Perry Bacon Jr.

I try to rely on facts and logic when thinking about politics. But that may be a mistake, according to a very smart column in today’s Washington Post by columnist Perry Bacon Jr.

Perry Bacon Jr.
[image_credit]Washington Post[/image_credit][image_caption]Perry Bacon Jr.[/image_caption]
Bacon lays out a perhaps common understanding of how politics works, then blows it apart. The common understanding, which I must admit often underlies my own thinking, is that there are liberals, who reliably vote for Democrats; conservatives, who reliably vote for Republicans; and moderates, who swing and who have the power to determine outcomes. That is not quite utter rubbish (I hope), but – according to Bacon’s column, which I found quite convincing – more wrong than right.

He lays out the shortcoming of that understanding across seven statements, which I’ll list, but you should read his full column for the explanation of each of the seven and why each of them undermines the unreliable common understanding described in the paragraph above:

    1. Americans’ views on most issues aren’t deeply held — and can be highly influenced by their party, the media or both.
    2. Besides, views on issues don’t predict election outcomes.
    3. Governing performance often doesn’t matter.
    4. Swing voters aren’t all centrists.
    5. And one of the biggest swings in elections is who turns out.
    6. Plus, wooing “swing” voters isn’t in tension with winning “base” voters.
    7. In the end, politics is unpredictable.

Perhaps you’ll look at that list and think: “I already know all those things.” But if you’re like me, you’ll also realize that you sometimes fall back on all of the myths, even though at some level you know they are myths.

The column is headlined:

“Why ‘moderation’ doesn’t guarantee electoral success for Biden and the Democrats.”

The full, really-quite-helpful Bacon column is here.

Join the Conversation

50 Comments

  1. You could apply Bacon’s judgements to his own column — since none of his statements comes with a 100% guarantee that it is always correct, then it is of no value.
    His logical error is assuming that if a prediction does not have a 100% chance of being correct then it has no value.
    Back in the real world predictions are always statements of probability, one can always find exceptions that prove (in the original meaning of ‘test’) the rule.
    And of course Bacon’s own statements do not come with numbers that show how accurate we can assume that they are.
    So while Bacon’s commentary contains some good questions about the limitations of our political assumptions, they do not make these assumptions invalid or useless.

    1. “His logical error is assuming that if a prediction does not have a 100% chance of being correct then it has no value.”

      I’m afraid the error in logic here it yours not Bacon’s. Nowhere does Bacon make the claim your comment is premised upon. On the contrary he explicitly refers to the unpredictability of politics and elections. All Bacon is doing here if pointing to some falsehoods typically expressed by conventional wisdom. He’s not adopting an absolute position about anything.

  2. I was initially inclined to dispute Paul Brandon’s commentary, but the more I think about it, the more I’m inclined to end up wandering off in a different direction. What Bacon does, I think, is provide an oft-unstated questioning of assumptions that many / most of us make in this political context, and my bias is that questioning assumptions typically has more benefits than drawbacks in that kind of flexible verbal and intellectual environment. Many on the left assumed Trump couldn’t possibly win in 2016, and many on the right similarly assumed that Biden couldn’t possibly win (or conversely, that Trump couldn’t possibly lose) in 2020. Bacon’s last point is one that those of us interested in politics should perhaps have tattooed on our person somewhere: politics is unpredictable. One of Bacon’s points that seems to me spot-on more often than not is that voters are often not deeply invested in a candidate’s policy positions. For many years, I’ve heard people respond to media questions about their support of candidate ‘X’ with something like, “I like his/her policies,” but in those exceedingly rare cases where the reporter then probes more deeply to ask which policies the voter likes, and especially, why the voter likes that particular policy position, the response often approximates the proverbial crickets. They like candidate ‘X’ because… well… they like candidate ‘X.’ Thus, there’s no real logical argument with which to take issue.

    1. Well, thanks, I think.
      One way to categorize all of Bacon’s points is that human behavior is multiply determined; there is never just one reason why a person does something. At most, we might say that one particular event is a tipping point; as the most recent of multiple causes it is the most obvious one, and might be assigned the credit or blame for a particular behavior.
      And I will agree that assumptions should always be questioned, as long as the question itself does not imply a particular answer.

  3. “Besides, views on issues don’t predict election outcomes.”

    “Governing performance often doesn’t matter.”

