Claims

The filibuster is not used more often than in the past, but because the parties have become ideologically non-overlapping (bimodal), filibusters are now along strict partisan lines on economic issues—whereas in the mid-20th century the 'conservative coalition' of Northern Republicans and Southern Democrats created cross-party overlap that made the filibuster a less purely partisan tool.

Party identification, conceived not as a fixed inherited trait but as a measure that people revise based on events, is a potentially better predictor of the president's party's fortunes than generic-ballot polling, because voters tend not to jump directly from one party to the other but move through 'independent' as a halfway station—so tracking movement toward independent signals trouble for the president's party.

Presidential-election outcomes are far more predictable than congressional ones: a 'guns and butter' model (real income/economic growth plus involvement in foreign wars) predicts whether the incumbent party wins because voters hold the president responsible for the economy and for foreign wars—the model fails mainly in 1952 and 1966.

The traditional explanation for midterm seat loss—that marginal voters drawn out by presidential-election hype stay home in off years, returning the electorate to its normal party affiliation—broke down around 1988 because Republicans kept winning presidencies while Democrats held the House, so political science lacks a single good explanation for why the president's party loses midterm seats.

Standard congressional forecasting regressions over all 435 seats are misleading because gerrymandering and incumbency fix the outcome of roughly 360 seats, leaving only about 60-70 genuine swing districts; the meaningful action is therefore in those swing districts, not the averaged national model.

No democratic system—including European parliamentary democracies and Japan—where representatives are elected on short terms has yet shown it can sustainably take back benefits it has granted, raising doubt about whether the U.S. (whose slow system resists radical change) can confront its over-promised liabilities, though the German cutbacks were a surprisingly impressive first round.

In the 2008 New Hampshire primary, all roughly 30 pre-election polls failed to show Hillary Clinton ahead of Obama, which she won, indicating something systematically wrong affecting all polls rather than random chance.

Apparent poll movements late in a campaign are often illusory: a methodology shift (e.g. Gallup switching from registered-voter to likely-voter weighting) creates a bump unrelated to real opinion change, and pollsters who err have a ready ex-post story of a late undetected surge—what Rivers calls the last refuge of scoundrels in polling—when in fact the same poll repeated would have produced the same wrong answer.

The convention bounce—a candidate gaining several points (about five per party, net swing ~ten) during/after their party's convention—is partly an artifact of differential response: supporters of the convening party are energized and answer polls while opposition partisans, irritated by the 'phony love fest,' decline to participate, so YouGov controls for prior party identification to neutralize the effect.

Party identification shifted dramatically: Democrats had an ~18-point lead from the New Deal through the 1960s, Republicans closed it to a few points by 2004, but 80% of those Republican gains were erased since 2004 (half of them in fall 2007/spring 2008), which in principle should make 2008 a terrible year for Republicans.

Massive consumer and voter databases that did not exist 25 years ago now provide detailed information (income, home value, etc.) on most people, allowing pollsters to draw a randomly selected target sample from a voter list and then match closely-similar respondents from a large opt-in internet panel, creating a sample that mimics a random sample across many dimensions and removes skews that simple demographic weighting cannot.

Averaging polls (as RealClearPolitics does) does not necessarily wash out error: if pollsters make the same systematic errors or herd toward each other, the average is no more reliable than a single poll; averaging only clearly helps when polls use the same methodology, in which case combining them is equivalent to a larger sample, and ideally polls should be weighted by the reciprocal of their variability.

In 2006, the matched internet-panel method produced election forecasts whose average error was substantially less than the average reported telephone poll, mainly because it removed biases better (not because sampling variability was smaller), and Rivers argues telephone polling could be improved by similarly substituting a respondent who resembles the missed person rather than drawing another number from the same skewed population.

Telephone polling, which arose in the 1970s, replaced in-person area probability sampling because in-person interviewing (door-to-door) was the dominant method from the 1930s until then.

Roughly 80-90% of polling errors come from sample skews rather than from sampling (random) error.

Random digit dialing became the dominant phone-polling method because roughly 30% of the population had unlisted numbers, making listing-based sampling infeasible; it worked well for 20-30 years but broke down in the last 10-15 years as marketing calls, cell phones, and distrust drove cooperation rates from ~70% down to ~20% or less, hurting accuracy.

