New Lows at Five Times the Rate of New Highs

The Breadth Number

On 2026-03-10, $24.37$ of the US equity universe printed new 20-day lows. Against that, $5.18$ printed new 20-day highs. The ratio is approximately $4.7:1$ in favor of deterioration. Breadth above the 50-day moving average sits at $35.35$. Risk pressure is $19.19$ — elevated, and unchanged from the prior session.

These four numbers, taken together, constitute a risk-off warning. Not a forecast. Not a prediction about what the market will do next week. A structural observation about what the cross-section is doing right now.

The vol shock ratio is $3.91$. The volume surge ratio is $9.72$. Volume is moving, but it is not moving symmetrically — elevated turnover in a deteriorating breadth environment is not confirmation of distribution; it is distribution.

The Signal Filter Cross-Section

Of $4{,}520$ symbols scanned, $3{,}860$ passed quality filters, and $2{,}660$ triggered at least one technical signal. That $68.9$ signal trigger rate is not remarkable on its own. What matters is the directional composition of those signals.

On the sell side, the single largest count is momentum_20_negative_cross at $377$ triggers. That is a 20-day momentum cross — a symbol whose price crossed below its 20-day momentum reference, indicating deteriorating intermediate-term price trend. Against this, the largest single buy-side signal is williams_r_buy at $1{,}072$ triggers and stochastic_buy at $918$. Both are oscillator-based. Both measure short-term oversold conditions.

The asymmetry here is important. Oscillator-based buy signals fire in downtrends — they are, by construction, triggered more frequently when prices are falling. Momentum-based sell signals are structurally different: they fire when a trend has changed direction and a threshold is crossed. The $377$ momentum_20_negative_cross triggers are not an artifact of oversold conditions. They represent $377$ symbols that have crossed a directional inflection point downward. The $1{,}072$ williams_r_buy signals include many of those same symbols firing an oversold alert on the way down.

Counting raw buy versus sell signal volumes overstates the bullish case in a declining market. The structure of the signals matters more than the count.

Sector Themes Within the Deterioration

The signal filter does not produce a uniformly bearish picture. Within the broader cross-section of deterioration, the top bullish names cluster into four identifiable sectors.

Pharma and biotech is the most concentrated pocket. Dyne Therapeutics (DYN), United Therapeutics (UTHR), and Xenon Pharmaceuticals (XENE) each score a composite decision of $100$ — the maximum — with six buy signals and zero sell signals apiece. Regeneron (REGN) scores $92$. These are not small speculative names with thin liquidity; UTHR and REGN are large-cap, cash-generative businesses. The technical picture across this group is consistent: price above the 200-day moving average, MACD golden cross in at least some names, and no distribution signals. In a risk-off environment, defensive healthcare with durable earnings is a structurally rational destination for rotation capital.

China technology ADRs appear as a second distinct cluster. Bilibili (BILI) scores $100$ with six buy signals including stochastic, RSI, CCI, Williams %R, and MFI — a multi-indicator confirmation across momentum and volume dimensions. Kingsoft Cloud (KC) scores $99$. These names have been structurally depressed relative to their US peers, and the current technical readings suggest that the selling pressure that characterized their recent history may have exhausted itself. Whether the fundamental case supports continuation is a separate question. The signal architecture does not require one to take a view on Chinese regulatory risk to observe that the selling dynamics have changed.

Semiconductors contribute STMicroelectronics (STM) at a composite of $99$. A European-listed semiconductor name appearing in the top decile of a US cross-section scan on a risk-off day is notable. STM's primary exposure is analog and embedded processing — not the high-performance compute segment that has driven the dominant semiconductor narrative for the past two years. Its presence here likely reflects both the technical setup and a degree of relative insulation from the thematic compression that is affecting US-listed AI semiconductor names.

3D Systems (DDD) is the top-ranked name with a composite of $100$ and seven buy signals — MACD golden cross, Ichimoku breakout, price above MA200, Bollinger upper breakout, Keltner breakout, RSI centerline bull, and positive momentum. Seven simultaneous buy signal triggers across structurally independent technical frameworks represent a high-conviction technical setup. DDD has been a persistently distressed equity; a clean technical breakout from a low base in a deteriorating broad market is precisely the kind of idiosyncratic setup that survives macro headwinds when the underlying name-level setup is strong enough.

The Bond ETF Signal

The clearest statement of regime in the entire signal filter is not in the equity names. It is in the bond ETFs.

USIG (iShares Broad USD Investment Grade Corporate Bond ETF) scores $95$. VCLT (Vanguard Long-Term Corporate Bond ETF) scores $95$. ISTB (iShares Short-Term Corporate Bond ETF) scores $92$. VTC (Vanguard Total Corporate Bond ETF) scores $92$. Four investment-grade bond ETFs triggering high-conviction buy signals on the same session that $24.37$ of equities print new 20-day lows is not coincidental.

This is a flight-to-quality signal. It is explicit in the data. Capital is rotating out of equities with deteriorating momentum and into investment-grade fixed income. The USIG and VCLT scores of $95$ are not marginal. They represent technical readings that would appear in the top tier of any cross-sectional scan regardless of the broader market environment. Against a backdrop of elevated risk pressure and record new-low breadth, their presence in the trade plan at $3.4$ allocation each is the system's acknowledgment that duration and credit quality are outperforming in the current regime.

The inclusion of bond ETFs in a systematic equity trade plan — at non-trivial allocation weights — is the trade plan communicating what the equity signals are only implying: this is a defensive rotation.