    Both very true. I expressed a little earlier that we are a 50/50 nation and getting to 60/40 or 70/30 on most major issues is not nearly as difficult as it seems:

    Immigration (Cato!!)
    69% say immigration over time contributes to more economic growth
    61% say immigration enriches American culture and values
    58% say immigrants strengthen our country because of their hard work and talents

    Gun Control (Pew)
    85% of Republicans and 90% of Democrats support preventing those with mental illnesses from purchasing guns
    70% of Republicans, 92% of Democrats support subjecting private gun sales and gun show sales to background checks
    Majorities in both parties also oppose allowing people to carry concealed firearms without a permit.

    Voting Rights (Monmouth & Vox)
    71%, supports that it should be overall easier to vote early in elections.
    91% of Republicans, 87% of independents and 62% of Democrats support requiring some form of identification to vote
    80% Of respondents said they supported preventing foreign interference in elections, limiting the influence of money in politics, and modernizing election infrastructure to increase election security.
    60% Of respondents supported requiring nonpartisan redistricting commissions, a 15-day early voting period for all federal elections, same-day registration for all eligible voters, automatic voter registration for all eligible voters, and giving every voter the option to vote by mail.

    And the list can go on and on…

    This politikin’ stuff is really easy, jus do what we wan ’em to do!

    Unfortunately, what we want and who we vote for have very little relation to one another. And we keep reelecting folks that are way more concerned about keeping their sweet work gigs than doing the people’s business. Never better described than by Bill Maher last Friday at the end of his show:

    https://www.youtube.com/watch?v=XyCNxugvs6s

  4. Bacon avoids the obvious. There’s little difference between the two major parties. Take health care. The choice is between insurance-based health care or more insurance-based health care.

    Issue by issue, we’re stuck fighting it out over Kang or Kodos. That’s how the big boys win.

    1. Look a little closer. In the comment above I listed public opinion on many of the key issues of the day:

      Immigration (Cato!!)
      69% say immigration over time contributes to more economic growth
      61% say immigration enriches American culture and values
      58% say immigrants strengthen our country because of their hard work and talents

      Gun Control (Pew)
      85% of Republicans and 90% of Democrats support preventing those with mental illnesses from purchasing guns
      70% of Republicans, 92% of Democrats support subjecting private gun sales and gun show sales to background checks
      Majorities in both parties also oppose allowing people to carry concealed firearms without a permit.

      Voting Rights (Monmouth & Vox)
      71%, supports that it should be overall easier to vote early in elections.
      91% of Republicans, 87% of independents and 62% of Democrats support requiring some form of identification to vote
      80% Of respondents said they supported preventing foreign interference in elections, limiting the influence of money in politics, and modernizing election infrastructure to increase election security.
      60% Of respondents supported requiring nonpartisan redistricting commissions, a 15-day early voting period for all federal elections, same-day registration for all eligible voters, automatic voter registration for all eligible voters, and giving every voter the option to vote by mail.

      You may also notice that almost all of these popularly supported positions are aligned with the Democrats and opposed by the Republicans.

      So, there are huge differences between the 2 parties and 1 party is in alignment with the majority of the people and the other in alignment with minority opposing interests. To say that both parties are essentially in agreement on healthcare is completely without a basis in practice and fact.

      1. Polls are meaningless. They had Trump getting blown out in both elections. But now we can believe them?

        As for health care, show me where they differ. One is for Obamacare. The other Obamacare-lite.

        1. That is false. The 2016 polls ended with a 2-3 point lead for Clinton, which is where it ended up. Trump won razor-thin victories in close states, and won the electoral college. Biden’s final polling put him up 7-8 points nationally, and it was a couple of points closer than that. Polling did not show a blowout in either race.

          1. Now we’re rewriting history? EVERYONE was told it would be a Clinton blow-out in 2016. Trump given no chance. I knew of no one, including Republicans, who weren’t profoundly shocked at the outcome. But you’re saying different. And you knew all along by the polls. Okay.

              1. Really? Any simple search reveals Trump was the longest of shots. This was literally the first site to come up https://projects.fivethirtyeight.com/2016-election-forecast/ Dozens more follow.

                2016 is considered one of the biggest, if not biggest, upsets in the history of presidential elections. This was all based on polls. All of it. How else do they generate a prediction? Putting a finger in the air?

                Bottom line: Polls are meaningless. They can be manipulated to suit anyone’s purpose at any time, and are.

