Masters in Business podcast

Barry Ritholtz returned to the Rational Reminder podcast for episode 421, roughly seven years after Benjamin Felix and Cameron Passmore hauled backpacks full of recording gear to his New York office for episode 57. The occasion is his book How Not to Invest, and the framing of the whole conversation is that avoiding a short list of expensive errors matters far more than finding brilliant ideas. Over roughly a hundred minutes it covers why billionaires are terrible forecasters, what the failure gap research says about how badly humans estimate their own odds, why financial media is structurally incompatible with a thirty year time horizon, how concentrated positions destroy retirements, and a closing stretch on cars, spending, and decumulation that is the best part of the episode. You can watch the full conversation here.

TLDW

Ritholtz defines investing as the art of using imperfect information to make probabilistic assessments about an inherently unknowable world, and builds everything else on that. He explains the halo effect that makes billionaire forecasts feel authoritative, offers the three Ds of doubt, depth, and Dunning-Kruger as a filter for spotting bad advice, and argues that experts are excellent at explaining what is happening and useless at predicting what happens next. He makes the case that financial news is not merely useless to long term investors but actively negative, since it biases you toward action when fewer decisions produce better outcomes, and points to the IRS having to publish a 42-point debunking of TikTok tax advice as evidence of how bad short-form has gotten. He walks through three forms of economic innumeracy: denominator blindness, survivorship bias in its modern form as the failure gap, where research across more than thirty life domains found failure occurs about 61% of the time while people estimate 41%, and our inability to intuit compounding. He covers secular cycles and multiple expansion, why valuations predict returns but not timing, why externalities cause markets to wobble and then resume, and why the COVID crash confused everyone: the sectors visibly dying were roughly 6% of the S&P 500 while the index rallied 69% off the March lows, an availability heuristic error he says people are repeating now with AI. He explains why indexing works using Hendrik Bessembinder’s finding that 1% to 2% of stocks drive essentially all equity value, catalogs the big behavioral mistakes including concentration, fees, taxes, and a lack of humility, and gets specific about Jack Welch, General Electric ESOPs, and Cisco taking 25 years to break even. He gives his prescription: a plan, dollar cost averaging, pay yourself first, a broad index core, and a 3% to 5% cowboy account to keep your lizard brain away from the real portfolio. He is skeptical of private credit reaching 401(k)s, thinks young investors should be all equity, and closes with an emphatic argument that you should buy a new car for the safety technology, that comparison is the thief of joy, and that the hardest problem his wealthy clients have is spending the money they spent a lifetime accumulating.

Thoughts

The definition Ritholtz gives around the twenty three minute mark is the load bearing idea of the entire episode, and it is worth sitting with because most people who repeat it have not actually unpacked it. Investing is the art of using imperfect information to make probabilistic assessments about an inherently unknowable world. Every word is doing work. Art, because it is not a formula, and he tips his hat to the quants while noting their math is not perfect either. Imperfect, meaning incomplete, sometimes inaccurate, sometimes stripped of context. Probabilistic, meaning the output is a distribution rather than a direction. And unknowable, which is the part people resist hardest. His three Ds framework earlier in the episode, doubt, depth, and Dunning-Kruger, is the practical version of the same claim: the tell of bad advice is not that it is wrong, it is that it is confident and specific. What makes the framework more than a slogan is the defensive driving metaphor he reaches for at twenty eight minutes. Every high performance driving school he has attended, and he has attended most of them, is a barely disguised defensive driving class, because the actual skill being taught is knowing the limits of your own ability and the performance envelope of the machine you are operating. Body-on-frame SUVs roll over at high rates because someone asked the vehicle to do something it could not do. That is a better model of investor blowups than any behavioral finance taxonomy, and it explains why his advice is almost entirely subtractive.

The media argument is the one where he is most obviously testifying against his own industry, and that is what makes it credible. Ritholtz hosts a Bloomberg podcast, writes a long-running blog, and is on the show as an author promoting a book, and his position is that for a long term investor the news is not neutral but negative, because the media urges you to action and history says the fewer decisions you make the better you do. He supports it with Hendrik Bessembinder’s work showing that if the S&P 500 had simply been left alone from inception, with no rebalances, additions, or deletions, it would have significantly outperformed the actively maintained version. That is a quietly radical finding, since the index most people treat as the passive benchmark turns out to be an actively managed portfolio that underperforms doing nothing. His prescriptions follow logically: lengthen the time horizon of your media consumption to match the time horizon of your money, build your own vetted list of trusted voices rather than outsourcing your thinking to an institution’s masthead, and read books. The line about a book being 2,000 hours and thirty years of experience for $21.95 is the kind of thing that sounds like an author selling books until you notice he then refuses to give out his own list of trusted sources on the grounds that you have to build your own. His observation that mass literacy is only a few centuries old, so declining book reading is closer to mean reversion than to civilizational decline, is bleakly funny and probably correct.

The failure gap research he cites at thirty eight minutes is the most quietly devastating thing in the episode and it barely gets a follow-up question. Researchers reviewed more than thirty life domains and found that failure occurs about 61% of the time, while surveyed people estimated 41%. That twenty point gap is not a rounding error in intuition, it is a systematic distortion in how humans model risk, and it extends far past investing into starting businesses, changing careers, and any decision where you only ever see the survivors. His example of the Wall Street Journal running features on which classic cars appreciated most over fifty years is exactly right and exactly infuriating: thanks for telling me who won the game after the game ended, now tell me which cars to buy today, and of course they never do, because that is the hard version of the question. The other two innumeracies he names are the familiar denominator blindness, illustrated by Shark Week and the observation that humans kill tens of millions of sharks a year while sharks kill three or four of us, and our failure to intuit compounding, which he admits still catches him. His framing of why this matters is the most quotable sentence in the whole hundred minutes: you do not have to be brilliant to succeed as an investor, we all just have to be less stupid. That is a genuinely different bar than the industry sells, and it is the entire thesis of the book compressed into one line.

