IA-AG (Selzer): Miller +2

[President Punxsutawney Phil]

Miller should still win, but a defeat is not hard to imagine, but even if he wins, this will be the closest win of his career.

[President Punxsutawney Phil]

Miller will likely lose, by the way. Same poll has Grassley ahead by a very underwhelming 53-41, and it’s not hard to figure out how the "undecideds" are leaning here.

Strongly suspect this is another Hubbell 2018 situation (where they had the losing Democrat "ahead" 46-44 in their final poll).

A 47-45 lead is a bit different from 46-44...that's a smaller number of undecideds. And it's frankly really lazy to just assign all or most of the undecideds to the Rs. Your criticism recently makes a considerable amount of sense but, the way you interpret it, leads towards some rather difficult-to-agree-with conclusions and/or result in something that goes too far down the other direction...
Reality is that absolute certainty is a mirage and ironically, you nurse it almost as much as the people you (rightfully) criticize.

[President Punxsutawney Phil]

A 47-45 lead is a bit different from 46-44...that's a smaller number of undecideds. And it's frankly really lazy to just assign all or most of the undecideds to the Rs. Your criticism recently makes a considerable amount of sense but, the way you interpret it, leads towards some rather difficult-to-agree-with conclusions and/or result in something that goes too far down the other direction...
Reality is that absolute certainty is a mirage and ironically, you nurse it almost as much as the people you (rightfully) criticize.

Here’s reality:

1. Your 'argument' that a 47-45 lead is different from a 46-44 lead is the definition of grasping at straws (margin of error is a thing, esp. as far as vote share is concerned).
2. Undecideds have consistently leaned Republican in this state.
3. Polling has consistently underestimated Republican strength in this state.
4. Your predictions have consistently overestimated Democrats virtually everywhere (that’s because you desperately want Democrats to win and aren’t willing to approach things in a more objective manner).
5. I never used "absolute certainty" in my post ("Miller will likely win" and "strongly suspect" do not convey absolute certainty), so that is a lie.

You’re not interested in arguing in good faith. Consider this my last reply to you.

You may not be the worst poster on this site, in terms of the most idiotic or asinine, but you almost certainly are the worst person who posts on this site. Just unbelievably unpleasant to the core.

I hope karma catches up to you some day. And I really hope that day is Tuesday. Then again, as I've said before, if KY 2019 didn't humble you, nothing will. There will always be excuses. You can, of course, never be wrong, because narcissists never can be wrong in their eyes.

Thanks for defending me here. Nonetheless, I do want to add a disclaimer that I think this is a more inexact "science" than most give it credit for and what he said about excessive trust in models is bang-on correct, no matter how repulsive you might or might not think the mouth saying that is.
If Miller loses in November I'll give him credit for getting this right if reminded, even if the thought process might not be one I agree with fully. I'd say more, but I'll leave it at that.

[President Punxsutawney Phil]

2020 WV-Treasurer redux? Although I'm expecting this for 2024, if Miller keeps running for reelection.

This race seems almost comparable to Manchin's 2018 reelection, when he won narrowly over Morrisey after having defeated Raese in a landslide six years earlier. Miller didn't have a Republican opponent in 2018 and was reelected overwhelmingly against a Libertarian candidate. But now, if he wins, he won't win by more than a few percentage points. As with Manchin, he might just have enough lingering goodwill to get over the finish line one more time.

I never thought up this comparison before, but it makes sense.