margin of 2.1%.

Indeed, for the first time a quite a while, no Obama state has a margin within 2%, with only two states with a margin within 3%, Colorado and Pennsylvania.  Indiana and Florida also have a margin within 3%, on top of the 4 other states already mentioned.

So, to round it up:

For states within 1%:

McCain 17, Obama 0.

States within 2%:

McCain 22, Obama 0.

States within 3%:

McCain 80, Obama 38.

And while I mark my Too Close To Call margin at 5%, I think 3% may be closer to the margin of where a candidate needs to be if they want to legitimately have a shot at picking off a state on election day.

Note:

I should include that this doesn’t include this morning’s Mason-Dixon Florida poll, since I don’t have info such as number of respondents and margin of error, but it’s quite possible that after that poll, Florida will make it a complete top 5 for McCain as far as closest states go.

Despite the surge towards Obama overall, New Hampshire has been going the opposite direction: the two most recent polls in that state have shown McCain ahead by 3% and 2%.  As a result, remarkably, a state which at one time was a strong Strong Obama state is now lean McCain (though still Too Close To Call).

As a result, the count with leaners is how showing the nightmare scenario: 269-269.

The slippage by McCain in the state polls is continued today by a Research 2000 poll showing McCain only ahead in Missouri 49-45.  This was finally enough to tip Missouri back into the Too Close To Call category from Weak McCain.

The official electoral college count now sits at Obama 202, McCain 173, Too Close To Call 163 as we still have an extroardinarily high number of close state remianing, after states appeared to be slipping to one side or another a couple weeks ago.

For the first time in about a week, obama is back over 200 Electoral votes with Washington moving from Too Close To Call to Weak Obama this afternoon.

Recent polls have also pushed lean McCain states pusher to lean Obama states, with Ohio only having a 1.7% lead for McCain and his lead in Indiana down to 2.6%

Meanwhile, Obama is rebounding in other states, including Michigan where he now leads by 3.9% and now may be in a position to push Michigan back into his count soon.

P.S.: If you notice on the sidebar, we’ve now passed 500,000 total respondents for polls conducted in 2008 (that I count).

I outlined the crazyness in the polling in North Carolina last night, but that crazyness has now apparently extended to Virginia and Florida.

Look at the last 5 polls in Virginia, all conducted between 9/10 and 9/15:

• McCain +9
• Obama +4
• Obama +2
• Tied
• McCain +7

Again, it is statistically inconsistent to have polls both showing a tie or slight Obama lead as well as a significant McCain lead.  It just doesn’t make sense.

This disease has also shown up in Florida where the last 5 polls there, all conducted between 9/11 and 9/17 show:

• McCain +5
• Tied
• Tied
• Tied
• McCain +6

That’s not as bad as North Carolina or even Virginia, but again, the chances that all of these polls are correct seems low.  If we went with the majority for all 3 states, then North Carolina would be within 3% or so, Obama would be ahead in Virginia, and Florida is tied.  But there is not only a poll contradicting these, but several polls doing so.

I’d really like to know how different pollsters manage to get so different results?  Part of it could just be weighting for Likely voters, and the +6 Florida poll is also of “adults” and not registered or likely voters, but the fact that we keep getting such different poll results is crazy.

Thanks to some new polls today by the Indianapolis Star and Survey USA, we have two states which have changed status, both in favor of Obama.

First, Indiana has flipped back to being Too Close To Call, taking 11 EVs away from McCain’s total.  In the meantime, a new Survey USA poll has pushed New Mexico back into the “Weak Obama” category, giving Obama 5 EVs, for a total swing of 16 Electoral Votes, currently giving Obama a 7 Electoral Vote lead.

Recent telephone polling in North Carolina has been so unbelievable that it’s hard to figure out how one can even get such results.

In the past six telephone surveys - all of which were at least started no later than September 6th - have the following results:

• McCain +20
• McCain +4
• McCain +3
• McCain +17
• McCain +11
• McCain +1

At first when the first 3 polls came out, and we had +4, +3, and +20, it was naturally assumed that the +20 poll was an outlier.  Then we got the +17 poll and we had two pairs of polls diametrically opposed to each other.  Now after today, we have two threesomes of polls which are diametrically opposed to each other.

However, when one looks at it closely, one can actually see how we can get the results that we do (or at least make them more plausible).  Here are those polls with the margin of error with 95% confidence, applied to both candidates (remember, the first number would be O+moe and M-Moe and the second number is O-Moe and M+moe, thus why a poll with a 4% MOE has a 16% margin):

• McCain +12.4% to +27.6%
• Obama +3.8% to McCain 11.8%
• Obama +5% to McCain +11%
• McCain +9% to McCain +25%
• McCain +3% to McCain +19%
• Obama +6% to McCain +8%

Taking these ranges, we get two pretty plausible ranges that the polls may fall into.  Both ranges include 4 of the 6 polls.

The first range leaves the best poll for McCain and the worst poll for McCain out of the mix.  If we do this, the four middle polls all fall into a range between 9% and 11% within their 95% confidence level.  The two remaining polls fall outside this range by only 1% and 1.4%

The second range takes the 4 most pro-Obama polls.  In this case, the range would be between McCain +3 and McCain +8 within their 95% confidence level.  It’s quite a big larger range, and the two polls left out are left out by 1% and 4.4%.

Obviously McCain wold prefer the 11% number while Obama wold prefer the 3% number, but i think the truth is somewhere inbetween, and the lead probably sits somewhere between 7% and 8% right now (my average has it at 7.9%).  In any case, for those who are desperate to know where North Carolina stands, we can probably say it stands somewhere between McCain +3% and McCain +11%.  Which of course is the difference between being a Too Close To Call state and being a Strong McCain state, so needless to say the range is pretty useless.

Also, having at minimum 1/3 of the polls as outliers, no matter how you cut it, is still a bit much.  And there is also another point here: there are in a way two type of outliers.  There are statistically significant outliers and polls which aren’t statistical outliers, but are effectively outliers.

This is what I mean:

If you have 5 polls, all with a margin of error of 4%, all showing a race of between 2% and 4%, and then a poll comes out with someone 22% ahead with a margin of error of 4%, there is no way that the 22% poll and any of the 5 other polls can jive and still all be within all of their 95% confidence levels.  I would consider that a statistical outlier.

But let’s say a poll came out which showed a race of 15% with a margin of error of 4%.  Given the exercise above, one could say “well, even the closest poll could possibly have a real margin of up to 10% while the 15% poll could be as low as 7%, so where’s the problem?”  Of course, the issue here would be: what are the chances that you have 5 polls whose results are between 2% and 4% when the actual margin is between 7% and 10%?  Sure, it’s possible, but it becomes quite a bit less likely.  It would make more sense to say that the 15% poll is effectively an outlier, even if it isn’t statistically one.

So where does North Carolina come into this?  In my example above with the two ranges for the NC polls, we’re kind of treading into the territory of the “effective outlier.”  The problem is that the 3 polls showing a race within 5%, the poll showing a 11% race, and the two polls showing a near or at 20% race are all effectively outliers of each other (not to mention that we’re guaranteed at least 2 statistical outliers as well).

But unlike the example above with 5 close polls and a 15% poll, enough polls are in each camp that we just can’t know which group of polls consists of the outliers.  Even though the 11% poll is by itself, it’s still taking up the middle ground between the two other groups.

My current score for North Carolina as of this writing - 7.9% - is 3.1% away from any of the six polls.  How oven do you have an average of polls where the average is so far away frm the results of any of the polls in the average?  Yet that’s what we have.