The most useful transferable model in the episode is his explanation of why the COVID market made no sense to people, and the fact that he applies it twice is what makes it valuable rather than merely clever. In 2020 everyone looked around at closing restaurants, empty hotels, and grounded airlines and concluded the market was unhinged from reality. He ran the numbers: all of those visibly dying sectors summed to roughly 6% of the S&P 500, while the index rallied about 69% off the March lows. The error was not optimism or manipulation, it was the availability heuristic colliding with market cap weighting. Your lived experience is unweighted and local. The index is weighted and global. Those two things can move in opposite directions for years without either being wrong. Then he applies the identical model to the present: how can markets rise if AI is going to take all our jobs? Because your income and the index are different objects, and the relevant question is which of the other 493 companies get more productive and more profitable because of AI, not which of the seven build it. He is not dismissing job displacement, he is separating two questions that get constantly fused in commentary. This is also where his bubble answer lands, and it is refreshingly empirical rather than rhetorical. Asked constantly on client calls whether AI is a bubble, he went to the data and found the four largest companies today trade at multiples nothing like 2000, with Nvidia’s price to earnings ratio back near where it sat in 2019. That is not a claim that nothing is expensive. It is a demonstration that the analogy people reach for first does not survive contact with the numbers.

The final quarter is where Ritholtz breaks with the genre he writes in, and it is the most interesting stretch of the episode. Asked whether people should buy new cars, a question the personal finance orthodoxy answers reflexively with no, he says absolutely yes, and the argument is not about enjoyment. It is that a twenty year old car has airbags whose actuators probably will not fire, no collision avoidance, no blind spot detection, and no seatbelt tensioners, and that treating this as a consumption decision rather than a safety decision is a category error the spending scolds trained us into. He has earned the right to the argument the hard way, having been T-boned in a 2017 Panamera that was totaled while he and his wife walked away with a chipped tooth. The Lee Cooperman story is the sharper version: a billionaire driving a twenty five year old Passat because the money is all going to charity and he does not want to spend the charity’s money, until Ritholtz points out that staying alive longer is how he keeps generating market-beating returns for those philanthropies. A week later Cooperman bought a Lexus. The same logic applied to Kawhi Leonard, whose knees and wrists are his capital asset, makes the point unmistakable. This connects directly to his closing answer about success, which he defines as freedom, opportunity, and optionality, and to what he says is the hardest problem his firm has with clients who have already won: getting them to spend. A lifetime of accumulation habits does not reverse on command, and the pivot to decumulation is genuinely difficult. Combined with his line that comparison is the thief of joy and his suggestion that giving the inheritance while you are alive to watch it enjoyed beats the alternative, it amounts to an argument that the last mile of personal finance is not a math problem at all. Most of the industry, including most of this episode, is about accumulation. The part almost nobody addresses well is what he is describing here.

