Gallup/USA Today Poll: McCain +4

[rayofreason]

Source: Real Clear Politics

This one should set the tongues wagging:

Republican presidential candidate John McCain moved from being behind by 6 points among "likely" voters a month ago to a 4-point lead over Democrat Barack Obama among that group in the latest USA TODAY/Gallup Poll. McCain still trails slightly among the broader universe of "registered" voters. By both measures, the race is tight.

Read more: http://time-blog.com/real_clear_politics/2008/07/gallupusa_today_poll_mccain_4.html

---

Is this the poll that can't shoot straight? Yesterday Obama was +9%. Only difference is "Likely voters" vs. "All voters".

[liberalmuse]

1. I don't think so.

I smell a rat, and usually that means there's a rat. I mean, how many people in how many agencies and corporations are loyal to Republicans, and absolutely would not want to see Barack Obama get the Presidency? I'm not talking conspiracy, but I wonder what their methodology was in this one? You can always manipulate questions in order to get the answers you want. Nope. Sorry. This poll is flawed.

[maseman]

2. Two different polls

One is the Gallup Daily tracking poll of registered voters. The other (The one with McCain +4) is with likely voters.

Please please PLEASE remember this folks...one poll is not right. Look at 10 polls like on RealCLear Politics and that is the best way to figure out what is going on.

[Selatius]

3. Let me post this graphic from that link:

[liberalmuse]

4. More telling. Thanks!

Anyway, it looks like McCain got a little bratwurst bounce there. Or was it the falling applesauce that made more people want to vote for him?

[radfringe]

14. I'm waiting for the poll that asks

Who would you rather have a bratwurst with... ;)

[underpants]

15. Still trailing and there is a out-fo-the-norm uptick there

as I expected. thanks for posting this

[liberalpragmatist]

5. Among registered voters it's 47-44

So the outlier reading is a result of their likely voter screen. Likely voter screens at this stage are useless -- at this point, nobody knows what the turnout will be like in November.

[eppur_se_muova]

6. BEE-yewel-SH!T.

McCain is probably not ahead even in his midafternoon dreams.

[NBachers]

7. All it's gotta be is close enough

for the voter caging and the machines to make a "plausible" surprise.

Read graywarrior's "Obama Doesn't Sweat. He should." in todays DU.

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=389x3688051

[melody]

8. Nice try -- look at the fine print

It was a very small subsample poll of "likely voters". Likely voters always are predominantly Republican and older.

McCain is still going to lose bigtime.

[curious one]

9. This a big time BS that can't pass the smell test. Do not trust the polls.

[still_one]

10. If people want a continuation of bush then they should vote for mccain

It is as simple as that, and if that happens, there would be no abiguity this time. It would mean the stupidity of the country is real

The choices could not be clearer

[noamnety]

11. I'm having flashbacks

Remember this?

CNN/USA Today/Gallup Poll
February 4-6, 2005
Bush Approval/Disapproval 57%-40%

Republicans: 378 (37%)
Independents: 353 (35%)
Democrats: 279 (28%)

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x1582259

[happygoluckytoyou]

12. its what you get when you poll outside the PussyCat Strip Club

[Delphinus]

13. I am just simply amazed

that there's any closeness in the polls. To me, anyone with a brain can see Obama is the ONLY choice.

[robcon]

16. Turnout will be the key.

If young people vote in good numbers in November - something that has never happened in a presidential election (although the turnout for Obama in the primaries among young people was strong) the election will be a cakewalk. Otherwise, it will be tight.

The pollsters don't know what to do. "Registered voters" have ALWAYS overestimated the Democratic vote, since young registered voters (more likely to be Democrats) are usually much less likely to vote than older registered voters. "Likely voters" is also an imperfect measure, since 2/3 of people say they will vote, while only about half actually vote.

I think the poster above is right: look at all the polls to see the preponderance of votes, and the trends among voters.