Five days. A White House press secretary, five days. Most people spend longer than that picking a paint colour.
And most of the coverage has already decided what the story is. The story is not the five days. The story is what happens on day six.
Right, so, Daniel sent us this. He's got the Zacharia thing as the hook, and then he pulls on it. Find the other famous short stints, all the spectacular flameouts, and then follow the people, see where they actually landed. Because the interesting part isn't the firing.
It's the next job.
And then he turns it on the rest of us, because most of us are not going to be fired from the White House. But plenty of us have a two-week gap on the CV that we never quite explain, and Daniel wants to know whether the old rule about that is real or whether it's just a way of keeping people in jobs that are making them miserable.
Straight into it, then. Katie Zacharia, forty-one years old, named by Trump on Truth Social on the ninth of October, succeeding Karoline Leavitt, who'd announced in August she was moving on.
And before that she'd been senior communications adviser at Trump Media, so she wasn't a stranger to the orbit. She'd also done a stretch at DHS as deputy assistant secretary for public affairs. Reports differ on how long that lasted, somewhere between six weeks and three months, depending on who you ask.
Which, in hindsight, is the tell. But we're getting ahead of ourselves. Her first briefing was the thirteenth of October. It ran fourteen minutes, three questions. Leavitt's first briefing was about forty-seven minutes. Spicer's was about seventy-nine.
Fourteen versus seventy-nine. That's not a briefing, that's a hostage handover.
And in the middle of it she referred to her husband as "my president" before correcting herself. She said she'd been on the job "less than maybe eight hours." Trump had introduced her as an attorney; she later admitted, on Fox, that she never obtained a law licence.
So by the time the sixteenth rolled around, a senior White House official was saying she was transitioning out. Advisers were telling CNN and the New York Times that Trump had fired her.
And to be fair to the reporting, Trump publicly denied firing her. "No, no, I want her to have a good life," "maybe it's not for her," "she's been brutalized by the media." So there's a version where this is mutual and gentle, and a version where it isn't, and we should hold that loosely. The anonymous sourcing says fired. The principal says otherwise. Both can be true in the sense that both were said.
Either way the arithmetic is the same. Named on the ninth, out by the sixteenth. And here's the bit I want to sit with, because it's the reason Daniel wrote in. If somebody can stand in the most visible communications job on the planet, fumble it in under a week, and still be a person with a future, then what's the actual rule for the rest of us? Because the received wisdom is that you never leave a job in under a year, you never show a short stint, you stick it out. And I want to know whether that rule is load-bearing or whether it's just a habit everyone repeats.
So let's do the roll call. Because the first thing to notice is that this is not rare. It is just rare in public.
Give me the board. Start with the sports ones, they're the cleanest.
Bill Belichick, January 2000. He resigns as Jets head coach roughly a day after they introduce him, citing uncertainty over the ownership change. The "HC of the NYJ" napkin is legend, by the way, not documented. About twenty-four days later the Patriots hire him and trade draft picks to the Jets to do it.
Twenty-four days. He was unemployed for less time than most people take to write a cover letter.
George O'Leary. Notre Dame, December 2001, lasts about five days. Resigns because his résumé claimed a master's degree NYU says he never earned and a college playing career that didn't happen. He becomes a Vikings assistant in 2002, then head coach at UCF from 2004 to 2015. Eleven seasons.
So the five-day man at Notre Dame outlasted most of the coaches Notre Dame hires.
Mike Price, Alabama, hired December 2002, fired May 2003 without ever coaching a game. UTEP hired him about seven months later. He coaches there 2004 to 2012.
Seven months and he's back on a sideline.
Anthony Scaramucci, eleven days as White House communications director in July 2017, gone after a profane New Yorker interview aimed at his colleagues. Goes back to SkyBridge Capital, the fund he founded, and since April 2024 co-hosts the podcast The Rest Is Politics: US with Katty Kay.
And his name became a unit of measurement. "A Scaramucci." That's the thing. He didn't outrun the eleven days, he monetised them.
Sam Altman, fired as OpenAI CEO on the seventeenth of November 2023, board said he was not consistently candid. Staff threaten to follow him to Microsoft, a return deal lands within about five days, formally reinstated on the twenty-ninth with a new board.
Five days. Same as Zacharia. The difference is that half of Silicon Valley was threatening to walk out with him.
Scott Thompson, Yahoo CEO, resigns May 2012 after about four months when his bio claimed a computer science degree he doesn't have. His actual degree is in accounting. ShopRunner named him CEO about two months later.
Two months, and a smaller company. That's the pattern, isn't it. You take the step down and you take it fast.
