This counter-response essentially says that any form of weighting in survey research is herding. If so, I love herding!

He's right about the motivation: All weighting is done to ensure that you don't get crazy results. But that's a feature of a good poll, not a bug!

https://twitter.com/NateSilver538/status/1853322147919180265

After all: Is there a principled difference between weighting on age (to ensure that your sample includes youngs and olds) and weighting on past vote (to ensure you get folks from across the political spectrum)?

Lately @NateSilver has been arguing that pollsters are "herding" —tweaking or hiding their results to avoid publishing outliers.

I don't know any (reputable) pollsters who do this.

And I think the problem here is Nate making a simple math/stats error 🧵 Let's explore binomial distributions and the standard errors of weighted samples.

TL, DR: Weighting across groups with different voting patterns should change how you calculate confidence intervals.

Fail to do so, and you'll falsely accuse pollsters of herding.

Focusing for a moment on the POLICIES at play.

Voters were asked whether they supported 100 different policies, but weren't told whether the policy was from Trump or Harris.

Nearly all of Harris' policies got majority support from the public, and... 🧵 Far fewer of Trump's policy proposals garnered majority support, and almost none were broadly popular (say, >75% support).

Senator, I think it's worth acknowledging that violent crime actually has falllen, and indeed it is lower today than it has been for decades.

https://twitter.com/JDVance/status/1844456663413293260

This decline in violent crime is evident in not just the FBI reports, but also an independent survey by the BJS.

Wowza: HUGE jobs report.

September payrolls grew +254k, well above expectations.

August payrolls revised up +17k to +159k, and July revised up +55k to +144k.

Unemployment fell to 4.1%

This economic expansion that is motoring along. Honestly, there's not much to say here other than that fears the job market had slowed turned out to be a statistical illusion due to incomplete data.

Over the past three months, payrolls is motoring along at +186k per month, on average, which is pretty much where you want it.

Senator, you're misleading folks again. I just read the CBO report you recommend. It actually says that an immigration surge boosts federal revenues quite dramatically, and has only a small effect on mandatory spending and interest.

https://twitter.com/JDVance/status/1841338117506109762

The CBO study @JDVance cites analyzes how immigration improves the *federal* budget.

Yet he pulls a misleading quote, talking about the effects of *state and local* budgets and ignoring the (likely larger) federal govt impacts.

Here's the CBO directly saying this isn't okay:

I posted yesterday about the latest crime data, which showed crime continuing on its decades long fall.

And a lotta reply guys turned up to dispute the usefulness of these data.

So let's dig in, shall we?

https://twitter.com/JustinWolfers/status/1838712445524668818

All crime data are problematic. The definition of crimes change over time, as does what people report, and which police depts provide data.

Social scientists don't ask: "Are these data perfect?", but rather "Do they add to our existing understanding?"

Our existing crime data are surely imperfect, but more informative than anecdotes, facebook vids, and political operatives.

New crime data just dropped (yesterday). It's your choice whether to believe these carefully collected statistics, or the unsubstantiated claims made by political hucksters.

The numbers say: Crime fell last year, it fell by a lot, and it fell pretty much across the board. Taking a longer perspective: Violent crime has been declining for decades. After a brief pandemic spike (more on that in a moment), it's back to falling again.

Senator, it appears upon digging a little deeper that we have misdiagnosed the cause of higher egg prices. They are higher because of the largest and longest outbreak of Avian flu in American history -- an outbreak that has led farmers to destroy 100 million birds (so far).

https://twitter.com/TrumpWarRoom/status/1837581418329002260

No yolk, bird flu can eggsplain the whole thing.

The 2014-15 outbreak led 50m birds to egg-spire, scrambling prices to $1-$1.50 above the normal pecking order.

This outbreak is twice as big (100m dead chooks), likely raising egg price $2-$3. That's the whole egg-stravagent rise

The Fed chose the big cut, reducing rates by half a percentage point to below 4-3/4 to 5 percent.

The economy is getting back to normal, inflation is back down to normal, and so now rates are heading back to normal.

