Momentum's Case Is Earnings-Backed — and That's Both the Point and the Risk

Momentum's Case Is Earnings-Backed — and That's Both the Point and the Risk

The momentum factor has been the dominant driver of equity returns in 2026, and the usual objection already sounds rehearsed: it's just a narrative trade, propped up by artificial intelligence optimism and liquidity rather than operating reality. Bank of America's latest research suggests that objection is losing its grip. The earnings revision ratio for the global momentum style — a measure of how many upward analyst estimate changes outnumber downward ones — reached 3.14 last week. That means upgrades are running at more than three times the pace of downgrades.

That's the data point investors want to hear. But the momentum trade is not a single stock, and revision data tells you something is moving — not whether the structure underneath it can hold. The question is whether earnings revisions are broad enough, durable enough, and distributed across enough of the index to justify how concentrated the momentum leadership has become.

The revision data is real. The earnings engine behind it is not slowing. S&P 500 EPS climbed more than 50% in the second quarter, the best growth rate since the post-pandemic surge of mid-2021. Beat rates are at five-year highs: as of late July, 86% of the roughly 305 companies that had reported topped consensus estimates, per FactSet. From March through June, bottom-up EPS estimates rose 3.4%, the largest intra-quarter upgrade since 2021. This marks a potential seventh consecutive quarter of double-digit earnings growth, the longest such streak since the Great Recession.

So momentum names are not coasting on stale positioning. They're being revised higher into the quarter, not dragged down. The factor stack has a genuine operating foundation.

Where the support is concentrated

The concentration problem doesn't disappear because the numbers are strong. Momentum leadership in 2026 has narrowed to a specific set of sectors, with Technology doing the heaviest lifting. In the first five months of the year, sentiment-driven returns in Technology were exceptionally elevated relative to both long-term averages and the post-COVID period. Investors paid up for growth and momentum while sidelining defensive characteristics, causing lower-volatility stocks within the sector to materially underperform.

This is not the same as broad market leadership. It's factor concentration within sector concentration.

The Mag7 — Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla — continue to pull the growth index. In Q1, Mag7 earnings are estimated to have grown nearly 60%, while the remaining 493 S&P 500 companies tracked closer to 19%. The anticipated "convergence trade," where broader market earnings catch up to the mega-cap AI beneficiaries, has been repeatedly pushed forward.

But the broader cohort is not standing still. All 11 S&P 500 sectors reported year-over-year revenue growth in Q1, and five posted double-digit revenue increases. The Q2 beat data shows momentum extending beyond Financials into names like General Motors, AT&T, Wabtec, and CME Group. The proportion of companies beating both EPS and revenue estimates is tracking at a 20-quarter high.

Still, the factor scorecard tells a different story from the headline earnings growth. Momentum is a relative system. It ranks stocks against each other, and the stocks with the strongest revision momentum continue to cluster in the same cohort. That clustering is what makes the trade work — and what makes it fragile when the revision flow slows.

The structural vulnerability

There are three structural headwinds that revision momentum alone doesn't address.

First, net profit margins across the S&P 500 are tracking near 12%, close to post-pandemic highs and well above the 7–8% range typical of the 2010s. High margins are the engine of current earnings growth. They're also the cushion that disappears first when input costs rise or revenue decelerates. The 2026 earnings boom is partly a margin story, and margin expansion doesn't repeat indefinitely.

Second, the market penalty for missing earnings has become sharply asymmetric. Companies that miss estimates have seen average excess returns of -3.9% on the first trading day after release, compared to a beat premium of roughly 0.7%. That asymmetry signals stretched long positioning in high-expectation names. When the bar is this high — and estimates have already been revised up by 3.4% — the risk-reward for the next miss is disproportionate. A few high-profile misses can undo the revision tailwind faster than a dozen small beats sustain it.

Third, BofA's own broader framework points to valuation compression as a parallel force. The bank has argued that 2026 is an era of "higher EPS, lower PE," with the S&P 500 experiencing more than 10% multiple compression since 2025. If valuations are tightening while earnings are surging, the net effect on momentum names — which typically trade at premium multiples — depends on which force is stronger. When earnings revisions outpace multiple compression, momentum wins. If the PE contraction accelerates, even strong revisions may not offset valuation headwinds.

What the momentum retreat told us

The July pullback is instructive. Momentum styles underperformed the MSCI AC World Index by 10.5% that month, while growth fell 7.2% and risk styles declined 13.3%. Dividend and value styles gained ground. Then, in August, growth, momentum, and risk styles turned back around and outperformed, buoyed by continued earnings strength.

That swing is exactly the kind of regime volatility that matters. Momentum doesn't need to be wrong forever to hurt you — it just needs to be wrong at the wrong time. The July retreat showed how quickly the trade unwinds when the narrative (AI investment themes) is questioned, even if the underlying earnings engine hasn't stalled. The August recovery showed it re-engaging as revision data improved.

The pattern isn't unusual. History shows that regimes driven by narrow signals and narrow sectors can persist for extended periods but are vulnerable to sharp reversals when leadership becomes crowded. Neuberger Berman flagged a significantly crowded momentum trade as a source of bumpier returns ahead. SSGA's analysis echoed this: the current dynamics are reinforced by AI enthusiasm, a benign volatility backdrop, and accommodative liquidity. All three can change without any single one being a bearish call.

The scorecard

Let's put the factor stack together.

• Revisions: A+. Upgrades outnumber downgrades 3.1-to-1 in the momentum universe. This is the strongest evidence supporting the trade.
• Earnings growth: A. Over 50% EPS growth in Q2, seventh consecutive double-digit quarter, broadening into more sectors.
• Breadth: C+. All sectors posting revenue growth is a positive, but Mag7 vs. ex-Mag7 dispersion remains severe. Momentum leadership is still concentrated.
• Margins: B-, trending down. Near-record profit margins support current earnings but leave little room for further expansion. Any compression hits the revision flow.
• Valuation: C. Multiple compression has been a parallel force. Momentum names carrying premium multiples are the ones most exposed to PE tightening.
• Miss risk: D. The -3.9% miss penalty versus a 0.7% beat premium is asymmetric. The bar keeps getting higher, and one miss matters more than many beats.

The positioning call

The momentum trade isn't a narrative scam. The revision data is legitimate, and the earnings growth is real. But legitimacy is not the same as unlimited runway. The concentration in Technology and the Mag7 means the revision tailwind is not evenly distributed across the index. When a trade runs on a narrow engine, the question is always how long the engine can carry the weight before breadth needs to widen or the market starts discounting what's already expected.

For a portfolio, that points to structure, not conviction. The momentum sleeve — if you run one — deserves a position proportional to how much you trust the revision flow to persist through Q3. Q3 will serve as a credibility test: management confidence in margins, hiring, capex, and demand will determine whether Q2's exceptional pace is a plateau or a trajectory.

If the revision ratio holds above 3-to-1 and breadth continues to expand, momentum names can be held as the growth bar of a barbell. Pair them with dividend or quality-value names that don't depend on multiple expansion — the staples, utilities, or financials that gained while momentum retreated in July. That structure gives you exposure to earnings growth without betting the entire account on revision continuity.

If the revision ratio drops materially — say below 2-to-1 — or if high-margin names start guiding into compression, the momentum sleeve loses its asymmetric edge. The miss penalty will compound. At that point, Hold is not Sell, but trimming premium-multiple names and rotating toward the broader cohort that's finally catching up on earnings would be the process-consistent move.

Narratives move quickly. The factor stack moves more slowly, and it usually tells you more. Right now, the stack says momentum has operating support but structural fragility. That's a trade worth holding, not one worth over-concentrating.