Saturday, 29 October 2022
We are now less than two weeks away from the 2022 US Congressional Midterm elections. These elections will determine the balance of power in Washington DC for the next two years and could well set the tone for the upcoming Presidential elections, including the widely expected return of you-know-who. If there is one thing we have learned from the past few years, it is that elections can have serious repercussions. All around the world, eyes will be on the United States.
Introduction
The Democratic Party heads into this midterm election with full control of Government. They hold the White House as well as majorities in the House of Representatives and the Senate. The White House, of course, is not up for grabs this year, but every seat in the House faces an election, as do 1/3 of Senate seats. Much to play for, as it were.
Today I will be taking an in depth look at the House and Senate separately. But before we get into that let's quickly go through the golden rules of midterm elections, which apply to both the House and Senate elections and are vital towards understanding the dynamics at play this year.
1. Midterm elections generally show lower voter turnout than Presidential elections. I have written in past election cycles about why this might be the case, but it essentially boils down to people just caring less. Rightly or wrongly people just see the President as a more important and more glamorous role. Everyone knows who he is, he's a celebrity. By comparison, very few Americans can even name their Congressional representative.
2. Low voter turnout generally favours the Republicans. This one might seem less obvious, but statistically it is undeniably true. We could spend an entire article discussing the reasons why this might be the case, but most experts will agree that it comes down to something quite obvious. In a low turnout year where people are less motivated to vote, the most likely people to still show up and vote are a) those who care the most (ie those who are most switched on to the latest political happenings, spend more time watching cable news, etc) and b) those with the easiest opportunity to vote (ie those with more free time on a Tuesday). So who watches a lot of cable news and is less likely to have no work or other commitments on a Tuesday? The elderly, and low-education voters, two voting blocs who have very heavily backed Republicans in recent years.
3. Midterm election turnout almost always favours the party in opposition. This is another fact that is very clear in the data, the party which holds the White House almost always comes off worse in the midterms. It makes perfect sense really. Fear and anger are great motivators to vote, and the party out of power is invariably the angrier.
1. Midterm elections generally show lower voter turnout than Presidential elections. I have written in past election cycles about why this might be the case, but it essentially boils down to people just caring less. Rightly or wrongly people just see the President as a more important and more glamorous role. Everyone knows who he is, he's a celebrity. By comparison, very few Americans can even name their Congressional representative.
2. Low voter turnout generally favours the Republicans. This one might seem less obvious, but statistically it is undeniably true. We could spend an entire article discussing the reasons why this might be the case, but most experts will agree that it comes down to something quite obvious. In a low turnout year where people are less motivated to vote, the most likely people to still show up and vote are a) those who care the most (ie those who are most switched on to the latest political happenings, spend more time watching cable news, etc) and b) those with the easiest opportunity to vote (ie those with more free time on a Tuesday). So who watches a lot of cable news and is less likely to have no work or other commitments on a Tuesday? The elderly, and low-education voters, two voting blocs who have very heavily backed Republicans in recent years.
3. Midterm election turnout almost always favours the party in opposition. This is another fact that is very clear in the data, the party which holds the White House almost always comes off worse in the midterms. It makes perfect sense really. Fear and anger are great motivators to vote, and the party out of power is invariably the angrier.
This last point, in particular, is really key. The track record of the President's party losing midterm elections is extremely strong and goes back decades. It's such a strong record that it holds true regardless of who the President is or how popular they are. Even Barack Obama, a popular and successful two-term President who left office with the highest approval ratings of any outgoing President, lost both of his midterm elections by wide margins.
Based on these factors, our starting presumption should be that the Republicans are a strong favourite to win. They typically benefit from the lower turnout and enthusiasm of midterm elections and this year have the added benefit of being in the opposition. But despite this, the available data heading into these final weeks does not appear to indicate the red-wave blowout we might expect.
In fact, despite the Republican party's apparent advantages this election cycle, Democrats appear to be leading in most national polls. The polls released this week (at the time of writing) have Democrats leading by +1 (Tufts), +3 (Echelon) and +4 (YouGov), with only the usual partisan pollsters showing a Republican lead. An average of the most recent ten polls (excluding partisan pollsters from both sides) has Democrats ahead by 1% and most non-partisan polls in recent weeks have had them 1-2% ahead on average.
The situation appears particularly worrying for Republicans if you look at other elections that have taken place so far this year, typically a good predictor of midterm success. Fivethirtyeight's analysis suggests that in special elections held since the summer, Democrats have outperformed their partisan baseline by 11% on average, an astonishing result that would be more in keeping with a blue wave than a Republican leaning or even a close election.
