HomeBlog › Regime Change
Read

Read vs Listen: Regime Change

Seventeen hours of reporting by two named correspondents, with a third person reading every word of it.

Read vs Listen·Verdict last reviewed 11 August 2026·6 min read
The verdict

Read it on Kindle

A room of named officials to place, and the back matter that would help you place them stays in the print edition.

Definitely readDefinitely listen
76%
Read

Confidence 76%, which puts it in the leaning band. Bands run from close call at 50 to 64, through leaning at 65 to 79 and clear at 80 to 94, up to clear cut at 95 and above.

Regime Change

Maggie Haberman and Jonathan Swan

Genre
Political non-fiction
Published
2026
Length
496 pages
On audio
17 hrs 6 mins
Reporting basis
More than 1,000 interviews on deep background

Why we reached this verdict

Every book on this site is checked against the same ten questions. They are based on research into what most people find easier to follow in each format, not on whether the book is any good. This is the first non-fiction title on the site, so the same ten are applied and the ones written with novels in mind are read across to reported work.

Three reasons to read it
There is a long roll of named officials to placeRead across here as a named-source load rather than a count of viewpoints. The reporting draws on more than 1,000 interviews conducted on the ground rule of deep background, and one reconstructed meeting alone is attended by nine named officials, only two of whom are quoted directly. Nobody has counted the named people in the book, so the total is unknown, and no cast of characters list is documented anywhere.
The vocabulary is institutional and nothing is spelled out for youThe terms in play are departments, agencies, military structure and legal process rather than invented or technical language, and no glossary is documented in any listing or review checked. Combined with the named officials, there is nothing a listener can turn to when a name or an office comes back forty minutes later. On the page you can go backwards; the recording keeps going.
The print edition carries apparatus a listener cannot openOne reader reports over sixty pages of notes, sources and references at the back, along with an index. Treat that carefully: it is a single uncorroborated count, and the same reader's figure for how long the index runs does not stand up for a book of 496 pages. What is certain is the four named printed part dividers and the methodology statement at the front, which a reader can turn back to whenever a passage rests on an unnamed source.
The chronology moves forwardBroadly linear, covering the first year and change of a second presidential term, with the four parts sitting inside that order rather than cutting across it. Nothing found describes flashbacks, braiding or dated sections running against each other.
Nobody has described the writing as something to slow down forNo source found characterises the sentences either way. The one relevant fact on record is about compression rather than texture: the authors say they cut thousands of pages, and that the manuscript would otherwise have run to about 2,000 pages.
One reason to listen to it
It is built to be gone through, not dipped intoRead across here as whether the book pulls you along or gets used as a reference. Four named parts, Whirlwind, Retribution, The Enemy Within and Plunder, a broadly linear run through one year and change, and no exercises, checklists or worksheets in any description. One Audible listener review opens by calling it well structured and covering a lot of event timelines. Nothing found states outright how readers actually use it, so this rests on the shape rather than on documented behaviour.
There is a great deal running alongside the authorial voiceNo source states whether the two authors write as one continuous third-person voice or take chapters and sections in turn, so the voice itself is unsettled. What is documented is the volume of named figures and sources moving through the reporting, which is the load this question is meant to catch.
Neither of the two reporters reads itRobert Petkoff reads the whole book and no listing credits Haberman or Swan with any recorded portion, including the introduction. This is not a neutral. The book is built on their own interviews, on an in-person interview with the subject in March 2026, and on an opening statement in which they set out their own rules for attribution, and none of that reaches the microphone.
The recording is a straight reading and nothing elseOne narrator, unabridged, out on the same day as the print edition. No second performer, no archival audio, no author-recorded introduction and no supplementary PDF advertised on any Audible, Apple, Kobo, Chirp or audiobooks.com listing checked.
Nothing here turns on how a line is saidThis question covers tone, humour, testimony and emphasis, and none of those is what the book is described as running on anywhere in the sources. No review or listing consulted discusses the delivery, so this is a not proven rather than a documented failure.
The deciding factor
Two reporters, and the third voice is the one you hear

The book is built on Haberman and Swan's own reporting, and neither of them reads a word of it. Set that against what a listener gives up, which is the index, the notes and the attribution rules at the front, and those are exactly the things a reader uses to keep a large cast of named officials straight. It goes to Kindle on that combination. It is a leaning call rather than a clear one, and the reason is the evidence: the strongest single argument for print rests on one reader's uncorroborated count of the back matter, and part of that same count does not hold up. Settle it against a physical copy and this could firm up or soften.

