If you look at the current FiveThirtyEight model for control of the Senate it looks like this.

That’s pretty close, especially compared to where the same model was a month ago.

You can see the entire change over time in that image at the top. Democrats had a good run in August and September but then something changed and now we’re almost at parity again. And that was the subject of an analysis Nate Silver wrote for his site today. It’s titled “Why I’m Telling My Friends That The Senate Is A Toss-Up” and in this case he means that he things the GOP’s actual chances of taking the Senate are a little better than what his own model currently shows.

If a friend asked me to characterize the Senate race, I’d say “it’s pretty f**king close,” and emphasize that neither party has much of an advantage. Here’s why.

For one thing, as of Thursday afternoon, Republicans realized a slight lead (of 0.1 percentage points) in the FiveThirtyEight generic ballot average for the first time since Aug. 2…

And all of that is before getting into the chance that the polls could overstate support for Democrats again, as they did in 2016 and 2020. This is a complicated subject; I mostly think the model does a good job of accounting for this, and one should keep in mind there’s also the possibility that the polls could be biased against Democrats. But I’m not entirely confident, so my mental model is slightly more favorable to the GOP than the FiveThirtyEight forecast itself…

But the main reason why I think of the race for control of the Senate as a toss-up — rather than slightly favoring Democrats — is because there’s been steady movement toward the GOP in our model over the past few weeks. In principle, past movement shouldn’t predict future movement in our forecast and it should instead resemble a random walk. (We put a lot of effort in our modeling into trying to minimize autocorrelation.) This year, though, the forecast has moved in a predictable-seeming way, with a long, slow and steady climb toward Democrats over the summer, and now a consistent shift back toward Republicans.

So the generic ballot looks even but that favors the GOP slightly. Then there’s a chance Democratic strength in the polls is being overstated again. And here I’m going to go into the weeds a bit because it’s pretty important.

Nate Silver has argued that while polls clearly favored Dems in 2016 and 2020 they did not do so in 2018 and therefore it’s too early to assume there’s a built in bias. The alternative theory is that polls only undercount GOP strength when Trump is on the ballot. However, as Vox pointed out last month, Silver is arguably wrong about 2018. It’s true that Democrats did about as well as polls predicted overall, but that’s mostly because a lot of races that year were in blue states. But if you look at the competitive races, things look different.

The Senate map that year had an unusually large amount of contests in solidly blue states, none of which proved to be competitive. Democrats outperformed polls in nearly all of those contests.

Yet if we look at 2018’s actually competitive races — which that year were in purple and red states — most Democratic candidates underperformed their polls, and often by quite a lot.

The final margin was more than 3 points more unfavorable to the Democrat than FiveThirtyEight’s final polling averages in Florida, West Virginia, Michigan, Ohio, Tennessee, Missouri, and Indiana. There was only one competitive state — Nevada — in which the Democrat outperformed polls by more than 3 points.

In other words, I think Nate Silver is wrong about 2018 and if so, then there is a pretty consistent polling problem from 2016 onwards. And that makes it (in my view) more likely that we could see that pattern continue again this year. In fact, as I wrote earlier today, we’re seeing a similar pattern where nationally Dems look pretty good (up a point or so) but if you look at battleground states, the GOP is leading by an average of 6 points. Maybe that means the polls are capturing the GOP strength in those battlegrounds this year better than they have in previous years or maybe they’re still underestimating it. If it’s the latter then GOP chances to control the Senate are probably a bit better than Nate Silver’s private mental model which has it as a 50-50 toss-up.

Some analysts are going further and trying to “calibrate expectations” such that anything short of a massive red wave would be a disappointment for the GOP. Here’s a Twitter thread from a political science professor at Vanderbilt University.

A standard model for House elections is Gary Jacobson's: presidential approval, income trends, and # seats controlled by the president's party.

Forecast: Democrats lose 45 seats.https://t.co/hOiyJ5COd1

— John Sides (@johnmsides) October 21, 2022

A model by James Campbell based off of @CookPolitical expert ratings: -36 or -42 Democratic House seats, depending on how you code the ratings.

For Senate, -1 Democratic seat.

— John Sides (@johnmsides) October 21, 2022

The same thing is true if you look at @FiveThirtyEight's polls-only ("Lite") forecast: -15 Dem House seats and +1 Dem Senate seat.https://t.co/dWT4GI0y70

— John Sides (@johnmsides) October 21, 2022

His bottom line: Anything less than an epic blowout is sort of a win.

So if the Democrats lose 20 seats in the House, that is bad for the party — no question. But it is also better than they might have expected. Both things are true.

— John Sides (@johnmsides) October 21, 2022

I don’t think that’s accurate, personally. The GOP overperformed expectations in the House in 2020 so there probably aren’t enough competitive seats left in the country to have a 40-45 seat win this year, not without a generic ballot nearing double digits for the GOP (though some have argued it is possible). What’s more realistic is 25 seats in the House and 1-2 in the Senate. But, hey, this is one case where I would love to be wrong.