The right decision, albeit a tad late. Fed statement is here. federalreserve.gov/newsevents/pre…

I think this critique is exactly right, so lemme try to put Harris's big econ speech in North Carolina in context next to Trump's big econ speech in North Carolina a couple of days ago...

https://x.com/BlueStTrav/status/1824556412342767848

1. Harris's econ speech was about the economy. Trump's econ speech barely mentioned the economy. Indeed, he had to keep reminding himself that he was meant to talk econ. He was rambling, incoherent, and said almost nothing recognizable as econ.

https://x.com/Acyn/status/1823904506494812539

Lemme 'splain... There's two stories about the economy, and two sets of storytellers. There's the view from Wall Street, and the reality on Main Street. Financial markets tell you about the former, and economists are about the latter. Imma give you the economist's version. And the truth is, when you dig into the economic numbers, the current economy is gloriously, splendidly, brilliantly... boring. This is what an economy that's returning to a healthy normal looks like.

I don't hate this policy as much as I thought I would bc this isn't traditional rent control.

Think of it as macro policy—setting a nominal anchor when the Fed's target isn't working—rather than as micro policy setting the relative price of housing.

The details really matter.

https://twitter.com/JStein_WaPo/status/1812918420423254480

The proposed controls appear not to apply to new construction. And so they still allow the price signal to provide an incentive for new construction.

Blunting price signal is bad — to the extent that it affects those at the margin. But this policy is mostly infra-marginal.

This striking pattern of Trump under-performing his poll numbers continues. According to 538, Trump was meant to win:
- Massachusetts by 37 (NYT says he's up by 23)
- Tennessee by 69 (he's up 57)
- Texas by 64 (he's up 56)
- Virginia by 49 (he's up by 29)
The exception:
- North Carolina by 46 (he's up 51) Plus in those states with more sparse polling there's no 538 average. But the final Trump-Haley poll suggested Trump would win:
- Alabama by 75 (he's up 68)
- Oklahoma by 77 (he's up 66)
- Maine 58 by (he's up 47)
- Minnesota by 64 (he's up 41)
- Utah by 27 (he's up 14)

Are we there yet?

Yes kids, we are: The inflation measure the Fed targets (core PCE) has run at an annualized rate of 1.9% over the past six months, *below* the Fed's two percent target.

This isn't a one-off blip; it's six months of sustained low inflation. Core PCE inflation over the past year was 3.2%, but that tells us a lot about what was happening in late 2022.

Over the past 6 months, inflation was 1.9%; over the prior 6 months it was 4.5%. As those high rates "fall out" of the 12 month window, year-ended inflation will fall.

There's no question people are telling pollsters they're miserable about the economy. But riddle me this: Why can't we find evidence of this pessimism in anything other that public opinion polls?

Every non-poll based indicator of confidence suggests folks are optimistic [thread] Take consumption. If folks were worried about their economic future, you might think they would be squirrelling money away for the hard times coming. But they're spending like they expect ongoing economic strength.

Lemme take a moment to fact check this.
(Spoiler: Every single claim is wrong.) 🧵

https://twitter.com/HouseGOP/status/1726240661022622159

Let's start with the claim "uncooked turkey" is 7.2% more expensive. That's actually the CPI for a broader category of "other uncooked poultry" which includes Turkey.

But turkeys? The USDA says the retail price of fresh Hens is down -7.6% and fresh Toms are down -8.1%

An absolutely bonkers example of the bad-news bias (and a failure to take Econ 101) in econ reporting: The @WSJ observes that the price of metals used in electric vehicles are crashing, and somehow infers that it will create shortages and snarl adoption of EV's. The best part: The headline assures us that this is bad for Biden.

Inflation's receding, baby.

Headline inflation was literally zero -- on average, prices were the same in October as September.

Over the year, headline inflation is 3.2%, beating expectations of 3.3%. The decline in headline inflation is partly due to falling gas prices.

Want to know where inflation is going?

You're better off focusing on core inflation (ex food and energy), which was 0.2% last month, and 4.0% over the past year. Again, better than the markets expected.

A steady-as-she-goes strong jobs report.

Payrolls rose +150k, and while unemployment ticked up a little, it remains below 4%.

Revisions subtracted -101k.

And so over the past 3 months, payrolls have averaged +204k, which is still extraordinary for this stage of the cycle. Remember that this month's payrolls number is probably about -30k lower than it otherwise would be, due to the UAW strike. And we expect that to bounce back as an extra +30k next month.

Blockbuster GDP report shows real GDP grew at an annual rate of 4.9% in Q3, blowing even optimistic expectations out of the water.

This economy is going gangbusters, and it's time for the doomers to apologize for being consistently wrong for two years. When the recessionistas were wrong in late 2021, they said it'll be 2022, and the economy kept growing so they said it'll be early 2023, only to see unemployment hit fifty year lows with inflation falling, so we were promised late 2023 and output is growing at nearly 5%.