                1. “Bottom line: Polls are meaningless. ”

                  If you make no effort to understand the margin of error, statistics and probability, polls are meaningless.

                  It is fine to show the 538 2016 results, how about 2012 when they nailed it? And click a little in to the 538 link you provided and you will see hundreds of different polling data sources and many show margin of error differences between Clinton and Trump.

                  And if polls are meaningless, you better let political campaigns know that because they make million dollar campaign dollar decisions based on their polling data. Just ask Paul Manafort as he delivered precinct level polling data to his Russian Oligarch paymaster. Why would the Russian be so curious about meaningless data?

                  1. I don’t know what polling political campaigns use. They may be more accurate, we may never know. If you’re losing, you don’t announce it.

                    I always assumed whoever orders polls sifts through them like a deck of cards, deciding which ones they’ll show us.

                2. They predicted a Clinton win 70% of the time; a nearly 30% chance – or 3 in 10 – for Trump. That is NOT a projected blowout. If you read Nate’s analysis at the time he explicitly called out the possibility that slight shifts in projections in a handful of states could drastically affect the outcome. In that sense he, 538, nailed it.

                  Scroll down to the “crazy and not so crazy scenarios,” they show a 10% chance that clinton wins the popular vote & loses the electoral college. But they also show a clinton “blowout” at 6% – or less likely than what happened.

                  What I think it’s fair to complain about is how pundits and some journalists misrepresented polling, particularly in treating national polling as projecting the next president. Thanks to the electoral college, they are not coupled as tightly as some might assume. Polling is not the problem – vacuous reporting is.

            2. It’s not uncommon for people to hear something different from what was said. If someone told you the election would be a blow out, you more than likely were not listening very well.

              Fivethirtyeight was very clear that Don Trump had a 1 in 7 chance of winning the Electoral College. Events of that probability happen every day.

              1. I listen fine. The polls had it dramatically wrong, and you can’t remake history.

                1. Are you interpreting the 70 to 30 as a 40pt win for HRC? Yes, that would be a blowout. But that’s not what it says.

                  1. There’s a lot of hair-splitting going on. Can’t we agree the polls were wrong– dead wrong– by whatever margin, and it was played full blast by the media?

        2. Well, there is this “margin of error” thing with any poll. If we have 52% vs 48% and the “margin of error” is 4%: Viola, that is a meaningless poll.

          If 80% of respondents said they support preventing foreign interference in elections, limiting the influence of money in politics, and modernizing election infrastructure to increase election security, we can safely say: FOLKS FAVOR IT, and it is a meaningful poll.

          And on healthcare, I missed the Trump “Obamacare Lite” solution. I do remember a few things though:

          1. Trump telling us: “No one will lose coverage. There will be insurance for everybody. Healthcare will be a “lot less expensive” for everyone — the government, consumers, providers.”
          2. Never seeing a Republican proposal that does that or anything else.
          3. 70 Votes moved forward by Republicans to eliminate Obamacare with NO replacement plan.

          I guess we can safely say “Obamacare Lite” is no insurance options beyond what existed in 2008. That sure is “lite”.

        3. Trump WAS ‘blown out’ in both elections — he lost the popular vote by millions.

          1. That’s not the point. The point is the polls whiffed, by wide margins, both times.

            1. That’s false. All the *national* Presidential polls — which were looking at popular vote, not electoral votes — were pretty accurate in both 2016 and 2020. The problem is that the President is elected by electoral votes. So believing those *national* polls and projecting a winner was wrong thinking both times. The way to project a winner is with state-by-state popular vote polls, and those still have a margin of error.

              But people already have posted these details above.

              1. If the polls were misreported as favoring the Democrats, or the margin of error ignored, it’s all the more reason to label them meaningless, which I have.

                1. You wrote, “The point is the polls whiffed, by wide margins, both times.” Then you changed the subject to talk about how the polls were reported or represented.

                  My point remains correct — and you didn’t respond to it — that the polls themselves were pretty accurate; they didn’t “whiff”. I’m sure I wasn’t the only person who understood exactly what the polls were about.

                  1. I changed nothing. You, and others are saying the polls were right. Show me an expert who believes this. Show me anyone, besides the other posters on this site, because you are way out in left field on this one.