We find ourselves in a very unusual position for a midterm election, one where the fundamental electoral presumptions are ostensibly at odds with the actual data and polling. This gulf is actually perfectly illustrated in the FiveThirtyEight forecast, where a polls-only analysis has Democrats as solid favourites to hold the Senate, whereas the expert opinion has it as a toss-up.
Now before we get into our final predictions, let's have a quick look back at how this blog did last time around. The Ephemeric's 2020 election predictions correctly called the winner in all three of the Presidential race, the House race and the Senate race. The Senate predictions were within 1 seat of the actual results and the Presidential predictions called every single state correctly except for Iowa and the lone electoral vote from Maine's 2nd Congressional district. The Ephemeric was one of the few places to correctly predict Democrats to win in Georgia and Republicans to win in Florida. I say all this not to brag but just to point out, before we get into this year's forecast, that this blog has a formidable track record in predicting US elections.
So what to make of the unusual 2022 situation? I think the Republican advantage still holds, but it won't be as golden a year as it should have been. I will dive deeper into my reasoning behind this, but first of all the headline prediction you all came here for: The Ephemeric predicts the Republican Party to retake the House of Representatives and the Democratic Party to hold the Senate.
Current House Map: Democrats - 220, Republicans - 212.
Predicted House Map: Democrats - 202, Republicans - 233.
Approximate Net Change: Republicans gain 15 - 20 seats.
So first thing is first. The uninitiated may well look at all the red on that map and assume that Republicans are steam-rolling these elections. A reasonable thing to think, but look closer and you will see that most of that red belongs to just a few very large seats, covering vast tracts of rural land where very few people actually live. If you zoom in to the denser population centres you will see dozens of much tinier, but far more populated blue districts.
On paper, 2022 should be a blow-out year for Republicans to retake the House. Currently Democrats hold only a small majority, 220 seats with 218 needed. As we have already discussed, the opposition usually performs well in midterm elections and the Republican Party in particular has (in recent history) benefitted from the lower turnout and enthusiasm of the midterm election cycle. The average gain for an opposition party in a midterm is around 25 seats, and recent cycles have tended to be even higher, 40+. They only need to pick up 6.
As a reminder for those new to American politics: gerrymandering is the process through which partisan actors draw up the borders of Congressional districts in such a way that voters more likely to vote for your opponent are pooled into as few districts as possible, while your voters are spread into as many districts as possible whilst still maintaining a lead in those districts. This process effectively allows you to increase the number of seats you win, even if you don't increase the number of votes you get. It results in some bizarre and shockingly manipulative district boundaries. For lack of a better description, it's a legal way of rigging a democratic election. For further clarity on how this trick can be used to manipulate election results, please see the diagram below.
Gerrymandering has been a huge factor in recent election cycles. The Republican control of the 2010 redistricting process allowed for Congressional maps to be gerrymandered to an absurd extent for these past 10 years, bad enough that it was estimated Democrats would need to win by around 5% nationally (a near landslide margin) just to break even in the House. This was even put to the test in 2020, where Democrats did win by about 5% nationally, and only just squeaked a majority in the House.
Now look back at those polls, what do you see? Democrats leading by 1-2%, almost exactly the hypothetical break-even margin. If we take the polls at face value, this race is essentially a toss-up. Sure, you could argue that historical precedent would tip the scales in Republicans' favour, but equally you could point to the recent special election results as tangible proof that this advantage may not in fact be materialising.
Current Senate Map: Democrats - 50, Republicans - 50.
Predicted Senate Map: Democrats- 50, Republicans - 50.
Approximate Net Change: Democrats gain 0-1 seats.
Key states to watch: PA, GA, NV, WI, NC, OH, AZ
If the House is looking like a relatively safe bet for the Republicans, the Senate right now is showing a small but clear lean in the other direction towards the Democrats. This really shouldn't be the case. The 2022 map may not be as absurdly one-sided as 2018 (a year where Republicans managed to hold the Senate despite a blue wave), but it is still pretty favourable for them with most of the competitive races in conventionally red states like Arizona, Georgia and North Carolina as well as 2016 Trump states like Wisconsin and Pennsylvania. But while there are enough close races here that a Republican majority is still very possible, Democrats head into the final stretch as favourites.
Conclusion
So there it is. By no means a foregone conclusion, but the polling right now strongly suggests an election night that results in a split Congress. In a year where Republicans were widely expected to sweep the legislature, that would have to be seen as a big disappointment. If I had to put numbers to it, I would say the Republican House is 75% likely, and Democratic Senate 60% likely. There is still time for this to change, and the fine margins are such that a wide range of realistic outcomes exist, but if I were a betting man this is how I would place my money.
House of Representatives Verdict: Republican Majority
Current House Map: Democrats - 220, Republicans - 212.Predicted House Map: Democrats - 202, Republicans - 233.
Approximate Net Change: Republicans gain 15 - 20 seats.