Key Takeaways

Billionaires making economic forecasts on television do the same thing everyone does, which is talk their book, plus something worse: they are usually unqualified to make the forecast at all.
The halo effect is what makes those forecasts dangerous. Extreme success in one domain gets projected onto every other domain, and Ritholtz’s blunt correction is that they are not all that successful elsewhere.
His formula for forecasting outcomes is skill plus outside events plus random luck, which places skill as a small component of a much larger equation.
The film industry is his favorite illustration, because greenlighting a movie is a prediction about where public taste will be three to five years from now, and taste moves faster than production schedules.
Experts are genuinely valuable for explaining what is happening and why, using data, history, and knowledge of every player’s track record. They are not valuable for prediction.
His cardiologist analogy makes the distinction concrete: a specialist can tell you exactly how healthy your heart is, and cannot tell you the date of your heart attack. Ask your advisor how your portfolio looks, not when the crash comes.
The most obvious tell of bad advice is an appeal to emotion, whether fear, greed, or FOMO, and any manufactured sense of urgency is a giant red flag rather than an insight.
Good advice sounds humble and probabilistic. Bad advice sounds self-confident and specific, which is unfortunately also more attention-grabbing.
The three Ds filter: doubt (people with no self-doubt are unaware of their blind spots), depth (a broad well of experience inside a repeatable process), and Dunning-Kruger (awareness of the limits of your own skill).
The most expensive sentence a novice investor can utter is “how hard can it be?”
Social media algorithms structurally reward the loudest and most extreme voices, so the selection pressure runs against exactly the qualities that mark good advice.
The news media is the daily beast that must be fed, with column inches and 24/7 broadcast hours to fill, and that clock is chronologically incompatible with letting a portfolio compound for decades.
For a long term investor, financial news is not merely useless but negative, because it feeds a bias toward action and the data says fewer decisions produce better outcomes.
Bessembinder’s research on S&P 500 rebalances, additions, and deletions found that leaving the index untouched from inception would have significantly outperformed the version that gets actively maintained.
The IRS had to publish a 42-point press release debunking TikTok and Instagram tax advice, some of which would cost you penalties, some a clawback with interest, and some of which could land you in jail.
Taking financial advice from an anonymous account is the financial equivalent of taking candy from a stranger, and Ritholtz’s phrase for the result is getting financially roofied.
Gell-Mann amnesia, from Michael Crichton, is reading an article about something you know well, recognizing it is completely wrong, then turning the page and believing the next one.
His remedy is to build your own vetted list of trusted voices over years, and he pointedly refuses to sell you his list, because outsourcing your thinking to a third party is the underlying problem.
Lengthen the time horizon of your media diet to match the time horizon of your money. Ask what anyone tweeting in 2001 could have said in a short burst that would matter today, and the answer is nothing.
Books are the great bargain of the information age: a thousand or two thousand hours plus a lifetime of experience for about $22, versus paying him a large fee for forty minutes and a Q&A.
His definition of investing: the art of using imperfect information to make probabilistic assessments about an inherently unknowable world.
A single year is simultaneously trivial in a thirty year horizon and long enough for random events to derail even the most careful forecast, which is why annual outlooks fail so reliably.
The December 2019 look-ahead pieces contained no pandemic, and even the most telegraphed policy move in memory, 525 basis points of Fed hikes, still caught the market by surprise.
Investing is simple but hard. Not easy, but simple, and mostly a matter of not getting in your own way.
Focus relentlessly on what you control: savings rate, asset allocation, having a plan, sticking to the plan, and managing your own behavior during drawdowns.
The list of things you cannot control, Fed moves, elections, payrolls, GDP, war timelines, inflation prints, quarterly earnings, happens to be exactly what fills print and broadcast media, precisely because it changes constantly.
Every high performance driving school he has attended is a barely disguised defensive driving class, because the real skill is knowing the limits of your own ability and your vehicle’s performance envelope.
There are only two ways to learn humility about your limits: age well and become wise, or lose a great deal of money in the market, which he calls pricey tuition.
Sturgeon’s Law, from science fiction writer Ted Sturgeon, holds that 90% of everything is crap. Ritholtz’s Corollary applies it to finance: 90% of all financial products are crap.
George Box’s line that all models are wrong but some are useful matters because every model has the assumption baked in that the future will resemble the past, so models work until the future changes.
William Goldman’s “nobody knows anything” explains Hollywood passing on Star Wars, The Princess Bride, and Raiders of the Lost Ark, and Keanu Reeves being unable to get John Wick made before it became a $2 billion franchise.
On whether AI breaks Sturgeon’s Law, Ritholtz says he uses Claude, NotebookLM, and other tools productively for research, but that asking models to write produces mostly unreadable output and the 90% figure is timeless.
Denominator blindness is the first innumeracy: Shark Week terror against roughly five attacks a year among 8 billion people, while humans kill tens of millions of sharks annually.
The failure gap is the modern form of survivorship bias. Research across more than thirty life domains found failure occurs about 61% of the time while surveyed people estimated 41%.
Financial survivorship bias started with 1990s mutual fund returns, where funds that closed or merged had their bad data pulled from the record, and now shows up as social media where only successes are visible.
The third innumeracy is compounding. Markets generate exponential returns while the physical world is arithmetic, so exponential growth is simply foreign to human intuition.
The synthesis: you do not have to be brilliant to be a successful investor, you just have to be less stupid.
Knowing whether you are in a bull or bear market is a psychological hack rather than a timing tool, since cycles are only identifiable in hindsight and bull markets run longer than anyone expects.
From 1982 to 2000 the S&P 500 multiple went from about 7 to about 32, meaning roughly three quarters of the gain came from multiple expansion rather than earnings growth, which is a psychological rather than fundamental factor.
The 1966 to 1982 bear market took the Dow from 1000 back to 1000 over sixteen years, a roughly 75% real loss, yet anyone dollar cost averaging through it never saw those prices again after 1982.
The same held for 2000 to 2013: purchases made the day before the October 2007 high and the day before Lehman collapsed all went on to rise three to five times over.
Valuations tell you about future expected returns, not about timing. Stocks were famously cheap in 1977 and 1978 and still took three or four years to get into the green.
Avoiding equities because they look expensive on price to earnings or CAPE would have kept you out of US stocks since roughly 2015 and out of large cap growth for decades. A valuation is a flexible snapshot, not the moving picture.
Externalities such as wars, terror attacks, assassinations, and natural disasters cause markets to wobble and then resume the prior trend, because they usually do not change corporate revenues and profits.
