RT Strategies (Various Congressional Districts)

[Sam Spade]

Although these Indiana results are interesting, they lead me towards further suspicion of the quality of these polls.

My main rationale here lies within the recent Research 2000 poll released a couple of days ago in IN-09 showing Hill up by 6%.  R2000, for any of you who have ever read Vorlon's comments, is a state polling firm that conducts polls similarly to M-D.  This usually means steady results, as opposed to SUSA for example, which tends to jump around a lot.  However, R2000 is, for all intensive purposes, a lesser quality firm than M-D; they tend to cut a lot of corners in the developement of their polls. 

In the polling world, this invariably means that their numbers tend to lean Democratic in comparison to the whole.  I'm not saying this is by much, but in 2004, their numbers were, on the whole, about 1-2 points too Democratic. (in 2000, when Dem turnout numbers at the end shot up dramatically, they were much closer to even. I consider that luck) And because their polling method is fairly consistent, these findings are fairly consistent as well.  I'm trying to think of a race where R2000 overestimated on the Republican side and none comes to mind.  It's just that consistent.

Now, I'm sure the response would be:  Well, how do we know that R2000 didn't just get a bad sample for once or an overly Republican sample?  This is quite correct in my opinion.  That's the reason why I called it suspicion, not certainty.  Only more polls can clarify this opinion.

[Sam Spade]

In the polling world, this invariably means that their numbers tend to lean Democratic in comparison to the whole.  I'm not saying this is by much, but in 2004, their numbers were, on the whole, about 1-2 points too Democratic. (in 2000, when Dem turnout numbers at the end shot up dramatically, they were much closer to even. I consider that luck) And because their polling method is fairly consistent, these findings are fairly consistent as well.  I'm trying to think of a race where R2000 overestimated on the Republican side and none comes to mind.  It's just that consistent.

So, Sam, does that mean you think Wilson is even further ahead in NM-1 then?

Well, their result matches ABQ polling, which about matches Madrid internal polling.  I know what the ABQ record was in 2004, so that makes me suspicious a bit, but I doubt we get any other polls out of this race to compare, so I gotta believe what I have. 

How's that for a non-answer.