                    1. The final 2016 FiveThirtyEight NATIONAL presidential poll had it Clinton 45.7% and tRump 41.8%. (https://projects.fivethirtyeight.com/2016-election-forecast/national-polls/)

                      The final 2016 NATIONAL popular votes totals were Clinton 48.2% and tRump 46.1%. (https://en.wikipedia.org/wiki/2016_United_States_presidential_election)
                      — —
                      The final 2020 FiveThirtyEight NATIONAL presidential poll had it Biden 51.8% and tRump 43.4%. (https://projects.fivethirtyeight.com/polls/president-general/national/)

                      The final 2016 NATIONAL popular votes totals were Biden 51.3% and tRump 46.9%. (https://en.wikipedia.org/wiki/2020_United_States_presidential_election)
                      – – – – – –
                      FiveThirtyEight is reliable, and they got these two results pretty accurately. They didn’t whiff, and the numbers show it.

                      It’s irrelevant to your “whiffed” comment how anybody *reported* the polls, because the polls were good. Somebody may have whiffed on some polls, but it wasn’t on the Presidential NATIONAL vote results.

                      If you still claim that the polling and how the polling was reported are the same thing, please explain how that can be.

                    2. BTW, Ms. Wicklow, about “you are way out in left field on this one”:

                      I think you’ve trimmed your wick so low that the light has gone out. If you knew anything about me, you’d know that I play in straightaway center field and I cover a lot of ground.

            2. You’re saying that the point is not that the polls accurately predicted the popular vote, which is all that they can predict, since the electors are not known beforehand.
              What IS your point?

              1. If they weren’t inaccurate then they were misrepresented. In any case, epic fail.

    2. That’s just silly. The more-insurance based care has enabled tens of millions of people to access affordable care. The shortcoming with the ACA are due to slim Democratic majorities and non-stop opposition from Republicans. One party has worked to expand affordable health coverage and one has not.

      1. “Affordable” sounds like something from a used car salesman. Giving a working person a $5,000 (or $8000 or $12,000) deductible isn’t “affordable”. It’s cruel. But that’s insurance-based care, endorsed by both parties.

        1. 12 Million people would disagree:

          “Marketplace Enrollment Tops 12 Million For 2021; Largest-Ever Funding For Navigators.
          On April 21, 2021, the Centers for Medicare and Medicaid Services (CMS) released the final Marketplace open enrollment report for 2021. Consistent with prior posts that showed an increase in enrollment through HealthCare.gov, overall enrollment across all 50 states and DC increased during the 2021 open enrollment period. This increase—of about 5 percent relative to 2020—underscores the importance of the marketplaces as part of the health care safety net during the pandemic.”

          1. Enrollment, insurance, deductibles, bronze, silver, and platinum plans don’t equal health care. What it equals is record profits for insurance companies.

            1. Oh, the BIG BANG theory of healthcare reform.

              Sorry UHG, you are out of business at the end of the month.

              If you recall , the ACA was to have a public option inside of it. All of the Republicans and semi-Republican, independent Joe Lieberman stopped it. Change to something you will find agreeable is incremental. A 3.5 Trillion dollar reconciliation bill will certainly have healthcare implications. You have to start somewhere and improve from there and 12,000,000 more people with benefits is a good start and is steadily getting better.

              Or just wait for the BIG BANG to occur, and wait, and wait…

              1. You can support insurance-based care. I do not. I would like a choice, however.

        2. As opposed to the previous system where many people had NO insurance and could be liable for hundreds of thousands, or worse.
          I agree that the type of social medicine provided in civilized countries would be better.

          1. “I agree that the type of social medicine provided in civilized countries would be better.” As do I, and almost everyone I talk to. But we’re not given that choice, which is a better way of understanding politics than us/them thinking.

        3. “All you need is MORE insurance. That’s your problem.” Any good insurance salesman knows that line.

  5. If one may look at the list cafeteria style, start with 1 and 4. Many people are prone to easily changing views and swing voters aren’t all centrists. Another way of saying Americans are pragmatic more than centrist, sometimes willing to look further left or further right to suit the high-anxiety moment, for better or worse. This is why the GOP has gone off the deep end over past decades. And why the Democrats’ “moderate” base is liking a lot of progressive ideas.

    1. Put another way “I vote for the person, not the party.” Though I doubt the veracity of that claim – I’ve made it myself, but haven’t voted for a republican since 1998 (have voted 3rd party, at times).

      Also, “one of the biggest swings in elections is who turns out” is underappreciated, I think.

      Trump brought out a lot of first time and infrequent voters. Clinton did not turn out the vote well in certain demographic groups – which did turn out for Biden.

