The Runoff Forecast

0:01 Edmund Hale — This is The Fathom Report, extended edition, with Edmund Hale.

0:04 Edmund Hale — A word before we start. This edition draws on reporting and analysis from The Leadsman, the analytical newspaper. Where something is the paper's judgement rather than established fact, we will say so.

0:16 Edmund Hale — The headlines. Russia's October strikes on Ukraine's power grid were described as the most intensive yet.

0:21 Edmund Hale — The Philippine Senate is expected to reach a verdict in its vice president's impeachment trial. And the Nobel in medicine rewarded switching on brain cells with light.

0:31 Edmund Hale — After three short reports, our editor, Rose, joins me to examine a forecast: the paper's reading of Brazil's presidential runoff, and the claim that history is already on one side.

0:41 Edmund Hale — First, Ukraine. The reporting says October has been the most intensive month yet of Russia's campaign against civilian infrastructure.

0:49 Edmund Hale — On the seventh, Russia launched one hundred and thirty drones and forty-eight cruise missiles, with ballistic missiles as well.

0:56 Edmund Hale — On the fourteenth, about thirty-nine ballistic missiles and more than one hundred and twenty drones struck twenty-six power facilities, leaving six hundred thousand households without electricity.

1:07 Edmund Hale — The analysis sees the targets moving from generation, to substations, to district heating.

1:13 Edmund Hale — It adds that high-voltage equipment takes twelve to eighteen months to supply, which limits what European help can do this winter. Those are the paper's assessments.

1:22 Edmund Hale — In the Philippines, the reporting has the Senate President expecting Vice President Sara Duterte's impeachment trial, which began in July, to end before the Senate's Christmas break.

1:33 Edmund Hale — The prosecution dropped three of the four articles on the thirteenth of October.

1:37 Edmund Hale — Only the charge of unexplained wealth remains. Conviction needs sixteen of the twenty-four senators.

1:43 Edmund Hale — The paper predicts a conviction, by sixteen to eighteen votes, at moderate confidence. The defence is due to present from the ninth of November.

1:51 Edmund Hale — And in medicine, the Nobel Assembly at the Karolinska Institute honoured Karl Deisseroth, Peter Hegemann and Georg Nagel for optogenetics, the use of light to switch specific nerve cells on and off.

2:02 Edmund Hale — The piece notes one patient, blind from retinitis pigmentosa, who regained enough sight to discern objects on a table and grasp them, a result it calls modest.

2:12 Edmund Hale — It cautions that outside the eye, no such therapy has yet reached the final phase of trials.

2:17 Edmund Hale — Now, Brazil. With me is our editor, Rose. Rose, a note for listeners first.

2:22 Rose Calloway — Please.

2:22 Edmund Hale — The Leadsman's piece was published on the twenty-fifth of October, while voting was still under way, so it contains no result.

2:29 Edmund Hale — In this conversation we are testing its reasoning, not announcing a winner. And to be clear, we are discussing evidence, not preference.

2:37 Rose Calloway — Agreed on both counts. A forecast is only interesting once you ask how it could fail.

2:41 Edmund Hale — Set the scene for us.

2:42 Rose Calloway — Brazil has about two hundred and fifteen million citizens, roughly a hundred and fifty-six million of them registered.

2:48 Rose Calloway — Polls were due to close at five in the afternoon Brasilia time, eight in the evening Coordinated Universal Time.

2:54 Rose Calloway — The voting is electronic, and the article says a projected winner is typically called within ninety to a hundred and twenty minutes of the last polling station closing.

3:03 Edmund Hale — And the final polls?

3:04 Rose Calloway — Datafolha and Quaest, which the article calls the two most established pollsters, both had Flavio Bolsonaro on fifty-one percent of valid votes and Lula on forty-nine.

3:13 Rose Calloway — The average of credible polls in the final week was roughly the same.

3:16 Edmund Hale — A two-point lead.

3:18 Rose Calloway — Which the article itself says is within the standard margin of error for that split. So, in its words, a Lula victory cannot be statistically excluded by the polls alone.

3:26 Edmund Hale — Then why does the piece lean so firmly towards Bolsonaro?

3:29 Rose Calloway — Two further things. A prediction market, Polymarket, that priced him above eighty percent throughout the final week. And what the piece calls the historical constraint.

3:39 Edmund Hale — Which is striking. Since the return of presidential democracy in nineteen eighty-nine, no candidate who finished second in the first round has gone on to win the runoff.

3:48 Rose Calloway — And Lula finished second, on about forty-five percent to Bolsonaro's forty-seven.

3:52 Edmund Hale — That seems to me a strong piece of evidence.

3:54 Rose Calloway — A strong pattern, which is not quite the same thing. The article does not say how many runoffs sit behind it, and that is the number I would want. A pattern is only as strong as its cases.

4:03 Edmund Hale — It offers a mechanism. First-round results act as a signal of credibility and momentum, which shapes where smaller parties' voters go and how energised the soft supporters are.

4:12 Rose Calloway — A plausible mechanism, but notice that it is argued, not measured. Fair enough for an analyst. Less fair as proof.

4:18 Edmund Hale — I think you are being hard on it. A mechanism, an unbroken record and the polls all pointing the same way. That is three separate supports.