The above map is based on data from a variety of aggregators, including Pollster, Fivethirtyeight, and analysts including Sabato and Cook Political, and shows the expected House map. The rest is pretty self explanatory: dark blue represents safe Democrat wins, light blue leans Democrat, grey is toss up. Meanwhile on the other side, light red to dark red represents lean to likely Republicans.
So first thing is first. The uninitiated may well look at all the red on that map and assume that Republicans are steam-rolling these elections. A reasonable thing to think, but look closer and you will see that most of that red belongs to just a few very large seats, covering vast tracts of rural land where very few people actually live. If you zoom in to the denser population centres you will see dozens of much tinier, but far more populated blue districts.
On paper, 2022 should be a blow-out year for Republicans to retake the House. Currently Democrats hold only a small majority, 220 seats with 218 needed. As we have already discussed, the opposition usually performs well in midterm elections and the Republican Party in particular has (in recent history) benefitted from the lower turnout and enthusiasm of the midterm election cycle. The average gain for an opposition party in a midterm is around 25 seats, and recent cycles have tended to be even higher, 40+. They only need to pick up 6.
In an ordinary midterm election cycle, these factors should make the requisite gain a mere formality. But what about those polls and special election results? In order to talk about those in greater detail we need to discuss the one additional key factor in House elections: gerrymandering.
As a reminder for those new to American politics: gerrymandering is the process through which partisan actors draw up the borders of Congressional districts in such a way that voters more likely to vote for your opponent are pooled into as few districts as possible, while your voters are spread into as many districts as possible whilst still maintaining a lead in those districts. This process effectively allows you to increase the number of seats you win, even if you don't increase the number of votes you get. It results in some bizarre and shockingly manipulative district boundaries. For lack of a better description, it's a legal way of rigging a democratic election. For further clarity on how this trick can be used to manipulate election results, please see the diagram below.
Gerrymandering has been a huge factor in recent election cycles. The Republican control of the 2010 redistricting process allowed for Congressional maps to be gerrymandered to an absurd extent for these past 10 years, bad enough that it was estimated Democrats would need to win by around 5% nationally (a near landslide margin) just to break even in the House. This was even put to the test in 2020, where Democrats did win by about 5% nationally, and only just squeaked a majority in the House.
However, Democrats' recent electoral successes have allowed for a much more equitable redistricting process in 2020. Republicans maintain their gerrymandered advantage, but it's much smaller than it was before, with current estimates suggesting Democrats now only need to win by about 2% nationally to break even.
Now look back at those polls, what do you see? Democrats leading by 1-2%, almost exactly the hypothetical break-even margin. If we take the polls at face value, this race is essentially a toss-up. Sure, you could argue that historical precedent would tip the scales in Republicans' favour, but equally you could point to the recent special election results as tangible proof that this advantage may not in fact be materialising.
To put it another way, even though the Republicans on paper only need to pick up 6 seats, the recent redistricting has actually moved the baseline further away from them. It's almost as if the Democrats have an extra ten or so seats. So while it may look as though Republicans only need to slightly outperform their 2020 results in order to take a majority, the less favourable gerrymander this year means they really need to pick up more ground. To be clear, the additional margin is not huge (well within the historical average for an opposition party gain) and they would still be solid favourites to do so. But if the less than stellar polling data proves to be accurate, it might be a bit closer than people are expecting. I should note that there has been some evidence of polls trending towards the Republicans in recent weeks, which may indicate they will outperform the current polling, but this seems to have levelled out in recent days.
As for the polls themselves. "How can we trust them?" I hear some people ask. "They're always wrong." Polling has gotten a bit of a bad rep in recent elections, unfairly so. The 2016 and 2020 polls were nowhere near as far off as conventional wisdom would have you believe. Clinton led by 4% in the polls and won by 2%. Biden led by 8% in the polls and won by 5%. Most significantly, the polling in the most recent midterm cycle, 2018, were almost exactly spot on, and in fact overestimated the Republican Party's support. To be clear, there have been certain states with consistently big polling misses in recent years (FL, OH and WI in the Republicans' favour, NV and arguably GA in the Democrats' favour), but in general and nationally the polls have tended to be reasonably accurate without any particular bias.
So despite the hype, I don't buy that there's enough evidence to support a theory that the polls will inherently underestimate Republicans. That simply isn't true historically and certainly was not true in the most comparable recent election cycle, 2018. You could well have a polling error in either direction of a few percentage points, and given the fine margins that could well be significant in determining the balance of power. But anyone who tells you they know which way such an error will go or that Republicans can be presumed to be outperforming the polls is talking pure nonsense.