Big wars are the exception, because they trigger a massive government response that historically produces both an inflation surge and a large market rally, seen after World War II, after Vietnam, and after the 2020 pandemic stimulus.
The COVID crash confused people because of the availability heuristic. The visibly dying sectors added up to roughly 6% of the S&P 500 while the index rallied about 69% off the March lows.
He argues people are repeating the identical error with AI, conflating anxiety about their own job with what drives a market, and points to the Magnificent 493 that stand to get more productive rather than the seven building the technology.
Bessembinder’s core finding is that essentially all equity value comes from 1% to 2% of stocks, which frames stock picking as roughly 50-to-1 odds before costs.
Indexing over a decade or two puts you in the top half of market performance, and over 25 to 30 years in the top quartile, which he says is not quite a sure thing but close.
Bogle’s framing survives: you cannot get alpha without first taking beta. The index is the Christmas tree and everything else is ornaments and garland.
The forecasting joke that carries real advice: give a price or give a date, never both, and always couch discussions of the future in probabilities rather than binary outcomes.
Traders lie to themselves about essentially everything, including what they own and what they sold at the top. Good ones keep a written thesis, a trading journal, and a detailed grip on their profit and loss.
His story of the broker whose one winner was 2% of the book while the four largest positions were losers captures why most self-directed performance disappoints.
Real traders do not predict. They manage losses and let winners run, which is why trend and momentum show up as genuine factors in the Fama-French framework.
The desk anecdote about a trader who claimed to nail every high and low, answered by a colleague noting that for a guy with that record he sure drove a cheap car, is the funniest thing in the episode.
To pursue active management successfully you need an edge that is reproducible, a process, military-grade risk discipline, and awareness of your own blind spots. There is no edge in public news reaching 400,000 Bloomberg terminals simultaneously.
Ritholtz notes that many of the best traders he knows are neuroatypical, and attributes it to an unusual capacity to manage social influence and emotion.
Much of FinTwit’s arguing is people misunderstanding each other’s timelines, with a three-hour trader and a three-decade investor talking past each other.
Concentration is the behavioral mistake he treats most seriously, arriving via founder stock, inherited low-basis positions, generous employer stock matches, or simply having bought Apple or Nvidia long ago and held.
He calls Jack Welch the most overrated CEO in history, arguing he rode the 1982 bull market, exited at the 2000 top, and left a stodgy industrial with a 47 multiple and a brewing GE Capital accounting problem for his successor to absorb.
General Electric’s generous ESOP match left employees with 401(k)s that were 50% to 70% GE stock, which is how a corporate blowup becomes a retirement blowup.
On refusing to sell winners for tax reasons: the surefire way to never pay capital gains tax is to not have any gains. Cisco peaked in March 2000, lost 93%, and took 25 years to return to break even.
His question for concentrated billionaires is what the difference is between one billion and two billion dollars, and his answer is nothing, since everything is already covered for multiple generations.
Sudden windfalls go wrong for the same reason any inexperience does, only with bigger numbers. He describes being surprised by the complexity of his own firm’s succession payouts after thirty years in the business.
A 25-year-old with $10 or $20 million will overspend, will not budget, is exposed to every scam, and then faces friends and family arriving with loan requests and business ideas.
You do not need a hundred people around a windfall. You need an accountant for taxes and budgeting and an attorney for trusts and estates, plus the patience to not buy a house and a Lamborghini the week the company IPOs.
When choosing an advisor, look for track record, process, and temperament, get referrals, and interview several. People spend more time researching a vacation or a refrigerator than their retirement plan.
Josh Brown’s rule of never hiring an advisor without a blog is really about wanting someone who can articulate how and why they invest, because a client who understands drawdowns in advance is less likely to act badly during one.
The seat-back safety card analogy: 30,000 feet with a flamed-out engine is too late to start reading. Advisors should be preparing clients for the end of the bull market while everyone is calm on the ground.
His prescription: start with a plan, dollar cost average as money arrives, pay yourself first, and recognize that money without purpose tends to disappear.
Purpose is what lets you match risk to goal, whether that purpose is retirement, education for children and grandchildren, or philanthropy.
Build a broad index core, then season it with whatever you believe in, and if you enjoy trading, carve out 3% to 5% as a cowboy account so your lizard brain has somewhere to go that is not your real portfolio.
On bonds: not for people in their twenties and thirties. He thinks the 40 in a 60/40 portfolio looks awfully high, and is comfortable with all equity for a half-century horizon.
The price of all-equity is volatility: 5% drops twice a year, 10% drops in most years, and a serious drawdown roughly every four years.
In retirement, bonds exist so a 4% distribution can come from the fixed income side when markets are down, which extends portfolio life by avoiding sales into weakness.
For people who have already won, he likes a slug of tax-free municipal bonds, using the example of $10 million out of $100 million throwing off $400,000 to $500,000 tax-free to cover expenses while the rest compounds.
On alternatives, he has moved from never to sometimes: the top funds are worth it if you can access them, but Renaissance Technologies’ Medallion fund fired its outside investors, which tells you how access works at the top.
Jim Chanos’s observation frames the problem: forty years ago there were fewer than a thousand hedge funds and they all generated alpha, today there are 15,000 and it is the same thousand generating alpha.
He is unenthusiastic about private credit reaching 401(k)s, noting that when an institutional product gets sold to retail it usually means the institutional buyers are exhausted.
His response to investors surprised by gates on private vehicles is that a seven-year lockup was disclosed in the words seven-year lockup.
On new cars he is emphatic and contrarian for the genre: buy one, because modern airbags, seatbelt tensioners, collision avoidance, and blind spot detection are safety purchases rather than consumption.
Lee Cooperman was driving a 25-year-old Passat because his money was earmarked for charity, and bought a Lexus a week after Ritholtz pointed out that staying alive is how he keeps earning returns for those charities.
Ritholtz was T-boned in a 2017 Panamera that was totaled while he and his wife walked away with a chipped tooth, and says a twenty-year-old car would have changed that outcome.
On lease versus buy: leasing means purchasing the three most expensive years of a car’s life. His preferred move is buying off-lease, and he paid roughly half of MSRP for a 2014 BMW M6 he has now owned for a decade.
He raises the larger question of why anyone will own a car at all in a world of Waymos and robotaxis, while admitting his own preference is analog cars with stick shifts and no screens.
Financial success, to him, is freedom, opportunity, and optionality, and the worst part of being poor is the endless stress of covering basic bills rather than the absence of luxuries.
Comparison is the thief of joy, which is why he treats social media as corrosive and offers the Zillow sold-prices trick as a reliable way to feel terrible about a house you like.
The hardest problem his firm faces with clients who have hit their number is getting them to actually spend, including giving an inheritance while alive enough to watch it be enjoyed.