      Similarly, we know presidential years bring out different voters than congessional races.

  6. Well, it’s nice have a summary of some basic and uncontroversial observations once and while. These observations are neither “new” or “original” but they’re quite valid.

    Not to pick on Eric, but I think the most revealing aspect of this commentary is the admission that these debunked assumptions are nevertheless still entrenched in his own thinking. I can see from other comments that Eric’s not the only one who has difficulty abandoning falsehoods and political myths. The question that arises from THAT observation is clearly… why? Why are the assumptions that Bacon is debunking here soooo difficult to abandon? Why do they continue to plague Eric’s mind?

    Other observers decades back answered this question. If you want to explain the intransigence of falsehoods sometime the best approach is to ask whether or not, and for whom do those falsehoods work? Who benefits from the dominant narrative? Mr. Bacon is debunking/revealing a dominant narrative here, who does that threaten?

    You can search for the answer to that question on your own, but I’ll skip to the end here and let you know it reveals a paradox of sorts. While both political parties rely on the dominant narrative to maintain power, they both lose elections because of that reliance. The assumption Eric has so much difficulty setting aside maintain existing power within the Parties. The seeds of their defeat lie in the their own pursuit of victory.

    Just ask yourself what happens if you set those assumptions aside? What happens to the existing regime if candidates and parties start working with a different narrative and set of assumptions like the one Bacon is offering? On one hand we’re just talking abandoning existing comfort levels, but beyond that we’re talking power and it’s maintenance.

    Someone pointed out the fact that despite popular opinion on various issues, our elections still break down 50-50 for the most part. Well, that’s because neither Party fully adopts the most popular positions. The bipartisan regime of failure may fail the nation, but it preserves existing power. These narrow election results are a key feature of both Parties narratives and promises to resolve crises with bipartisan cooperation. If you adopt popular positions and win election by large margins… who win and who loses in that scenario? What happens when or if bipartisanship becomes unnecessary? When you put popular candidates on the ballot, margins and participation increase. When you keep popular candidate off the ballot…

    So this brings us back to Bacon’s thesis that “moderates” may not be the salvation Democrats assume in their existing narrative. Here’s the thing, the problem with the moderates Democrats idealize is that… they aren’t actually THAT moderate. The “centrists” they chase are actually nowhere near the center. This is why predictions based on political models with these assumptions fail so frequently.

    And of course Eric is making: “we we” here… who is this “we” that holds onto these false notions? I know a lot people who abandoned them long ago.

  7. He may be correct that “Americans’ views on most issues aren’t deeply held — and can be highly influenced by their party, the media or both.” There are, however, numerous voters for whom one issue or group of issues is all-important (ending abortion, stopping same-sex marriage, etc.). They tend to be the most vocal in any election campaign or debate, and, more importantly they are the ones most inclined to vote.

    Also, saying that “[s]wing voters aren’t all centrists” just emphasizes the point that terms like “centrist” or “moderate” are overused to the point that they have become meaningless. I suspect that most voters would describe themselves as centrist or moderate, yet their voting will be erratic, based largely on the last emotional appeal they saw.

  8. As to Bacon’s last point, about the unpredictability of politics, I leave this, from a WaPost story:

    “90 percent of Democrats say they have received at least one dose of a coronavirus vaccine, compared with 54 percent of Republicans.”

  9. Not to belabor the point but listen:

    “I try to rely on facts and logic when thinking about politics. But that may be a mistake, according to a very smart column in today’s Washington Post by columnist Perry Bacon Jr.”

    Of course we understand that Bacon’s article isn’t anti-fact or illogical right? Eric’s mistake (and those who share his mentality) isn’t that he relies on facts and logic… the mistake may be that the facts and logic he relies on are not valid, that’s an important distinction.

    To be honest, what Eric what should be saying here is that Bacon is challenging some established assumptions, that would look something like this: “In a very smart article, Perry Bacon challenges the facts and logic I usually depend upon when I analyze politics.”

      1. Paul, I never miss a chance to say: “Sunstein shmunstein”, and then giggle a little bit. As for cognitive illusions and errors (Tversky), they’re fun to think about but the discussion here probably makes more sense in terms of power relations. THESE cognitive illusion if you will, empower one group of people while marginalizing another, it’s not just psych phenomena, these illusions work for some and not others in a very concrete way.

Leave a comment