4:25 Rose Calloway — Separate? My reading is that the market is trading on the polls and on that very history, so I would count it as an echo rather than a third witness. The article itself calls it a probabilistic indicator, not official data.

4:36 Edmund Hale — A real-money echo, though. And the article notes that the market already accounts for the transfer-vote argument, and still prices him above eighty percent.

4:45 Rose Calloway — True, it says so, and I concede that. The market is not ignoring the case for Lula.

4:49 Rose Calloway — It still does not tell us whether the market is right, and a price above eighty percent for a race inside the margin of error means the history is carrying a great deal of weight.

4:57 Edmund Hale — Or the market has priced something the polls cannot see.

5:01 Rose Calloway — It might. We are guessing at the same thing from opposite ends, which, frankly, is the state of the evidence.

5:07 Edmund Hale — Let us do the sums on the other side. The article's steel-man for a Lula win rests on voters of the eliminated first-round candidates.

5:14 Rose Calloway — About eight percent of valid votes. Forty-seven plus forty-five is ninety-two, which leaves eight. Now, fifty-one to forty-nine in the polls implies those eight points splitting evenly, four each.

5:25 Edmund Hale — And Lula needs more than fifty.

5:27 Rose Calloway — So he must take more than five of the eight points, leaving Bolsonaro under three. Call it a split of better than five to three, against four to four in the polls.

5:35 Edmund Hale — Your own arithmetic, I take it.

5:36 Rose Calloway — My back-of-envelope, from the article's numbers, and not the paper's claim. It assumes everyone turns out again and votes validly, which the article does not discuss.

5:44 Rose Calloway — But it shows what the steel-man needs: a clear lean among those eight points.

5:48 Edmund Hale — The article says such transfers have historically favoured the left when the pool is large enough.

5:54 Rose Calloway — And in the same breath says it is uncertain, and not directly sourced, whether eight points is large enough to overcome a two-point deficit. It is honest about that.

6:02 Edmund Hale — It also suggests polls may under-sample Lula-aligned constituencies.

6:06 Rose Calloway — A claim with no number attached. So the case for Lula rests on something unmeasured, and the case for Bolsonaro on a pattern with an uncounted sample.

6:14 Rose Calloway — Neither is comfortable. That is what a race inside the margin of error looks like.

6:18 Edmund Hale — The article does commit to a forecast. Flavio Bolsonaro wins with roughly fifty-one and a half to fifty-two percent of valid votes. Confidence, moderate to high.

6:27 Rose Calloway — And it names its own failure mode, a transfer surge for Lula that the polls missed. I would add one thing. Two pollsters, identical numbers. That is either reassuring agreement or shared assumptions. The article does not say.

6:39 Edmund Hale — Is that a fair worry? Independent pollsters landing in the same place is usually good news.

6:45 Rose Calloway — Usually, yes, and I have no evidence beyond the coincidence, so I will leave it as a question rather than a charge.

6:50 Edmund Hale — If you were watching the count, what would you look at?

6:54 Rose Calloway — The article gives an order. First, the Sao Paulo metropolitan result, the largest bloc of votes in the country. Then the states of Bolsonaro's biggest first-round surplus.

7:02 Rose Calloway — Then the states where the eliminated candidates' coalition was largest, because that is where the transfer story is tested.

7:08 Edmund Hale — Then there is what a result would mean. The article spends real time on governing.

7:12 Rose Calloway — Lula's party and its allies hold about two hundred and ninety of the five hundred and thirteen seats in the lower house, which is more than half. So a Bolsonaro presidency would face an opposition-controlled lower house.

7:23 Rose Calloway — The article says that would be a first under the nineteen eighty-eight constitution, with veto points over the budget, appointments and major policy, and coalition-building that has meant handing ministries to partners.

7:34 Edmund Hale — And on foreign policy?

7:36 Rose Calloway — The article says his campaign has stressed closer alignment with the United States on trade and defence, a revised stance on China over technology standards, and a harder line on Venezuela and Cuba.

7:45 Rose Calloway — It stresses those are campaign statements as reported, not enacted policy.

7:49 Edmund Hale — And a Lula presidency?

7:51 Rose Calloway — Historically unprecedented, it says, but continuity abroad, and a legislature on his side. Executive and legislature aligned for the first time since his last term.

7:59 Edmund Hale — The article's bottom line is that the race is, on the numbers, Bolsonaro's to lose.

8:04 Rose Calloway — Which is the sort of sentence you write about the favourite on the morning of the race. Favourites sometimes lose, and the article is careful to say so.

8:11 Edmund Hale — So where does that leave your verdict?

8:13 Rose Calloway — A careful piece, leaning on one historical pattern and one market that are doing more work than their evidence quite supports. It says as much, if you read the qualifiers.

8:22 Edmund Hale — I would say it earned its lean.

8:24 Rose Calloway — And I would say it earned a footnote. I think we can agree on that much, Edmund, and no further.

8:29 Edmund Hale — Quite right. Rose, thank you.

8:31 Rose Calloway — Thank you. It was a pleasure.

8:34 Edmund Hale — That was The Fathom Report, extended edition. Our thanks to Rose, and to The Leadsman, whose analysis we drew on throughout. This is Fathom FM.