Detailed Summary

Why Billionaires and Experts Are Bad at Forecasting

The conversation opens on billionaires making economic forecasts on television, and Ritholtz separates what they do that everyone does from what they do that is worse. Talking their book is universal and, in his view, forgivable, since the bias is unavoidable and usually visible. The problem is that they are typically unqualified to make the forecast at all, and the halo effect ensures the audience does not notice: extreme success in one narrow sphere gets projected across everything the person touches. His formula for what actually determines outcomes is skill plus outside events plus random luck, and skill is the smallest term. He extends this to the creative industries, where greenlighting a film is functionally a prediction about public taste three to five years out, and even a small indie project carries a two to three year lag. Asked what experts are useful for, he is generous and specific: people with genuine domain expertise are excellent at explaining what is happening and why, because they have done the data analysis, know the history, know every player’s track record, and can supply context, color, and nuance. The cardiologist analogy draws the line cleanly. A specialist can tell you how healthy your heart is right now and cannot tell you the date of your heart attack, which is exactly the distinction between asking an advisor how your portfolio looks and asking when the crash arrives. His John Kenneth Galbraith line about two kinds of forecasters, those who do not know and those who do not know they do not know, does the rest of the work.

Spotting Bad Advice: The Three Ds and the Algorithm Problem

Ritholtz’s tells for bad advice start with appeals to emotion. Anything trying to frighten you, activate greed, or trigger FOMO is disqualifying, and any manufactured urgency about a crash next month is a red flag rather than an insight. Good advice, by contrast, is humble and probabilistic, sounding like a range of outcomes with odds attached rather than a confident forecast. He adds the standard follow-the-money question about what the source is being paid to do, then offers the three Ds as a compact filter. Doubt, because people who exhibit no self-doubt are usually unaware of their own limits. Depth, meaning a broad and deep well of experience plus a track record that comes from a repeatable process rather than a lucky streak. And Dunning-Kruger, which he treats as more than overconfidence: it is specifically about metacognition, awareness of the boundaries of your own competence. He names the most expensive sentence in investing as “how hard can it be?” Felix raises the uncomfortable implication, which is that every characteristic of bad advice is more attention-grabbing than every characteristic of good advice. Ritholtz agrees and pushes it further: the algorithms across every major platform actively select for outrage and excitement, so the distribution mechanism itself is biased against the humble and probabilistic voice.

The Daily Beast: Why Financial Media Is a Negative for Long Term Investors

His framing of the media is that the phrase “the daily beast that must be fed” has become so common that people forgot it describes a real mechanism. There are always column inches to fill and 24/7 broadcast hours to occupy, and that relentless demand for attention is chronologically incompatible with a portfolio compounding across decades. Asked how useful the news is to investors, his answer is unusually blunt for someone in the media business: useful for cocktail party conversation, essential for professional traders, and for a long term investor not merely useless but actively negative. The mechanism is that media urges you to action, humans already carry an action bias, and the historical record says most financial decisions are poor and fewer decisions produce better outcomes. He supports it with Hendrik Bessembinder’s work on the S&P 500, which examined every rebalance, addition, and subtraction and found that leaving the index untouched from inception would have significantly outperformed the maintained version. On short-form specifically, he offers two examples. The IRS was forced to issue a 42-point press release debunking TikTok and Instagram tax claims, some of which carry penalties, some clawbacks with interest, and some of which lead to prison. And the older lesson applies directly: never take candy from strangers. Taking advice from an anonymous account with no known track record, methodology, or temperament is the financial equivalent, and his phrase for the outcome is getting financially roofied.

Gell-Mann Amnesia and Building Your Own List of Trusted Voices

Gell-Mann amnesia, which Ritholtz credits to Michael Crichton, describes reading coverage of an event you personally witnessed, recognizing the reporter got it completely wrong, then turning the page and extending full credibility to the next article. He is careful not to overclaim, saying the media generally does a decent job and that the real error is granting institutional credibility to what is actually a collection of fallible humans behind a masthead. His remedy is to build a personal list of voices you have vetted and tested over years. The notable part is what he does with his own list: he has published it, and he explicitly refuses to promote it as a product, on the grounds that outsourcing your thinking to a third party is the underlying disease rather than the cure. Asked how to get more signal and less noise, he says lengthen the time horizon of your media consumption to match the time horizon of your money, since tweets and short video operate on a now-now-now clock while your goals sit ten, twenty, or thirty years out. His test is to imagine Twitter existing in 2001 and ask what could have been said in short bursts that would matter today, and he concludes the answer is nothing. The alternative is books, long-form podcasts, and deep dives. He notes that mass literacy is only a few hundred years old, so declining book reading is closer to mean reversion than collapse, which makes reading a genuine competitive advantage. His pitch for books is the arithmetic of it: two thousand hours and thirty years of accumulated experience, available for about $22, versus the substantial fee he commands for forty minutes on a conference stage.

