So what should be done about Nebraska?

 So...I came across a post online with Nate Silver's polling averages and it had Kansas at like R+3 for the senate. i know the governor's race is expected to be in play, but is Kansas REALLY in play? Well, looking at the polls, it's mostly a lot of dem aligned internal polls. Low quality. But wait, Nebraska is the same thing. No polls on realclearpolitics, and the polls reported by other sites are low quality internal polls that don't pass muster on more reputable outlets. I also checked out south carolina, same thing. Annie Andrews floods the averages with internal polls suggesting she's winning even though South Carolina is basically safe R. 

So...this is making me rethink including Nebraska in my model. The reason I've been so interested in this race is because Dan Osborn got within 7 points in 2024. If we actually have shifted 8 points since then, it's plausible he can win. At the same time, it's also possible the 2024 matchup was more favorable to him. After all, the state went by 20 for Trump in 2024, Osborn overperformed by 13 points. But if we're seeing R+13 states like Ohio, Iowa, and Texas performing close to 50-50 as we have been, that means under NORMAL conditions, we should still see Nebraska go like R+12 or something. Osborn is an overperformer here. And he could overperform again. I don't know. 

The problem is, if I include it in my model as is, I should also add kansas and south carolina and their junk data. Seriously, the polls on Nebraska are just as bad. All Osborn sponsored polls, low reliability. Probably way too D optimistic. 

So...what should be done with Nebraska? Well, I could revert to some placeholder data and just throw a number in there that may be realistic or not realistic, but I dont know what that number should be. I'm seeing anything from D+1 to R+12 as realistic in the current environment, and if I split the difference it would be somewhere in the ballpark of R+5.5. Osborn last time performed around R+6.6. He could perform better than that, but I don't know where.

If I were to rate the race in terms of general probability, I'd be inclined to say either lean R or likely R. Lean R is the range between R+1 and R+4 in my model. Likely R is R+4 to R+8. 

If I did a kneejerk probability rating, I'd probably be inclined toward the upper part of lean, giving the possibility of a flip somewhere around 1 out of 6. That's 83% which is just short of the 84% of an R+4. So I'm leaning toward either an R+4 or an R+6. Maybe an R+5 to split the difference. If I were to estimate a range of performance I'd say it'll likely go between like R+11 or so to D+1. Once again, R+5. 

So....with that said, I'll axe the polling data which is partisan, low quality, and unreliable, and assume an R+5 rating. I mean, all of the factors I'm considering like the range of possible performance, to relative position of Nebraska vs other states, to Osborn's past performance, seem to converge around R+5. So until I get reliable data, as in, reputable data that RCP actually reports, I'll consider it R+5. 

So with that said, here's my updated senate prediction:

Yeah, that looks a bit more realistic. Anyway, that's my updated senate prediction.