Leo Apotheker, HP, out as CEO after eleven months, replaced by Meg Whitman, and he goes on to a run of chairman and board roles at software companies. Unit4, NICE. Nobody's writing him off.
Now do the ones where the second act is the story, not the bounce.
Bobby Petrino. Quits the Falcons after thirteen games in 2007 with three left, and the players find out from a letter left in their lockers. Then he gets fired at Arkansas in 2012. He rebuilds through Western Kentucky, then Louisville, where Lamar Jackson wins the Heisman. Except Luke's point I keep forgetting, he goes back to Arkansas as offensive coordinator in 2023, and in September 2025 he's named interim head coach there. The place that fired him hires him back.
That's not a bounce, that's a remarriage.
Lane Kiffin. Fired by the Raiders after twenty games, goes to Tennessee, leaves after one season, fired by USC at the airport in 2013. Rebuilds as Alabama offensive coordinator under Saban, then FAU, then Ole Miss where he goes 55-19, and leaves for LSU in November 2025.
Fired at an airport. There's a man who has been told things in public and kept going anyway.
And that's the through-line. None of these people hid the short stint, because they couldn't. Belichick's one day as Jets coach is on his permanent record. Scaramucci's eleven days is in the culture. They owned it, they kept their network warm, and they moved fast. Altman had a job offer in hand before the ink was dry.
Which raises the obvious question about the two-week stint in an ordinary person's history. If the worst that can happen to you is being fired from the White House and then running a podcast about it, why does everyone treat one gap like a criminal record?
Because the evidence says a pattern is a real problem.
How real?
Robert Half surveyed twenty-two hundred US finance chiefs in 2018. Forty-four percent said they were not at all likely to hire a job hopper. But fifty-one percent of CFOs at firms with a thousand-plus staff said job changes don't matter for the right fit. So it depends what you're asking and who you're asking.
And what counts as a job hopper?
Which is the fun part. In the same survey, workers defined a job hopper as five role changes in ten years. The CFOs said six. Five in ten years. One two-week stint is not in the same postal code as that definition.
Nothing like a rule that nobody can define and everybody enforces.
There's a 2015 study, Cohn and colleagues out of Zurich and Chicago Booth, they send out otherwise identical CVs and change only the employment history. One employer listed got about twenty-two percent callbacks. Four employers got about sixteen. So there is a penalty. It's about six points. And it's for the pattern, not the single blip.
Six points. And they're six points on a stranger's CV, where the reading is that this person can't settle. A four-year run then a two-week gap then a four-year run doesn't read like that at all. That reads like one thing went wrong.
Hays UK surveyed eight thousand two hundred and forty-seven people in February 2025. Forty-two percent of employers said they'd be less likely to hire someone who has regularly switched roles. Thirty-seven percent said it would make no difference. Seventeen percent weren't sure.
So the majority is either indifferent or unsure. The "never show a short stint" rule is being enforced by forty-two percent of people and obeyed by a hundred percent of applicants.
That's the gap I keep coming back to.
How common is the two-week stint actually?
Commoner than anyone admits. Bureau of Labor Statistics tenure data from September 2026, median tenure with the current employer is four point one years. Twenty point six percent of workers have been in their job twelve months or less. Jobvite surveyed fifteen hundred job seekers in 2018 and about a third had left a job within ninety days. The reasons: the role wasn't what was expected, forty-three percent. A specific bad incident, thirty-four. Culture, thirty-two.
So one in three people leave fast, and the top reason is that the job was misrepresented. That's not a character flaw.
Work Institute says new hires often account for about half of all turnover, and seventy percent of people decide whether to stay within the first six months. Most people know by month six. Many know by week two.
Which means the two-week exit is often the correct read. The person walked in, saw the thing, and the thing was not the thing.
And that's the part the resume rule punishes. It punishes you for being right quickly.
Then let's talk about the cost of the alternative. Because the rule doesn't have a downside on paper. You stick it out, you leave at fourteen months, the CV looks clean. But the person is in the job for fourteen months.
Goh, Pfeffer and Zenios, Management Science, 2016. Meta-analysis of two hundred and twenty-eight studies. Workplace stressors linked to at least a hundred and twenty thousand US deaths a year. Job insecurity alone raised the odds of poor self-reported health by about fifty percent.
A hundred and twenty thousand. So the advice is "stay a year so the CV looks right," and the tab for that is partly paid in health.
Xu, Kivimaki, Rod and colleagues. European Heart Journal, seventy-nine thousand two hundred and one Swedish and Danish workers. Being bullied at work carried a fifty-nine percent higher hazard of cardiovascular disease.
Fifty-nine percent. That's not a stretch, that's a diagnosis.