What Investing Actually Is, and What You Can Control

Asked simply what investing is, Ritholtz gives the definition the book is built on and then decomposes it. Art signals that it is not a clean mathematical formula, with a nod to quants who use math for an edge that is still imperfect. Imperfect information means incomplete, sometimes inaccurate, and sometimes stripped of context. Probabilistic means the work is assessing a range of outcomes and asking how the portfolio is positioned to survive all of them, rather than making binary up-or-down calls. And unknowable is the hardest part, because the world keeps intervening. A single year is simultaneously trivial within a long horizon and long enough for random events to derail careful forecasts, which is why the December 2019 outlook pieces contained no pandemic and why even 525 basis points of Fed tightening, the most telegraphed policy shift in memory, still surprised people. His summary is that investing is simple but hard. Asked how good investors think, he lists recognizing complexity, thinking probabilistically, staying humble, and then the step he says most people never make: concentrating entirely on what you control. Savings rate, asset allocation, having a plan, having the discipline to keep it, and managing your own behavior instead of panicking at every 5% drawdown. The counterpart is recognizing what you cannot control, and his list of uncontrollables, Fed decisions, elections, payrolls, GDP, war timelines, inflation prints, quarterly earnings, is precisely the content that fills financial media, for the simple reason that it changes constantly and therefore makes good fodder.

Defensive Driving, Sturgeon’s Law, and Nobody Knows Anything

Asked why not knowing is a viable posture, Ritholtz reaches for cars. He has taken essentially every advanced driving course available, from Skip Barber to Monticello to Sebring to the manufacturer schools, and his conclusion is that all of them are marketed as high performance racing schools while functioning as defensive driving classes. The actual curriculum is the limits of your own skill and the performance envelope of your vehicle, and he notes that tall body-on-frame SUVs show elevated rates of single-vehicle rollover fatalities precisely because someone asked the machine to do something it could not do. In a portfolio that knowledge saves money, on the road it saves lives. On how to acquire that humility, he offers only two paths: age well and become wise, or lose a great deal of money, which he calls pricey tuition. Turning to the ideas that shaped his philosophy, he starts with Sturgeon’s Law, coined by science fiction writer Ted Sturgeon in response to critics, that 90% of everything is crap, and his own corollary that 90% of financial products are crap, covering most mutual funds, SPACs, most individual stocks, and plenty of bonds. Second is George Box’s line that all models are wrong but some are useful, which matters because every model has embedded in it the assumption that the future resembles the past, so models work until the future changes and then everyone is perplexed. Third and his favorite is William Goldman’s “nobody knows anything,” which explains studios passing on Star Wars, The Princess Bride, and for years Raiders of the Lost Ark, and Keanu Reeves failing to get John Wick made before it became a $2 billion franchise. Combine “nobody knows anything” with “90% of everything is crap” and you have a compact explanation for why consistently beating the market is so difficult. Asked whether AI breaks Sturgeon’s Law, he says he uses Claude, NotebookLM, and other tools productively for research, finds their writing mostly unreadable, expects rapid improvement, and thinks the 90% figure is timeless.

Three Forms of Economic Innumeracy

Ritholtz names three innumeracies that drive bad decisions. Denominator blindness comes first, illustrated by a Shark Week advertisement he had just seen. Roughly five shark attacks a year against 8 billion people makes the personal risk effectively zero, and the correct denominator runs the other way: humans kill tens of millions of sharks annually while sharks kill three or four of us. Survivorship bias is second, and he frames its modern version as the failure gap. In finance it originated with 1990s mutual fund performance data, where funds that closed or merged simply had their records removed from the aggregate. Today it appears as social media showing only the successes. The research he cites reviewed more than thirty life domains and found that failure occurs about 61% of the time, while roughly 32,000 surveyed people estimated 41%, a systematic overestimate of success rates. He notes the survey even had hockey losses at 44% despite the NHL having no ties, which he flags as suspicious. His illustration is the Wall Street Journal feature listing which classic cars appreciated most over fifty years: useful for telling you who won a finished game, useless for telling you which cars to buy today, and nobody ever runs that harder version of the piece. The third innumeracy is compounding, and he includes himself among those who still fail to intuit it, because markets generate exponential returns while the physical world is arithmetic and human intuition was built for the latter. His conclusion from all three is the line that anchors the book: you do not have to be brilliant to succeed as an investor, we all just have to be less stupid.

Cycles, Valuations, and What Externalities Actually Do

On knowing whether you are in a bull or bear market, Ritholtz says the value is psychological rather than tactical. Bull markets run ten to twenty years, feature expanding activity and rising wages, and are characterized by investors paying progressively more for each dollar of earnings. His numbers from 1982 to 2000 are striking: the S&P 500 multiple went from roughly 7 to roughly 32, meaning about three quarters of the gain came from multiple expansion rather than improving fundamentals, which makes the largest driver of that bull market a psychological one. The 1966 to 1982 stretch is his counterexample and his best argument for persistence: the Dow started at 1000 and finished at 1000 sixteen years later amid high inflation, a real loss around 75%, yet anyone dollar cost averaging throughout owned shares at prices never seen again after 1982, so their worst purchases still sat far below the eventual price. The same held from 2000 to 2013, where buying the day before the October 2007 high or the day before Lehman collapsed still produced three to five times returns. He credits his colleague Nick Maggiulli’s Just Keep Buying framing for the discipline. On secular cycles he counsels Solomonic acceptance that every expansion, recession, bull, and bear market ends, and that cycles are only identified in hindsight. His answer to the client question of whether AI is a bubble is empirical: he went to the data and found the four largest companies today trade at multiples nothing like 2000, with Microsoft’s around 20 today against roughly 50 then, and Nvidia’s back near its 2019 level. On valuations, the honest use is as a signal about future expected returns, since expensive markets imply below-average forward returns and cheap ones imply the opposite. As a timing tool it fails, and he notes stocks were cheap in 1977 and 1978 and still took three or four years to reward buyers, while avoiding equities on valuation grounds would have kept you out of US stocks since roughly 2015. On externalities, history says markets wobble and then resume their prior trend after wars, terror attacks, assassinations, and disasters, because these events usually do not change corporate revenues and profits. Big wars are the exception, generating massive government responses that produce both inflation surges and large rallies, as after World War II, after Vietnam, and after the 2020 stimulus that was the largest as a share of GDP since World War II.

