Barbell Wealth - Why the Safe Middle Is the Most Dangerous Position

Before the exit, you were already running a barbell strategy.

Salary on one side. Maybe some consulting revenue, early customer payments, and a bit of rental income. Stable, predictable, enough to keep the lights on. And on the other side, the company. Massive, concentrated, illiquid bet on yourself. Most of your net worth is tied up in one thing that could go to zero or make you wealthy.

That structure worked. Not because it was comfortable, but because it was honest about where the risk lived. Stability covered the downside. Company carried the upside. Two extremes. Nothing in the middle pretending to be both.

Then the exit happened. Someone in a well-appointed office showed you a chart with a smooth upward line and explained that the responsible thing to do with your new capital was to put it in a "balanced portfolio." 60% stocks for growth. 40% bonds for protection. Simple. Elegant. Nobel Prize-backed.

And just like that, you abandoned the exact structure that made you wealthy.

I find this pattern maddening. Not because balanced portfolios are wrong, but because nobody explains what "balanced" actually means in practice. It means concentrating your entire portfolio in a narrow band of risk where both sides can lose money simultaneously. Which is precisely what happened in 2022.

Pitch sounds reasonable. Diversification. Risk-adjusted returns. Historical performance charts that slope gently upward. Wealth managers sell the middle because it's easy to explain, easy to monitor, and easy to charge fees on. A barbell structure (large cash positions on one side, illiquid asymmetric bets on the other) is harder to sell. It requires more explanation, more patience, and generates fewer transactions. From a business model perspective, balanced wins every time. From a wealth preservation perspective, it's more complicated than anyone admits.

What "Balanced" Actually Delivered

In 2022, the S&P 500 fell 18.1%. Bloomberg's Aggregate Bond Index dropped 13%. For the first time in over 40 years, both halves of the 60/40 lost money simultaneously. Combined, the portfolio declined roughly 17.5%: its worst year since 1937, its fourth-worst in 200 years.

One bad year doesn't kill a strategy. But 2022 wasn't random. It exposed something structural that had been hiding in plain sight for decades.

Bond yields fell almost continuously from 16.5% in 1981 to 1.05% in July 2020. During that entire stretch, bonds reliably moved in the opposite direction to stocks. When equities dropped, bond prices rose. That negative correlation was the whole engine behind "balance."

When inflation returned and rates reversed, the correlation flipped positive. Stocks and bonds started falling together. That "diversification" wasn't a permanent feature of these two asset classes. It was a feature of one rate regime that lasted roughly 40 years. Regime changed. Diversification vanished.

Anyone looking at where bond yields started 2022 (1.51% on the 10-year) could see the ballast had almost no room to rise and enormous room to fall. "Safe" was priced for perfection in a world that was visibly imperfect.

What does fragility actually look like? Not something obviously risky. Something stable for years, predictable, respectable on paper. Something that breaks at the exact moment we need it most. Quarterly statements looked fine. Right up until they didn't.

Where Fragility Hides

Nassim Taleb's thinking on this has stayed with me since I first read Antifragile years ago. He makes a distinction I haven't seen anyone else make as clearly: the moderate middle isn't a compromise between safety and aggression. It's a concentration of risk dressed up as responsibility.

Consider what a "balanced" portfolio actually does in a crisis. Stocks drop. Normally, bonds would rise to cushion the blow. But when correlations flip (as they did in 2022 and repeatedly before the 1980s), both sides move together. Not balanced at all. Doubled up on the same bet: that the macroeconomic environment stays benign.

Obviously, fragile positions are easy to spot. Leveraged portfolio. Business with one client. Concentrated stock position with no hedge. Most people recognise these and take precautions.

Moderate positions are worse in one specific way: they feel safe while being fragile. Quarterly statements look orderly. Volatility is manageable. Everything hums along for years, building false confidence, until a regime change reveals that "moderate risk" was never moderate at all. The danger isn't the risk itself. It's the mismatch between the risk a portfolio appears to carry and the risk it actually carries.