Sell-Side Concentration: Fintech and Consumer Platforms

The bearish side of the signal filter is as thematically coherent as the bullish side. Five Point Holdings (FPH), Icahn Enterprises (IEP), and Kodiak AI (KDK) each score $20$ with five sell signals. PayPal (PYPL) scores $24$ with Ichimoku breakdown, Williams %R sell, MFI sell, and RSI centerline bear. S&P Global (SPGI) scores $24$. Affirm Holdings (AFRM) and Booking Holdings (BKNG) score $28$.

The composition of this list — PayPal, Affirm, Booking Holdings, S&P Global — describes a specific sector rotation. Fintech payment processors, buy-now-pay-later platforms, consumer travel intermediaries, and financial data infrastructure are all under simultaneous technical pressure. These are not speculative small-caps. PYPL has a market capitalization above $60$ billion; BKNG above $150$ billion. Their simultaneous appearance in the bottom decile of a systematic signal scan reflects genuine deterioration in intermediate-term price structure, not noise.

The ichimoku_cloud_breakdown signal appears across multiple bearish names. Ichimoku cloud breakdown — where price closes below the Kumo cloud — is a structurally meaningful signal because the cloud incorporates both momentum (Tenkan/Kijun lines) and support/resistance structure (the Senkou spans). A confirmed cloud breakdown in names like PYPL and SPGI means the multi-dimensional Ichimoku structure has failed, not just a single moving average.

Energy Build-Up: Balanced Tension

The volatility energy scan identifies $44$ symbols with energy scores above $85$ and explosion probability above $80$. Among the top $40$ names by energy score, the directional bias is nearly symmetrical: $18$ upside, $17$ downside, $5$ neutral.

This is a compressed market waiting to move in two directions simultaneously across different names — which is exactly what a defensive rotation looks like from the perspective of energy accumulation. The speculative and higher-beta names are coiling toward downside: BKKT (energy $98.43$, explosion probability $98.67$, downside bias), BLNK (energy $96.62$, probability $97.13$, downside bias), ARWR (energy $95.45$, probability $96.13$, downside bias). These are names with thin or deteriorating fundamental anchors, and the energy accumulation is skewed toward resolution lower.

The large-cap names on the energy watchlist are telling a different story. Bristol-Myers Squibb (BMY) shows energy of $94.13$ with upside bias — a defensive healthcare name coiling for a potential breakout. Academy Sports and Outdoors (ASO) shows energy of $94.54$ with upside bias. Archer-Daniels-Midland (ADM) shows energy of $88.84$ with upside bias. ADC Therapeutics (ADCT) shows energy of $93.15$ with upside bias.

The pattern is legible: speculative and growth-oriented names accumulate downside energy; defensive, cash-flow-positive names accumulate upside energy. The energy build-up data is not producing a market-directional signal. It is producing a sector-rotation signal.

The Trade Plan Structure

The trade plan for 2026-03-10 is $30$ buy positions, $0$ sells, $0$ holds, representing a fresh portfolio build with estimated turnover of </span>332{,}500$. Top allocations of approximately$3.8%$each go to DDD, DYN, UTHR, XENE, and BILI. KC, AL, and STM follow at$3.7%$. USIG and VCLT receive$3.4%$ each.

The portfolio construction is self-consistent with the signal environment. The highest-conviction names (composite score $100$, zero sell signals) receive the highest allocations. Investment-grade bond ETFs receive material weight — not a token hedge, but a structural allocation that acknowledges the flight-to-quality dynamic directly. There are no tech-growth names at meaningful weights. There are no fintech or consumer platform names. The portfolio, taken as a whole, describes a defensive long book in a risk-off environment: pharma, China ADRs trading below intrinsic technical levels, a European industrial semiconductor, and investment-grade duration.

Air Lease (AL) at $3.7$ is the one name that does not fit a purely defensive narrative. AL is a commercial aircraft lessor — cyclically sensitive, though with long-term contracted cash flows. Its presence in the top tier of the scan despite the risk-off backdrop reflects a name-specific technical setup rather than a macro view. GE Vernova (GEV) at $88$ composite provides a similar industrial exposure at lower weight.

What the Data Describes

A $4.7:1$ ratio of new lows to new highs is not ambiguous. Neither is a breadth reading of $35.35$ above MA50. The risk pressure reading of $19.19$ has been elevated and stable — which in this context means the selling has not been a single-session event. It has been a sustained process.

The signal filter's bullish pockets — pharma, China ADRs, bond ETFs — are not evidence that the broad market is about to reverse. They are evidence of where money is going as it exits the names on the sell side. PYPL and BKNG capital does not disappear; it relocates. The bond ETF signals and the healthcare signals identify the destination.

The energy build-up data confirms the tension. $44$ symbols near explosion with a near-even split between upside and downside bias describe a market that is not uniformly breaking down — it is differentiating sharply between names, sectors, and risk profiles. The high-energy speculative names are coiling toward resolution lower. The high-energy defensive names are coiling toward resolution higher.

A market where $24$ of names print new 20-day lows can still contain genuine long opportunities. Those opportunities are not distributed randomly. On 2026-03-10, they are concentrated in exactly the sectors that benefit when risk appetite contracts: defensive healthcare, investment-grade credit, and names whose individual technical structures have decoupled from the broad tape.

The question the data does not answer is how long the rotation continues before the breadth deterioration exhausts itself into a mean-reversion bounce. In the Korean market five days ago, a 92.2% single-day advance ratio followed exactly that kind of sustained breadth compression. Whether the US cross-section is approaching that kind of reflexive inflection — or whether the deterioration has further to run — is not visible in today's signal architecture. What is visible is the direction of the rotation and the names it is favoring.

That is enough to act on.