Looks like Gallup is going to get sued again by the Holder Justice Department. David Axelrod is probably on the phone haranguing them right now, talking about “how we do things in Chicago.”

As I have been able to figure it out, here is what Gallup does: Gallup conducts automated telephone interviews with 3,050 registered voters. Then they cull out the likely voters, according to their likely voter procedures. In this case, the numbers shown above are for the likely voter response from 350 such voters culled out of the total, leaving 2,700 classified as registered voters.

The likely voter sample is deemed more useful because it more accurately reflects what will occur on election day.

The 2,700 in the response group that constitute registered voters went 48 for Romney, 46 for Obama.  Obama typically does better in polls of registered voters than he does with likely voters.

The overall margin of error for the total sample is plus or minus 2%.

Rasmussen Reports 3-day tracking poll of 1,500 likely voters has Romney 49% — Obama 48%.

Neither poll represents the influence of last night’s debate, if any there be.