Pollsters typically select 10 to 20 times as many phone numbers as the number of completed interviews they want, meaning the people who actually respond are a small, non-random slice that skews toward more women, higher education, and higher income, requiring weighting and adjustment.

Final pre-election polls tend to be too similar to each other relative to their known sampling variability, suggesting pollsters herd—choosing among multiple defensible weighting schemes the one that places them in the pack rather than the one that yields a divergent answer—though Rivers attributes this to risk-aversion and the art of weighting rather than dishonesty.

Of about 1,300 polls Rivers examined from the 2008 presidential primaries, fewer than 50 reported anything other than the margin of error for a simple random sample with no weighting—meaning the reported margins of error are systematically misleading, which Rivers calls scandalous.

Typical media polls correct only for age, race, and gender (sometimes education, rarely income) using weighting methods about 60 years old, well behind modern statistical theory, partly from a long-standing belief that random digit dialing was good enough.

Likely the New Hampshire 2008 polling miss came from over-representation of college- and graduate-degree voters (who favored Obama, over-represented by 2-3x and under-corrected) and from unreliable self-reports of voting intention, rather than from racism (the Bradley effect).

Self-reported voting intention is unreliable: about 90% of people say they will vote in a primary but actual turnout is much lower (US presidential turnout has hovered in the mid-50% range, congressional under 40%), and likely-voter screens cannot fully fix this because people answer based on what they think they should say.

Interactive voice response (robocall) polls make up roughly 80-90% of polling done in 2008 because newspapers cut back on expensive live-interviewer polling, and despite low response rates and not knowing who is answering, their record is not bad—largely because IVR organizations pay closer attention to weighting than traditional phone pollsters.

There would seem to be an incentive for a pollster to trust divergent weights and look like a genius (as the lone correct caller of an upset), but in practice small subsamples (e.g. a NH poll with ~100 Republicans) make divergent results look like noise that gets discounted, which is why having 30 polls all wrong in New Hampshire was an extraordinary wake-up call.

Weighting pollsters by their historical accuracy (as fivethirtyeight.com does) is questionable because past success may be partly random luck and because pollsters who erred likely changed their methodology afterward, so historical outcomes don't reliably predict future accuracy—analogous to picking mutual funds by past manager returns.

The Bradley/Wilder effect—the claim that voters tell pollsters they will vote for a black candidate but then don't—was a real overstatement of the black-candidate vote in the early 1980s but disappeared by around the late 1990s; a study Rivers cites found no systematic over-reporting across black congressional and Senate candidates and no comparable 'Whitman effect' for women candidates.

Experiments Rivers ran with Shanto Iyengar (lightening/darkening Obama's skin, and swapping black vs white candidate photos in a hypothetical congressional race) found essentially no candidate-race effect on voting, though about 15% of Democratic primary voters—even outside the South—held racial views considered conservative or racist.

Network exit polls (high-quality, ~50% response rate) have consistently overstated the Democratic share of the vote in presidential elections (2000, 2004), but because this bias appeared uniformly nationwide—including areas controlled by neither party—it indicates a systematic polling tendency, not the voter fraud that naive observers inferred.

Internet panel respondents are compensated with a reward they are not told in advance, and are not told what answers qualify them for a poll, in order to avoid biasing results; this contrasts with the phone tradition of not compensating, where rising marketing-call fatigue may eventually require incentives that introduce their own biases.

Obama's extraordinarily high favorability in fall 2007 faded by spring 2008 as Hillary Clinton ramped up rhetoric, leaving him looking like a run-of-the-mill (not weak) candidate without the magical appeal he had even among Democratic core constituencies; black Democrats, who initially didn't strongly favor him because of the Clintons' ties to blacks, swung to ~90% support once they recognized he was black.

No reason to segregate psychology from politics

There is no reason behavior should differ when examined at the individual level versus at the level of political society, so the mapping of human behavior and its biases onto economics should extend to political science and other disciplines that currently assume classical rationalism.

There are three types of societies ('orders'): the primitive order (small hunter-gatherer bands of 25-200 with little specialization or wealth accumulation), the limited access order or 'natural state' which solves violence through limited access, and the open access order.

A key 'doorstep condition' enabling incremental movement toward open access is the shift from pure privilege/personal judgment (the King deciding in favor of the more powerful) to rule of law for elites—where different lords can sue one another in courts—because such rules can be incrementally extended to a growing commercial class, making their organizations more efficient and the economy grow.