The recording at a glance

One narratorMulti-voiceFull castDramatised
Narrated by
Robert Petkoff, reading the whole book, and the only performer credited on every listing checked
Length
17 hrs 6 mins at Simon & Schuster UK and 17 hrs 7 mins at Apple Books, or roughly 11 hours 24 minutes at one and a half times speed
Publisher
Simon & Schuster Audio in the US and Simon & Schuster Audio UK, released 23 June 2026, the same day as print
What audio adds
Nothing documented beyond a single unabridged reading. No award for this recording was found as at 11 August 2026
What you miss
The index and the notes, sources and references at the back, the four printed part dividers, and the opening statement of attribution rules that a reader can turn back to

What this book actually is

Four hundred and ninety-six pages, published on 23 June 2026 in both territories, built in four named parts: Whirlwind, Retribution, The Enemy Within and Plunder. The authors have talked about that four-part shape in interview and credit their editor at Simon & Schuster with it. The chronology is broadly linear and covers the first year and change of a second presidential term. Both authors are New York Times White House correspondents.

It opens with an extended methodology statement rather than with narrative. A reader who has written about that opening describes it as setting out the categories of people interviewed, how direct quotes were arrived at, where background was used and why some passages are paraphrased. Behind the book sit more than 1,000 interviews conducted on the ground rule of deep background, and its subject was interviewed in person in March 2026. The introduction states that direct quotes come from people who were in the room, or from contemporaneous notes, recordings or transcripts. Haberman has said the pair jettisoned thousands of pages of reporting and that the manuscript would otherwise have run to about 2,000 pages.

Two things worth flagging because they are not documented rather than because they are absent. Chapter count and typical chapter length appear nowhere we checked, and neither does whether the authors write in one continuous voice or take sections in turn.

The case for reading it

Three arguments, and they stack. The first is the sheer number of named people. A single meeting reconstructed in one review is attended by nine named officials, of whom two are quoted directly, and that is one scene in a 496-page book drawing on more than a thousand interviews. No one has counted the named figures, so we will not put a number on it, but density per scene is on the record.

The second is that there is nothing to hold on to when a name recurs. The vocabulary is institutional, meaning departments, agencies, military and legal process, and no glossary, chronology or cast list is mentioned on the Simon & Schuster pages, the Waterstones or Bookshop.org listings, the Kirkus review or the trade entry. Absence from a listing is not proof of absence in the book, and this is the first thing to settle against a physical copy.

The third is the back matter. One reader reports over sixty pages of notes, sources and references, plus an index. The notes figure is plausible for a reported book this length; the same reader's account of how long the index runs is not, and we are not treating either as settled. Whether the citations sit as footnotes on the page or as endnotes at the back is not stated by any source, so we make no claim about that. What is not in doubt is that the four part titles are printed dividers, and that a reader can flip back to the attribution rules at the front at any point.

The case for listening to it

The structure is the argument for audio. This is a book to go through in order rather than one to consult, and audio handles that shape well: a broadly linear year, four parts, no exercises or checklists, and one Audible listener review opening on how well structured it is and how much event timeline it covers. Seventeen hours is a long sitting, though at one and a half times speed it comes down to roughly eleven and a half, which is comfortably a week of commutes. Nothing found says whether people read it cover to cover or use it as reference, so the shape is doing the work here rather than any documented reading habit.