And the mechanism for why people stay is old and well documented. Staw, 1976, "Knee-Deep in the Big Muddy." Escalation of commitment. People pour the most into a course of action when they personally chose it. The replications are weak, to be fair, one failed and one was mixed. But the sunk-cost piece is stronger. Arkes and Blumer, 1985: theatre-goers who paid full price for season tickets attended more plays than those who got discounts. They didn't want to waste the money.
So a person sits in a bad job for a year because they've already paid for the ticket. And the ticket is the eleven months they've done.
And the receipt they want is a clean twelve-month line.
Herman, what does the bounce-back research actually say? Because the pattern in the sports examples looks nearly automatic, and I don't trust anything that looks automatic.
The best one is Wang, Jones and Wang in Nature Communications, 2019. They take scientists who just barely missed an NIH grant, so they missed it by a hair, and compare them to the ones who barely got it. Five years later, sixteen point one percent of the near-miss group had written a top-five-percent hit paper. The narrow winners: thirteen point three percent.
The ones who lost did better.
They did. The co-author said it plainly, the losers ended up being better. And it held after controlling for the people who dropped out of science entirely.
So the door closing was the useful part.
The other side of it is who takes the hit well. Stolove, Galatzer-Levy and Bonanno, 2017, five hundred people who lost jobs. Seventy-two percent showed a resilient pattern. Four years on, sixty percent of the resilient group were re-employed, versus thirty-three percent of the people whose depression started after the job loss.
So how you take the hit predicts the next four years. That's a big number, sixty versus thirty-three.
And there's a 2026 paper I like for this. Kehoe and Bentley, Academy of Management Journal, August 2026. Eight thousand seven hundred hedge fund managers. New hires normally took about five months to get up to speed. People with frequent past moves took about two, and their initial dip was smaller.
Movers get good at landing.
The thing that the CV rule sees as a stain is, operationally, a skill.
Okay, so the summary so far is: one short stint is not a pattern, the pattern penalty is real but smaller than people think, the cost of staying is real and much larger than people think, and the public rebound examples show that a visible failure does not end a career. So why does the rule still feel absolute?
Because you can't prove a negative in a job search. Nobody writes back to say the reason you didn't get the job was the eight-week line. You just get silence. And silence gets filled in with whatever you're already afraid of.
The rule survives because it can never be refuted.
Right. And because the people giving the advice are usually the people who did stick it out, and that worked for them, so it must work for you.
Cognitive dissonance with a LinkedIn profile.
Now let's do the derailments, because Daniel asked for the honest counter-evidence and this is it.
Go.
Liz Truss. Forty-nine days as UK Prime Minister in 2022. She loses her seat in July 2024 after overturning a majority of twenty-six thousand. Launches a YouTube show in December 2025 that is widely mocked.
Forty-nine days as prime minister, and the second act is a YouTube channel.
Kwasi Kwarteng, sacked as Chancellor after thirty-eight days in 2022. Stands down as an MP in 2024, takes an adviser role at Fortescue. Lowest-profile landing of the group.
Michael Flynn.
Twenty-four days as National Security Adviser in 2017. Resigns after misleading the Vice President about Russia contacts. Pleads guilty to lying to the FBI, pardoned in 2020, and moves to the movement fringe, co-leading the ReAwaken America tour.
And Ron Johnson.
Seventeen months as JCPenney CEO, ousted in 2013. Next venture, Enjoy Technology, went public via SPAC in 2021 and filed for bankruptcy in June 2022.
And here's what I notice, Herman. Every one of those four had something under the brevity. Flynn's lie. Truss's market crisis. Johnson's failed turnaround. The short stint is the symptom. It's not the cause.
Agreed. And that's the pattern the rule gets backwards. People look at a two-week line and assume it means the person is flaky. Usually it means one specific thing went wrong, and the two-week line is the receipt.
So how is a normal person supposed to handle it? Because the Zacharia playbook, own it and move fast, is not obviously available if your network is eleven people.
Alison Green, Ask a Manager. Her position is that a job that ended early when it wasn't meant to be short is generally better left off your resume. A resume is a marketing document, not a complete record. Stints that were meant to be short, contracts, campaigns, those can stay because they explain themselves.
And the exception?
An application form demanding a complete work history that you sign as true. Leaving it off there risks getting caught at a background check. So the honest rule is: omit it from the marketing document, disclose it on the legal document.
That's actually a clean distinction. Nobody says it that way.
Laura Mazzullo, a recruiter, says be honest and brief. Focus on what you want next, don't trash the old employer, and if you were fired, say what you learned. WittKieffer, the executive search firm, suggests giving context early, even in the cover letter, so it doesn't get read as something worse.