The COVID Lesson and Why Indexing Wins

The COVID crash produced the same question over and over: how can the market be rallying when everything around me is closing? Ritholtz went down a research rabbit hole and came back with the availability heuristic as the answer. People assess the world from personal, local, unweighted observation, while the index is market cap weighted and global. He added up the visibly dying sectors, restaurants, retailers, travel, hotels, hundreds of companies, and found they came to roughly 6% of the S&P 500, while the index rallied about 69% from the March lows. Apple, Microsoft, Amazon, Target, and Walmart were fine, and the companies large and agile enough to pivot toward home delivery did more than fine. He argues people are running the identical error now with AI, asking how markets can rise if everyone loses their job, and points to what he has called the Magnificent 493, the companies positioned to become more productive and more profitable because of AI rather than the handful building it. His broader point is that your personal experience and the drivers of the market are two genuinely different objects. On indexing, he grounds the case in Bessembinder’s finding that essentially all equity value is produced by 1% to 2% of stocks depending on the region, which sets the base rate for stock selection at roughly 50-to-1 before costs. His analogy is a plane going down with one parachute among fifty people. Against that, indexing over a decade or two puts you in the top half of market performance and over 25 to 30 years in the top quartile, which he calls not quite a sure thing but close. He closes with Bogle: you cannot get alpha without first taking beta, so the index is the Christmas tree and everything else you add is ornaments and garland, and Passmore supplies the companion line about buying the haystack rather than hunting the needle.

Traders, Behavioral Mistakes, and the Danger of Concentration

Asked what traders lie to themselves about, Ritholtz answers everything: what they own, what their winners and losers were, what they sold at the top. Genuinely good traders are brutally honest, keep a written thesis and a trading journal, and understand their profit and loss in detail. His example from his strategist days is the broker complaining about poor performance whose winning stock was 2% of the book while the four largest positions were losers and the middle ten were flat, churning constantly. Real traders do not predict; they manage losses and let winners run, which is why trend and momentum appear as real factors in the Fama-French framework, and adding to a rising position feels counterintuitive to almost everyone. He tells the story of a desk colleague who claimed to nail every high and low, until the head trader observed that for a guy with that record he sure drove a cheap car. To pursue active management you need a reproducible edge, a process, military discipline about risk, and awareness of your blind spots, because 400,000 Bloomberg terminals receive the same public news simultaneously. He adds that many of the best traders he knows are neuroatypical, attributing it to an unusual capacity for managing social influence and emotion. On behavioral mistakes he lists having no plan at all, the timeline confusion that drives most FinTwit arguments where a three-hour trader argues with a three-decade investor, not understanding your own needs and risk tolerance, fee drag, tax management after a long bull market, and a general lack of humility in an industry he describes as fake it till you make it. Concentration gets the most attention. It arrives via founder stock, inherited low-basis positions, employer matches, or simply having bought Apple when the iPod launched and held. He names a billionaire client with 98% of his net worth in his own now-public company and asks what the difference is between one and two billion dollars, answering: nothing. General Electric’s roughly 15% ESOP match left employees with 401(k)s that were 50% to 70% GE stock, and he calls Jack Welch the most overrated CEO in history for riding the 1982 bull market, exiting at the 2000 top, and leaving a stodgy industrial carrying a 47 multiple and a brewing GE Capital accounting problem for a successor who absorbed the blame. On the tax excuse for never selling, his line is that the surefire way to never pay capital gains tax is to not have any gains, and Cisco is the proof: it peaked in March 2000, fell 93%, and took 25 years to return to break even.

Windfalls, Choosing an Advisor, and the Actual Prescription

Cash windfalls go wrong for the ordinary reason that inexperience produces mistakes, with the complication that big piles of money produce expensive ones. Ritholtz makes the point against himself: setting up a succession plan at his firm that pays out equity over ten years turned out to be surprisingly complicated even after thirty years in the business, requiring budgeting, quarterly IRS payments, moving cash to the highest-yielding money market because none of it can carry risk, and a new expense structure. If that is work for him, a 25-year-old who suddenly holds $10 or $20 million will overspend, will not budget, and is exposed to every scam plus the arrival of friends and family with loans to request and businesses to fund. His prescription is modest: an accountant for taxes and budgeting, an attorney for trusts and estates, and the patience to not buy a house and a Lamborghini the week the company goes public. On selecting an advisor, the same checklist as for information sources applies, with referrals, track record, process, and temperament, plus the AQR concept of organizational alpha, which is that the value is in the plan, the discipline, and the behavior rather than the portfolio. He notes people spend more time researching a vacation or a refrigerator than their retirement plan, and that with roughly 400,000 advisors in the United States there is no excuse for not interviewing several. Josh Brown’s rule about never hiring an advisor without a blog is really about wanting someone who can articulate why they invest as they do, since a client who was told in advance that markets fall is less likely to act badly when they do. His seat-back safety card analogy captures it: 30,000 feet with a flamed-out engine is too late to start reading, which is why his firm tells clients on calm days that this bull market will end, possibly next Tuesday and possibly in 2032. His own prescription is a plan, dollar cost averaging as money arrives, paying yourself first, and the recognition that money without purpose disappears, because purpose is what lets you match risk to goal. Build a broad index core, season it with whatever you believe in, and if you enjoy trading, carve out 3% to 5% as a cowboy account where, as he puts it, he does plenty of dumb things, specifically so his lizard brain stays away from the portfolio that is actually compounding.