There is a practical argument alongside it. This is a 496-page reported book with sixty-odd pages of apparatus behind it, and the format that actually gets finished is worth something. One narrator straight through is also the least demanding audio arrangement there is, with no cast to distinguish and no switching between performers. What the recording does not do is intervene in the text in any way that is documented, so the listen side rests on this single point rather than on the production.

The audiobook in detail

Simon & Schuster Audio put it out on 23 June 2026, the same day as print, at 17 hours 6 minutes on the Simon & Schuster UK figure. Apple Books and audiobooks.com list 17 hours 7 minutes and Kobo US lists 17.1 hours, so allow a minute either way. At one and a half times speed that works out at about 11 hours 24 minutes, which is our arithmetic rather than a published figure. Robert Petkoff reads it, and he is the only performer named on Audible US, Simon & Schuster UK, Apple Books, Kobo, Chirp and audiobooks.com. One narrator, unabridged. Neither author records any part, including the introduction. Only one recording turned up, with the same narrator and near-identical run times on both sides of the Atlantic, although no source states that the territories share it. No audio award for this recording was found as at 11 August 2026, and no supplementary PDF is advertised on any retailer page we checked, so there is no documented route from the recording back to the printed apparatus.

Who should ignore this verdict

Anyone who already follows this beat closely, because the naming load only bites when the names are new, and if the offices and the officials are familiar the audio loses most of its disadvantage. Anyone who will listen at speed rather than read at all, since seventeen hours got through is worth more than 496 pages not started. And anyone who wants the reporting for the story it tells rather than for the sourcing behind it, because the notes and the index are the part that stays on the page. One more group: anyone buying it to check what it cites should hold off on either format until the back matter is confirmed in a physical copy, since the shape of the notes is the single least settled thing about this book.

Get Regime Change
Both editions, so you can go with the verdict or ignore it entirely.
Affiliate links. If you buy through one of these we may earn a small commission, at no extra cost to you, and with no influence on the verdict above.

Related verdicts

Questions people ask

Who narrates the Regime Change audiobook?

Robert Petkoff reads the whole book. He is the only performer credited on every listing checked, including Audible US, Simon & Schuster UK, Apple Books, Kobo, Chirp and audiobooks.com. It is one narrator, unabridged.

Do Maggie Haberman and Jonathan Swan read the audiobook themselves?

No. No listing credits either author with any recorded portion, including the introduction. Both are New York Times White House correspondents and the book is built on their own reporting, but the recording is handed entirely to a professional narrator.

How long is the Regime Change audiobook?

17 hours 6 minutes at Simon & Schuster UK, with Apple Books and audiobooks.com listing 17 hours 7 minutes and Kobo US listing 17.1 hours. That is roughly 11 hours 24 minutes at one and a half times speed, a figure worked out from the published run time rather than published anywhere.

Does the Regime Change audiobook include the notes and the index?

No listing checked says what Simon & Schuster Audio did with the back matter, and no supplementary PDF is advertised on any retailer page we looked at. One reader reports over sixty pages of notes, sources and references in the print edition plus an index, and that count is uncorroborated.

How this verdict was reached

Every book on this site is checked against the same ten questions, set in advance and applied identically to every title. They are based on research into what readers and listeners generally find easier to follow in each format, not on personal preference, and they are published in full so you can check the reasoning yourself. Verdicts are never paid for, sponsored or influenced by affiliate partners.

  • The criteria: read the ten questions in full.
  • What we checked for this book: the narrator credit across six retailers and both territorial run times, the publisher listings for page count and format, the reported basis and sourcing rules, and what is and is not documented about the back matter.
  • Verdict last reviewed: 11 August 2026. Editions and narrators change, so verdicts are rechecked annually.
  • Prefer different criteria? The app lets you change which of the ten questions matter to you.

Deciding about a different book?

Search the site to see if it has been written up.

Not on the site? Download the free app, which can tell you whether any book is better to read or listen to, and lets you turn off any of the ten checks that do not match how you read.

Download the free app