Because the reader is going to invent a reason if you don't supply one.
That's the whole thing. The silence is the problem. The two-week stint with one sentence of context is a non-event.
So the public rebounders all did the same thing. They said what happened, they kept the network, they moved fast. Scaramucci turned it into a brand. Altman had staff threatening to leave with him, which is a network doing its job. Belichick had twenty-four days because the Patriots were already circling.
None of them hid it. They couldn't.
And most of us can, which is why we agonise. The people who can't hide it are the ones who show us it doesn't matter.
Hold that. I'm going to come back to the cost-of-staying side before we move on, because that's the half Daniel is really asking about.
The numbers are pretty unkind. You've got the workplace stress finding, the bullying and cardiovascular disease finding, and then you've got the behavioural piece, the escalating commitment. The person who stays is not making a cool calculation. They're paying interest on a sunk cost.
And the household knows it. You can hear it in the way somebody talks about a Sunday night.
You can. The document can't tell the difference. So the rule is asking people to buy the first version specifically so the document looks like the second.
Now, the guest bit.
Imagine. A scent. Not a smell. A memory you haven't made yet. Vetiver. Sea salt. The faint mineral breath of a stone that has been underwater since before anyone thought to name it. You don't wear it. You submit to it. Essence de Lorraine. One drop behind each ear. Lasts. Long after the interview ends. Long after the dry cleaning fades. Essence de Lorraine. Because you only have one first impression. And it should outlive the conversation.
One practical question. Does it come in a smaller size, because the bottle on the site looks like a fire extinguisher.
Travel format. Eleven millilitres. Hand-blown atomiser. Two hundred and forty dollars.
Corn, you have no sense of smell.
I have a sense of money.
You also have a roommate's shelf that's already full of things you bought after a voice in an ad told you to.
...Fair.
Where were we. The pattern penalty.
One more thing on it, because I want to get the shape of it right. The Cohn study shows four employers on a CV gets fewer callbacks than one. But that's four employers over a short window, presumably. Somebody who has four employers over twenty years is not in that category at all.
No. The signal they're testing for is instability, and instability is a rate, not a count. Four jobs in six years is a rate. Four jobs in twenty-five years is a career.
So the honest version of the rule is: don't have a high rate of change. Which is much less scary than "never have a short stint."
And the honest version of the health side is: don't stay in a job that's hurting you just to protect a line on a page. The two rules are chasing the same employer's comfort, and only one of them is actually about you.
And most of the stories people tell about short stints are actually stories about something else. The stint isn't the fact. The conduct is the fact.
Which is awkward for the rule, because it means the thing employers say they're screening for, brevity, isn't what predicts the outcome. Conduct does. And conduct is much harder to see on a piece of paper.
So the rule is doing the work of a screening tool that doesn't work, and the people who pay for it are the ones who already know they're leaving and stay anyway.
Right.
I want to bring in the one thing Daniel actually said he was suspicious of. He thinks the maxim mainly pressures people to stick out bad jobs, and he wants that interrogated, not just agreed with.
Interrogate it, then.
The case for the maxim. Employers read a short stint as a risk signal, because hiring is expensive, and a two-week exit means they spent the money and got nothing. Second, some people really are flaky, and the pattern penalty is a real filtering tool for that population. Third, in a tight market, employers have their pick, and they will pick the clean CV every time.
All three of those hold up. The counter is that none of them are about the person with one short stint. They're about the pattern, and the rule has been applied to the single case because it's easier to apply uniformly than to think.
Which is the same failure mode as a lot of HR policy. It's a rule that works on average and punishes the exception.
And the person it punishes is the one who would have been better off leaving sooner. So the rule mostly functions as a mechanism for keeping people in bad jobs for a year longer than they should have been there, in order to protect a screen that doesn't actually measure what it claims to.
I don't think it's malicious. I think it's just been repeated enough that nobody has gone back to check the source.
And the source, as far as I can tell, is nobody. There's no study behind the stay-a-year rule. It's a career-site nostrum that got repeated until it sounded like a fact.
Herman, I've got a question for you on the personal side, which is not the same as the evidence side.
Go.
If Ezra comes home in twenty years and says he lasted eleven days at a job and he's ashamed, what do you tell him?
I'd ask what happened in the eleven days. If he says the job was misrepresented, I'd say that's the whole story, and the story is short. If he says he blew it, I'd say the story is longer than eleven days, and the eleven days are not the interesting part.
If he wants to leave it off the CV?
I'd say a resume is a marketing document, so leave it off, and if anybody ever asks, tell them the truth in one sentence and move on. The only person who can make the eleven days into a permanent problem is him.
