The S&P 500 printed 7,818.93 at Monday's close — a record — and with it, the fourth birthday of a bull market that has now stacked 119% on top of its October 2022 floor. Every aggregator in my feed is running the same lede: six of six prior bull markets kept rising after year four, so this one will too. I've watched that sentence copy-paste itself into a dozen newsletters in the last twelve hours. Speed is the only currency that never inflates — and right now the crowd is spending it on a statistic with a sample size of six.
Here's what nobody's pricing. The structure holding this equity rally up looks identical to the structure holding up a DeFi protocol that got valued on TVL it hasn't earned yet. And if you trade crypto — nothing more than the highest-beta expression of global risk appetite — that structure is your problem, magnified threefold.
Let me lay out the numbers, because the numbers are where the story actually lives. The S&P's trailing price-to-earnings ratio sits at 26x. Its ten-year average is 23.6x. On the earnings companies have already booked, the index is expensive — roughly 10% above its own decade-long norm. But the forward PE, the ratio built on the next twelve months of expected earnings, is sitting almost exactly on that same ten-year average. Read those two facts together and you get this market's entire thesis in one sentence: the index is only fairly valued if earnings accelerate meaningfully from here. Strip out the acceleration and 26x is what you're actually paying. That isn't a valuation. That's a bet.
I've spent thirteen years watching crypto re-learn this in public. When a Layer 2 launches and its token trades at a fully-diluted valuation built on "projected" fee revenue three years out, nobody calls it expensive — because the forward multiple looks reasonable. Then the fees don't arrive on schedule, the forward multiple quietly becomes the trailing multiple, and the price corrects to wherever the real earnings justify. Same mechanism. Different ticker.
The historical framing is the other half of the story. Six prior bull markets reached their fourth birthday. All six kept climbing afterward. But four of those six rolled into a bear market soon after, and one — the 2006 cohort — saw an investor who bought the birthday print sitting on roughly a 50% loss by March 2009. The "6 for 6" headline and the "4 of 6 then crashed" reality describe the same dataset. The framing does the work. The data just sits there.
This is where I stop reading the equity tape and start reading what it means for the market I actually trade. Crypto doesn't have its own macro cycle; it has a leveraged copy of everyone else's. When the S&P's valuation rests on a single support pillar — forward earnings — crypto rests on that same pillar with no shock absorber at all. Equities carry dividends, buybacks, and a legal claim on cash flow. A governance token carries a roadmap. Governance isn't a feature you bolt onto a token; it's the entire load-bearing wall, and when that wall is made of projected fees, the whole structure flexes with the macro weather.
Look at April 2025. The S&P saw an intraday drawdown north of 20% — a technical bear market by the intraday measure — driven by a policy shock the commentary treated as a footnote. On a closing basis, the same stretch only touched -18.9%. Same market. Two conclusions. One is a bear market; the other is a bull market that "held." That definitional gap isn't a technicality — it's the whole game. And in crypto it's worse, because we don't even agree on the benchmark. Is BTC in a bear market because it's down from its all-time high, or in a bull market because it's up year-over-year? Both answers are defensible, and whoever shouts loudest sets the narrative for the week.
Now apply the single-pillar problem to my own backyard. Dencun made blob space cheap, and every rollup suddenly looked like a business because its data-availability costs collapsed. Sequencer margins ballooned, token models got repriced upward, and the forward multiples looked fantastic. But blob space is a subsidized resource, and it will saturate — my working estimate is inside two years — at which point the cheap-DA tailwind reverses and rollup costs step back up. Every L2 token currently valued on "cheap data forever" carries the exact flaw of an S&P priced on "earnings accelerate forever." Both are forward multiples wearing a trailing multiple's clothes, and both snap back the moment the projection misses.
The same pattern runs through DeFi, just quieter. Every few quarters, a fresh cohort of VCs funds a "liquidity fragmentation" narrative — the idea that capital is dangerously scattered across chains and needs a unifying product to rescue it. I've audited enough of these pitches to know the fragmentation isn't the disease; the pitch is the product. Fragmentation is what a competitive market looks like. The "solution" is a sales motion. And the sales motion is priced on future liquidity that, like the S&P's forward earnings, has to show up on schedule or the valuation unwinds.
That's also why I keep circling back to the exchange layer when I talk about survival. Binance's $4.3 billion settlement didn't weaken it — it licensed it. In a bear market the moat isn't speed or product; it's the regulatory ticket nobody else can afford. The CEX market leans on a single license-holder the way the S&P leans on a single pillar of earnings, and that concentration is its own kind of fragility.
Everyone's celebrating the fourth birthday. Almost nobody is asking whether the birthday means anything. A calendar anniversary is not a causal mechanism. The S&P doesn't check its age and decide to stop. What actually ends bull markets is a change in the discount rate, a break in earnings, or a policy shock — and this commentary is conspicuously silent on all three. It never mentions rates. It never mentions inflation. It never mentions the ten-year Treasury yield, the single most important number for a market priced at 26x trailing earnings. It frames a four-year cycle as a rule of nature while admitting, in its own text, that the calendar "has almost no predictive value for calling a top." That contradiction is the tell.
And the asymmetry is the part that should scare you. The 2006 birthday buyer lost half their money — not because they were wrong about direction, but because they were right about direction and wrong about timing, entering exactly where the risk-reward had flipped. That's the trap crypto repeats every cycle with more leverage. Buying the top of a cycle tail isn't a strategy. It's a coin flip with a loaded downside.
So what do I watch? Forward earnings revisions — if they start getting cut, the forward PE snaps toward the trailing PE and the "fairly valued" story evaporates. The ten-year yield — if it runs, high-multiple assets break first, and crypto breaks hardest. And in my own house, whether L2 fee revenue actually materializes before blob space saturates. Speed is the only currency that never inflates, but it's also the only currency that can't buy you out of a bad entry. I don't predict the market; I ride its heartbeat — and right now the heartbeat is a forward multiple betting on an earnings acceleration nobody has confirmed yet. Watch the revisions. That's where this turns.
