# The Leverage Behind the Rally

*Published 2026-09-14 | tags: Workflow*

> A Stocklake walkthrough on a real portfolio: which corners of a rally run on debt rather than fundamentals. A portfolio risk read, not a trade call.

Canonical: https://www.stocklake.dev/blog/the-leverage-behind-the-rally

Most mornings that feel like “the market's down” aren't actually one thing happening everywhere. They're one or two industries rolling over hard while the tape as a whole barely moves. On the morning behind this post, VIX sat at 15.6 and the Fear & Greed Index read 34 (“fear”) - not a panic reading, nowhere close to a crash. Treating that as one downturn misses the more useful question sitting underneath it: which corners of the market actually turned, and did the stocks that ran the hardest get there on real improvement, or on debt and an already-stretched chart? That second question is what a leverage-and-momentum screen is for. It isn't a signal to buy, sell, or short anything - it's the same due diligence you'd want on any position before deciding what, if anything, to do about it.

**What this is, and isn't.** Everything below is a research and screening workflow - reading fundamentals, technicals, and sector context together through Stocklake's data, cross-referenced with a brokerage account. It is not investment advice, not a recommendation to buy, sell, or short any security, and Stocklake doesn't generate trade calls. A few figures (account balances, buying power) are described in rounded, illustrative terms rather than as a current personal snapshot. Do your own diligence, and talk to a licensed advisor before acting on anything involving real money.

## Step one: read the tape before you read a stock

The instinct on a red morning is to jump straight to a stock. The more useful first move is a level up: is this a broad move, or is one industry doing all the work? Stocklake's `get_sector_intelligence` call answers that directly - each of the 11 sectors gets a `sector_score` from 0 (Lagging) to 100 (Leading), the same continuous read we wrote about in [The Number Behind the Label](https://www.stocklake.dev/blog/macro-score-and-sector-score), not just a bucketed word.

Which sectors are actually moving today, and is this a real risk-off morning or something narrower?

Assistant

Recent

Sector rotation check

Leverage screen

Account risk check

get_market_pulse({})

get_sector_intelligence({ sort_by_strength: true })

VIX is 15.6, Fear & Greed 34 - that's cautious, not panicked. It's narrower than “a downturn”: Energy is Strong/Markup (score 63) on a mid-cap refiner rally, while Technology is Weak/Markdown (score 27) on an AI/speculative-tech unwind. Two sectors, opposite directions, same day.

## The tell: concentration, not breadth

Both readings above carry the same warning sign once you look past the headline score. Energy's cap-weighted return that month was +0.2% - basically flat - while its equal-weighted return was +5.9%. The gap is concentration: 82.5% of the sector's move sits in its top five names, led by a handful of refiners up 30-40% in a single month while the integrated majors sat still. Technology's number is more extreme in the other direction - 94.2% of the sector's cap is doing the work of masking a broader equal-weight slide underneath it. Neither “Strong” nor “Weak” is describing something broad. Both are describing a handful of names carrying (or dragging) the average, which is exactly the setup where a fundamentals-and-technicals screen on individual names is worth more than the sector label alone.

## Screening for leverage plus exhaustion

The actual filter is three conditions, cross-referenced rather than run separately: an outsized price run over the past several months to a year, a Debt/Equity ratio that's genuinely elevated (read straight from a company's own filed balance sheet, not a headline multiple), and a technical exhaustion signal that's already *completed* - a finished DeMark TD Sequential sell countdown, or an Elliott Wave five-wave-up completion - rather than one still building. Any one of those three shows up constantly on its own; all three landing on the same name at once is rarer, and it's the combination that turns “this stock had a good year” into “this stock's balance sheet is doing a lot of the lifting, and the chart is already flagging it.”

Inside that Energy strength, is it broad, or a few over-levered names dragging the average up?