A key advantage of open access societies is the ability to provide benefits impersonally—by virtue of being a citizen or being in a certain condition (unemployment insurance, poverty relief, health benefits)—whereas natural states distribute benefits through who you know and what patronage system or organization you belong to.

When FDR proposed court-packing in 1937 at the height of New Deal popularity, weekly public opinion polls show no majority ever favored it—even beneficiaries of the New Deal opposed it—demonstrating the consensus condition; whereas in Latin America court-packing tends to get majorities because people want the benefits.

Natural states are best understood not as a single dictator controlling access but as a dominant coalition of many specialists in violence and economic activity who respect one another's privileges and thereby prevent violence.

In natural states no one holds a monopoly on violence; instead violence is distributed among many specialists, so the key to peace is ensuring these specialists in violence don't fight by creating rents that give them a stake in peace.

The distribution of rents and privileges must reflect the distribution of access to violence (physical power), because if someone has much greater access to violence than to privileges, they will fight rather than cooperate.

People are far better off being exploited under a peaceful limited access order than living under conditions of civil war, as seen in the collapse of Somalia, Chad, or Rwanda into violence.

The 'consensus condition' is that citizens police constitutional rules by acting in concert against violations; when people agree on what the rules are, they can credibly threaten those in power with removal, making the rules defensible—a key difference between open and limited access societies.

Open access orders rest on three pillars: open access to organizations of all types (economic, political, social, religious), rule-of-law institutions and constitutions, and shared beliefs of inclusion and equality before the law.

The 'tragic brilliance' mechanism distinguishes elections in limited access orders: in open access orders certain goods (water, electricity) are provided to citizens regardless of how they vote, but in Mexico under the PRI, ~80% of local government budgets came from higher governments conditional on the locality voting for the PRI, so they could literally cut your water off for not voting correctly.

Natural states progress through stages distinguished by their organizations: fragile (only the state organization, prone to collapse—Chad, Somalia, Rwanda, Afghanistan), basic (differentiated organizations but all part of the state), and mature (private but elite-based organizations with privilege); effective development helps a country move along this progression rather than leap to open access.

Bueno de Mesquita's 'selectorate' theory fits the limited/open access distinction: small selectorates (e.g., a Politburo) lead states to distribute private goods, while large selectorates (all adult citizens) lead to public goods; but his theory takes the state's monopoly on violence and the selectorate size as given, whereas Weingast argues you must explain how states form by how they solve the problem of violence.

William Riker reinvented the study of politics as the originator of the serious mathematical and game-theoretic study of politics; in 'Liberalism Against Populism' he explains that the best we can hope for from democracy is a system that makes it easy to throw the rascals out.

Ethnically homogeneous countries (like traditional Scandinavia) tend to have larger, more generous welfare states because citizens see recipients as 'their own kind' and trust them not to abuse it; therefore increasing ethnic heterogeneity through immigration actually undermines native support for the welfare state, meaning immigration's net political effect on welfare-state size may be negative.

Voter Participation Rises With Anger

Counterintuitively, voter participation is highest when people are pissed off and lowest when the system is working—as seen in the record-high turnout of the tense 2020 and 2024 elections—which means popular anger meaningfully impacts the system rather than leaving the status quo to continue unchallenged.

Thermostatic Public Opinion Shifts Against Incumbent

Thermostatic public opinion means the electorate moves opposite the party in power—when a left-wing president holds office the country drifts right, and vice versa—as visible in the swing of economic attitudes the moment Biden replaced Trump, which is why no political majority is permanent.

Three Models of Government Decisions

Government decision-making can be understood through three simultaneously valid models from Allison's Essence of Decision: the rational actor model (government as one entity maximizing national interest), the organizational/standard-operating-procedure model, and the bureaucratic/government politics model of internal power struggles; none is solely true—you can only say one is more true than another.

The Principles Are Nonpartisan

The principles Taylor outlines are not partisan: historically both parties have deviated from them (Johnson, Nixon, Ford, Carter in the bad period) and both have adhered to them (Reagan, Bush 41, Clinton in the good middle period), so voters across parties should elect officials who understand and can deliver the principles rather than treating them as the property of one side.