I've seen this play out in how portfolios get presented to founders after exit. Risk metrics look reassuring. Standard deviation is within acceptable bounds. Maximum drawdown projections use historical data from the benign period. Nobody models the scenario in which the fundamental relationship between stocks and bonds changes, because that hasn't happened in the dataset they're using. When it does happen, the "moderate" portfolio behaves like an aggressive one, and the founder learns the difference between modelled risk and lived risk.

Taleb's alternative is to skip the middle entirely.

Barbell Strategy: Two Extremes, Nothing Between

Radical safety on one end. Calculated asymmetry on the other. No blended, moderate, "growth and income" middle.

Safety first, and boring on purpose. Cash, short-duration sovereign bonds, maybe physical assets. Instruments that don't lose serious value under any scenario: recession, inflation, geopolitical crisis, whatever comes. This side isn't an investment. It's a guarantee that the portfolio survives no matter what happens to everything else. If every other position went to zero tomorrow, this side keeps a founder solvent, liquid, and calm.

On the other end, asymmetry. Positions where the downside is defined (lose what was put in, no more) but the upside is open-ended. Venture allocations in industries where a founder has genuine domain knowledge. Co-investments alongside institutional managers in specific deals. Real estate development with hands-on involvement. Concentrated conviction bets where specific expertise provides an actual informational edge, not the perceived edge that overconfidence creates.

What's absent matters more than what's present. No "moderate growth" funds. No balanced mandates. No medium-risk bonds that carry more downside than the yield compensates for.

The combination is counterintuitive but more resilient than any blend. Safety makes the portfolio indestructible. Asymmetry gives it exposure to outsized gains. During dislocations, when asset prices drop, and the moderate middle is bleeding, the barbell holder has cash to deploy. Buying when everyone else is being forced to sell. Barbell doesn't just survive volatility. It creates the conditions to profit from it.

We spent years embracing uncertainty, making asymmetric bets, building through chaos. That instinct is right. What's wrong is parking it at the door when someone hands us a glossy pitch about balanced allocation. Barbell is the same shape that built the wealth in the first place, just translated into a portfolio that doesn't depend on one company.

Subtract Before You Add

Taleb has a principle I keep coming back to: via negativa. Improve by removing, not adding.

Applied to portfolio construction, the first step after exit isn't deciding what to buy. It's deciding what to eliminate. Before adding a single new position, the questions that matter: what in the current portfolio could cause serious damage? What positions have poorly defined downside? Where are hidden correlations — assets that look diversified in a spreadsheet but will move together when it counts? What structures depend on a single counterparty, a single jurisdiction, or a single assumption about rates continuing in one direction?

Most post-exit portfolios are full of things that need removing before they need replacing. Private credit allocations that looked safe until redemption gates began to appear. "Diversified" fund of funds charges layered fees for access to positions available elsewhere at a fraction of the cost. Three "different" equity funds that all hold the same mega-cap tech stocks in their top ten positions.

And there's a subtler form of fragility that rarely shows up in portfolio analytics: correlation clustering. Five holdings that appear diversified by asset class but are all sensitive to the same risk factor. Three real estate funds in different geographies that all depend on low interest rates. A mix of growth equities and venture capital that both collapse when the risk-off switch flips. Diversification that only works in calm markets isn't diversification. It's an untested assumption.

"The good is mostly in the absence of the bad." That line from Antifragile has shaped how I think about wealth management more than any allocation model. Most founders are tempted to add. The better move, almost always, is to subtract first.

What This Looks Like at $10M-$50M

At the institutional scale, barbells get sophisticated. Treasury inflation-protected securities, systematic trend-following, long-volatility options strategies, venture portfolios with hundreds of positions. Dalio's All Weather approach (allocating risk equally across economic regimes so no single environment can sink the portfolio) is the institutional cousin of barbell thinking. It works because each component responds to different economic conditions, so the portfolio stays resilient regardless of which regime shows up. The principle translates to smaller portfolios. Specific instruments change. Architecture doesn't.