I have ultimately gone for a Republican majority. The polling and special election results may not be indicative of the red-wave that history suggests they should achieve, but the margin they need to gain is still small enough that a modest polling improvement from 2020 should be enough. I think Republicans will be able to win the majority, but it will be a smaller gain than would have been expected in such an advantageous electoral climate, 20 or so seats compared to recent cycles' 40+ gains.
Senate Verdict: Democratic Majority
Current Senate Map: Democrats - 50, Republicans - 50.
Predicted Senate Map: Democrats- 50, Republicans - 50.
Approximate Net Change: Democrats gain 0-1 seats.
Key states to watch: PA, GA, NV, WI, NC, OH, AZ
If the House is looking like a relatively safe bet for the Republicans, the Senate right now is showing a small but clear lean in the other direction towards the Democrats. This really shouldn't be the case. The 2022 map may not be as absurdly one-sided as 2018 (a year where Republicans managed to hold the Senate despite a blue wave), but it is still pretty favourable for them with most of the competitive races in conventionally red states like Arizona, Georgia and North Carolina as well as 2016 Trump states like Wisconsin and Pennsylvania. But while there are enough close races here that a Republican majority is still very possible, Democrats head into the final stretch as favourites.
The Senate right now is split evenly, 50/50, with the Democratic Party controlling a majority by way of Vice President Harris' tie-breaking vote.
The Republican Party has three clear pick up opportunities: AZ, GA and NV.
AZ and GA are conventionally red states, while NV is one of those states that is always viewed as a toss up, even though it has reliably voted Democrat in recent years. Interestingly, of these three it is Nevada that is currently polling the closest, but to be quite frank I am skeptical. There haven't been many high quality polls there and polls in recent electoral cycles have pretty consistently overestimated Republicans. The polls are close enough that I will call this a toss up for now, but I would still expect a Democratic hold here and so far the early voting data would seem to support this. Everyone is looking at NV as the most likely pick up for Republicans, I think it is overrated.
Meanwhile, Democrats have actually led pretty consistently in the conventionally red states of AZ and GA. There are two main reasons for that. 1) These states are far more purple now than conventional wisdom has accepted. 2) The Republican candidates are exceptionally weak. But while I think Republican chances in NV are overrated, they may actually be underrated here. In a year where the national electoral environment favours Republicans, they could well outperform polls in conventionally red states such as these. In addition, they lead comfortably in the polling for the Gubernatorial races in both of these states, and at a time where ticket splitting is becoming increasingly rare, I could easily see the two Senate candidates riding their coattails to a win (as happened in a few races in 2018).
To be clear, Democrats are narrowly favoured in all three of these states. But the margins are narrow enough that the law of probability suggests that there's a good chance Republicans pick up at least one.
Democrats, meanwhile, have four realistic pick up opportunities: NC, OH, PA and WI.
One of these, PA appears to be leaning in their favour. As above, the main reason for this is the gulf in quality between candidates, with Republicans inexplicably tapping celebrity quack and noted New Jersey resident Dr. Oz to run against the popular, salt of the earth Lieutenant Governor John Fetterman. There have been doubts raised about Fetterman's health in recent weeks, but so far there isn't any credible polling to show it having an impact on the race. At the same time we have an inverse of the situation in GA and AZ, with a Democratic Gubernatorial candidate leading polls. I expect the combination of Fetterman's sustained polling lead and the coattails of the Democrats' popular Gubernatorial candidate to secure a pick up here.
NC, OH and WI are more interesting. Polls had generally been showing these races as competitive, or even leaning in the Democrats' favour over the summer. There has been some movement back to the Republicans in these final days, but the margin is very narrow. The races may appear competitive, but I am skeptical. OH and WI in particular significantly overestimated Democrats in 2016 and 2020 (although not in 2018). I think Democrats are in a similar situation in NC, OH and WI as Republicans are in NV. The polling is solid, but the reality may not be quite so favourable.
Ultimately, despite the favourable electoral climate, the odds appear to be stacked against Republicans when it comes to the Senate. If Democrats do manage to pick up PA, Republicans will need to win two of NV, AZ and GA. This is absolutely achievable, but with Democrats leading polls in all three it's a tall order. They would also need to hold off Democrats in NC, OH and WI, where the margins are narrow enough that a Democratic pick up is entirely plausible. Republican hopes of taking the Senate most likely rest on holding that PA seat, which they are not currently favoured to do. Democrats retaining a majority with either 50 or 51 seats seems the most likely.
Conclusion
So there it is. By no means a foregone conclusion, but the polling right now strongly suggests an election night that results in a split Congress. In a year where Republicans were widely expected to sweep the legislature, that would have to be seen as a big disappointment. If I had to put numbers to it, I would say the Republican House is 75% likely, and Democratic Senate 60% likely. There is still time for this to change, and the fine margins are such that a wide range of realistic outcomes exist, but if I were a betting man this is how I would place my money.