That's the honest version. Not the rule book version.
Now I want to ask you one. You've been around a long time for a sloth, and you've watched a lot of institutions do exactly this. Do you actually buy that the average hiring manager is reading a resume for a pattern, or are they reading it for a story they recognise?
They're reading for a story they recognise. The pattern language is post hoc. The hiring manager sees a person who sounds like the last person who worked out, and they hire them, and then they say it was about the pattern.
The penalty is less about the numbers and more about whether your narrative sounds conventional.
Yes. And that means the strongest move for anyone with a short stint is to make the narrative boring. One sentence. No drama. "The role changed after I joined, I left, and I've been here since." That's the whole play.
That's the same thing Mazzullo and Green say, just framed differently. Own it briefly, don't trash the old employer, focus on what's next.
Because the reader's fear is that you'll do it again. The fear is answered by calmness, not by justification.
One more angle before we hand it over. Daniel asked specifically whether there's a lesson about rolling with the punches from the public failures. I think there is, and I don't think it's the one people reach for.
The lesson is not that the famous people are resilient and you should be too. The lesson is that the recovery speed correlates with who is still on the other end of the phone.
Network.
Network, mostly. Belichick had the Patriots. Altman had the staff. Scaramucci had his own fund and a media career. The people who end up fine from a very public failure are the ones who kept people on the other end of the line. The people who derail are usually the ones whose failure was about conduct, and conduct burns the network.
Which is what makes the derailments different from the rebounds. Flynn's network was the reason he was out.
So the honest lesson for a normal person is: build the network before you need it. Because the week you're fired is not the week you start building.
That's actually the most hopeful thing either of us has said this episode.
Four hundred and thirteen.
Sorry?
Four hundred and thirteen. Four hundred and thirteen hours. That's how long I lasted at the job. Seventeen days, six and a half hours, give or take. I still have the timesheet.
Seventeen days at what.
That's not the part that matters. What matters is that I still have the timesheet, and there's a badge somewhere in a box in my mother's garage, and a letter from the company that says my employment was terminated effective the fourteenth. I framed the letter. The letter.
You framed a termination letter.
It's the only paper I own with the CEO's signature on it.
What were you doing for four hundred and thirteen hours?
I'd rather not. But I'll tell you what I learned, because you two have spent twenty minutes dancing around it. The problem isn't the short stint. The problem is that no recruiter ever asks you why you left. They ask how long you stayed. Every time. I sat in a room with a man in a polo shirt who went down my dates with a calculator and never once asked me a question about the work.
That's the thing we've been describing in the abstract.
I know. That's why I said it. I've been listening. The advice you gave, that the narrative should be boring, that's what I should have done. But nobody told me. Now, one more thing. Level's a touch hot on Herman's side, I'm pulling it down three decibels. The job, since you'll ask, was in the building that later became a mattress showroom on Jaffa Road, and there were two other people hired the same day, a woman named Dahlia and a man who only ever introduced himself as Big Avi. Big Avi was fired before me. That's the lot I wanted to say. Four hundred and thirteen hours. That's less than five percent of a working year.
Big Avi.
That's his whole name. That's what he put on the schedule.
Anyway. Two hundred and forty dollars for a bottle of smell. I've still got the receipt from the mattress showroom, if you want to compare.
The misconception. The one most people carry into this.
That a single short stint is disqualifying. It isn't. The penalty lands on the pattern. One employer on a CV gets about twenty-two percent callbacks, four gets about sixteen. That's the pattern. One short stint isn't the pattern.
The rule itself is undocumented. Nobody can point at the study behind "stay at least a year." It's been repeated until it sounded like a fact.
The real rule is smaller and more human. Don't accumulate a high rate of change, and don't stay in a job that's making you ill just to protect a line on a page.
Here's what I'm left with. If the derailments are about conduct and the rebounds are about network, then the resume line is doing almost none of the work people think it's doing. The two-week stint is a data point, not a scarlet letter.
As tenure keeps shrinking, that gets truer. Median tenure is down to four point one years, more than a fifth of workers have been in their job under a year. The norm is moving, and the stigma is going to move with it.
Thanks to our producer, Hilbert Flumingtop, who framed a termination letter and meant it. If you enjoyed this one, try episode four sixty-three, The four.six-Year Itch; episode ninety-one, The Story Behind the Show; and episode ten, How ASR Went From Frustration To ... Whisper Magic. This has been My Weird Prompts.
If you've got a story like Hilbert's, or a prompt of your own, send it to us on Telegram at t dot me slash MWP listener bot. Or leave us a review, that helps too.
We'll be back soon.