On the safety side, a founder in this range might hold 12-24 months of living expenses in cash or near-cash. Beyond that, short-duration sovereign bonds or high-quality money market instruments. A paid-off property. Return isn't the goal here. This is insurance that pays the holder (via yield) rather than charging a premium.

On the asymmetric side, the specifics depend entirely on what the founder actually knows. Someone who sold a SaaS company might angel invest in adjacent software businesses where they can evaluate product, team, and market better than any generalist fund manager. Someone with real estate operating experience might take on development projects with hands-on involvement. Founders with deep sector networks often find co-investment opportunities in specific deals rather than blind-pool funds.

The positions that work on the asymmetric side tend to share two characteristics: a defined maximum loss and open-ended upside. They also tend to be areas where the founder has a genuine informational advantage. Not a fund they heard about at a dinner party or a pitch deck that landed in their inbox.

What I see in practice, working inside wealth management, is the opposite pattern. Founders get steered toward the moderate middle almost by default. "Growth and income" buckets, balanced mandates promising steady returns in normal conditions — returns that then correlate with everything else when conditions stop being normal. The middle is where fragility concentrates, because that's where the gap between expected risk and actual risk widens most during a crisis.

One useful filter before committing to any investment approach: how long has it survived? Diversification across genuinely uncorrelated real assets (property, productive businesses, physical commodities) has preserved wealth for centuries. The 60/40 worked for one rate regime. Forty years is a long time for a career. It's nothing for a wealth strategy that needs to survive generations.

I wrote about portfolio construction and the broader investment philosophy behind this approach in more detail in the Playbook. The barbell strategy is the overarching architecture. Implementation details vary widely depending on each person's situation, knowledge base, and risk tolerance.

Why the Barbell Feels Wrong (and Works Anyway)

Nothing about this is psychologically easy.

A quarter of a portfolio sitting in something earning 4-5% while markets rise and everyone around seems to be doing better. Nobody brags about their money market allocation at dinner. The safety side is invisible. It only proves its value when everything else breaks.

Meanwhile, the asymmetric side includes positions that will lose money. Some will go to zero. That's not a risk to be managed away. That's the design. The barbell works because the safety side absorbs the losses while the winners on the other end more than compensate over time. But "over time" means years. Sometimes a decade. The interim experience involves watching some bets fail while safe money earns a modest yield and does nothing dramatic.

There's also the social dimension, which nobody talks about. Post-exit founders are part of a peer group where investment conversations are constant. Everybody has a deal, a fund, an allocation they're excited about. Sitting on a large cash position while others talk about their returns requires a specific kind of discipline. It feels like falling behind. It isn't, but it feels that way. The identity pressures that follow an exit don't stop at the office door; they follow founders straight into their investment decisions.

By comparison, the 60/40 is psychologically easier. It goes up most years. Movements are moderate. Quarterly statements look orderly. It feels like something a responsible person would own.

Structurally, the barbell is sound. It survives scenarios that the balanced portfolio can't. Founders who apply it are positioned to buy when others are forced to sell. And their wealth doesn't depend on one correlation regime holding steady for the rest of their lives.

As Morgan Housel wrote in The Psychology of Money: "Risk is what's left over when you think you've thought of everything." A balanced portfolio prepares for normal volatility, a stock correction, a rate move, and the risks that are easy to imagine. The barbell prepares for the ones we can't imagine. The ones that only become obvious after they've already arrived.

What makes it work long-term isn't the theory. It's what happens during the dislocations that hit every 7-12 years. While balanced portfolios bleed from both sides and their owners sit paralysed, a barbell holder has dry powder. Cash to deploy at prices that looked impossible six months earlier. Properties, businesses, and equity positions are available at discounts that only appear when everyone is selling at the same time. Over a 20-30 year horizon, these moments — buying when forced sellers flood the market — account for a disproportionate share of total wealth creation.

For people who spent careers betting on themselves through uncertainty, this shouldn't feel foreign. Same instinct. Different domain. The hard part isn't understanding it. The hard part is sitting still while it works.