Bonds, Alternatives, Cars, and What Success Means

On bonds he is direct: not for people in their twenties and thirties, and the 40 in a 60/40 portfolio looks awfully high to him for someone with a fifty year horizon. The trade-off is volatility, with 5% drops twice a year, 10% drops in most years, and a serious drawdown roughly every four years. Fixed income earns its place near retirement, where taking a 4% distribution from the bond side during a down market avoids selling stocks into weakness and extends how long the assets last, and for people who have already hit their number, where he likes a slug of tax-free municipal bonds, describing $10 million out of $100 million throwing off $400,000 to $500,000 tax-free to cover expenses while everything else compounds untouched. On alternatives he opens by reminding everyone of his corollary about 90% of financial products, then softens his historical position. Access is the whole question: Renaissance Technologies’ Medallion fund returned its outside investors’ capital once it had enough, and Jim Chanos observes that forty years ago fewer than a thousand hedge funds existed and all generated alpha, while today 15,000 exist and it is the same thousand generating it. He is unenthusiastic about private credit reaching 401(k)s, noting that institutional products being sold to retail usually signals exhausted institutional demand, and his response to investors surprised by gates is that a seven-year lockup was disclosed in the phrase seven-year lockup. Then the car question, where he is emphatically contrarian for the personal finance genre. Buy the new car, because modern seatbelt tensioners, airbags with functioning actuators, collision avoidance, and blind spot detection are safety purchases rather than consumption, and this is what the spending scolds have cost us. Lee Cooperman drove a 25-year-old Passat because the money was earmarked for charity, until Ritholtz pointed out that staying alive is how he keeps generating returns for those charities, and bought a Lexus a week later. Kawhi Leonard driving an old SUV with a $103 million contract is the same error, since knees, wrists, and ankles are the asset. Ritholtz himself was T-boned in a 2017 Panamera that was totaled while he and his wife walked away with a chipped tooth. On lease versus buy, leasing means paying for the three most expensive years of a car’s life, and his preferred move is buying off-lease, where a 2014 BMW M6 cost him about half of MSRP and has held roughly half of that over ten years. Asked to define success in a portfolio, he answers freedom, opportunity, and optionality, notes the worst part of being poor is the constant stress of covering basics rather than the absence of luxuries, and warns about declining marginal utility and the guy with the bigger boat. Comparison is the thief of joy, he says, which is why he calls social media corrosive and offers Zillow’s sold-price filter as a reliable method for feeling terrible about a house you were perfectly happy with. His closing note is the one his industry rarely addresses: the hardest problem with clients who have won and entered decumulation is getting them to spend, whether that means handing over an inheritance while alive enough to watch it be enjoyed, or simply building a structure that reduces stress.

Notable Quotes

“There are two kinds of forecasters. Those who don’t know and those who don’t know they don’t know.”
Barry Ritholtz, quoting John Kenneth Galbraith on Wall Street’s forecasting record

“Is there a better bargain in the world? Someone takes a thousand hours, two thousand hours, and a lifetime of experience. You could buy it for $21.95.”
Barry Ritholtz, on why books beat every other form of information consumption

“Investing is the art of using imperfect information to make probabilistic assessments about an inherently unknowable world.”
Barry Ritholtz, giving the definition the entire book is built on

“The paper’s authors reviewed over 30 life domains and determined that on average, failure occurs about 61% of the time. When they surveyed all these people, they guessed that failure occurred 41% of the time.”
Barry Ritholtz, describing the failure gap research and how badly humans estimate their own odds

“To be a successful investor, you don’t have to be brilliant. We all just have to be less stupid.”
Barry Ritholtz, summarizing what denominator blindness, survivorship bias, and compounding add up to

“Your personal experience and what drives the markets are two totally different things.”
Barry Ritholtz, on the availability heuristic that confused everyone during COVID and is confusing them again about AI

“Let me tell you a surefire way to never pay capital gains tax. Don’t have any gains.”
Barry Ritholtz, on investors who refuse to trim concentrated positions for tax reasons

“Which part of illiquid seven-year lockup confused you? Was it the seven-year lockup part or the this is not liquid part?”
Barry Ritholtz, on retail investors surprised when private funds gated withdrawals

“This is what the spending scolds have done to us. If you love your family, get them the latest greatest protection. You won’t regret it.”
Barry Ritholtz, arguing that a new car is a safety purchase rather than a consumption decision

“Comparison is the thief of joy. If you’re always looking at everything else and comparing yourself to other people, you’re always going to be disappointed.”
Barry Ritholtz, in his closing answer on what financial success actually means

The full episode runs about a hundred minutes and the back half carries more weight than the front, particularly the sections on concentration risk, sudden windfalls, and the closing stretch on cars, spending, and decumulation. Watch the full conversation here.

Tag: Masters in Business podcast

Barry Ritholtz on How Not to Invest: The Failure Gap, Sturgeon’s Law, Why Financial News Actively Hurts Your Returns, and Why He Says You Should Buy a New Car