« American Crossroads Ad Eviscerates Obama (JWF) |
Main
|
Seasoned To Death »
July 25, 2012
To Answer Drew's Post Yesterday About Michigan
Per Mitchell Research, you're damn right it's in play.
Last poll- Obama by 1.
This one? Romney by 1.
Two firms in a row showing more movement to Romney (Rasmussen went from O+8 to O+6 yesterday). For what it's worth, Mitchell has the largest sample size of Michigan polls released this week, 825 Likely versus 500 for Rasmussen, and with it a smaller margin of error.
Marist (Obama+4), Epic-MRA (Romney+1), WAA (Romney+2) and a host of other firms who found Obama-Romney within MOE either way a few weeks back have yet to release new numbers. When they do they will either confirm this trend or...ok...
Before anyone points out the giggle-inducing PPP release today, remember Wisconsin when they came in sixth for accuracy. They have a long, long way to go before I take their independent polling seriously (like when they switch over to a likely voter model in September). When they are paid (ex- for DailyKos), they sober up. Now? Funnytime at the crazy bar. Give em traffic if you like, but a 14-pt outliner like the one they are smarming about? No linky today. Also amusing- this time in 2008 they had no problem using likely voter models for polling. I wonder why they are holding out this go-around...