Assistant

Recent

Sector rotation check

Leverage screen

Account risk check

get_screener({ sector: "Energy", ... })

get_stock({ symbol: "CVI", pro: true })

Concentrated. CVI and UGP are each up 30-40% this month alone against a sector that's cap-weighted flat - that gap is the tell. CVI's Debt/Equity is over 2.4x, its RSI is 82.7, and its DeMark TD Sequential sell countdown already completed. Same shape shows up on MPC, and on DELL over in Technology - a leveraged-buyout balance sheet with negative book equity, which is its own kind of leverage flag a plain ratio can't even express.

Sorting the results into two groups makes the pattern legible: names where all three conditions are fully met, and names that are close but missing one leg - either the technical signal is still building rather than complete, or there's a real offsetting factor (genuine earnings strength, or leverage that's only moderate) that weakens the “debt is doing the work” thesis.

Tier 1 - all three conditions met

Outsized run, genuinely elevated leverage, and a completed exhaustion signal - not one still building

[CVI](https://www.stocklake.dev/stock/CVI), [MPC](https://www.stocklake.dev/stock/MPC), [DELL](https://www.stocklake.dev/stock/DELL), CFLT

Tier 2 - two of three, or an offsetting factor

Signal still building, leverage only moderate, or the run is backed by real earnings rather than pure re-rating

[VLO](https://www.stocklake.dev/stock/VLO), [DINO](https://www.stocklake.dev/stock/DINO), [FRO](https://www.stocklake.dev/stock/FRO), [UGP](https://www.stocklake.dev/stock/UGP), QRVO, [SWKS](https://www.stocklake.dev/stock/SWKS)

A tiering, not a ranking - Tier 2 isn't “worse,” it's missing one leg of the same three-part filter

| Symbol | Sector | Run | Debt/Equity | What tripped it | Tier |
|---|---|---|---|---|---|
| CVI | Energy · refining | +153% off its low | ~242% | TD Sequential sell countdown complete | Tier 1 |
| MPC | Energy · refining | +126% (12M) | ~133% | Elliott Wave 5-up complete; insider selling | Tier 1 |
| DELL | Technology · hardware | +358% (12M) | Negative (LBO balance sheet) | TD Sequential sell countdown complete | Tier 1 |
| CFLT | Technology · software | Overbought scan | ~94.5% | TD Sequential complete; negative margin at ~48x fwd P/E | Tier 1 |
| UGP | Energy · refining/distribution | +109% YTD | ~98% | Sell setup complete, countdown 8 of 13 - not finished | Tier 2 |
| VLO | Energy · refining | +150% (12M) | ~40% | Sell setup only, countdown 2 of 13 - real ROE 27.6% | Tier 2 |
| DINO | Energy · refining | +114% (12M) | ~32% | Setup still building, not complete | Tier 2 |
| FRO | Energy · shipping | +114% (12M) | ~77% | Sell setup, countdown 3 of 13 - backed by real earnings (54.8% margin) | Tier 2 |
| QRVO | Technology · semiconductors | +34% (12M) | ~44.6% | RSI 83.8, above upper Bollinger Band - leverage only moderate | Tier 2 |
| SWKS | Technology · semiconductors | RSI 83.4 | ~11.9% | Elliott Wave 5-up complete - leverage case basically absent | Tier 2 |

Ten names from the same screen, priced the week of this post - a risk map, not a watchlist to act on

Debt/Equity vs. 12-month price change

The five names with a clean, matched 12-month figure - CVI, UGP, SWKS and CFLT aren't plotted here since their run was measured over a shorter or differently-dated window and would misstate the axis if forced onto it. DELL is excluded for a different reason: its book equity is negative, so there's no ratio to place on this axis at all.

MPC sits alone in the upper-right quadrant - the biggest run *and* the most leverage of the five plotted here

## Why the sector's own direction matters

Three of the four Tier 1 names - CVI, MPC, and the Tier 2 refiners around them - aren't just *in* Energy, they're the reason Energy reads Strong/Markup at all. That's a materially different situation from CFLT and DELL, which are extended inside a sector that's already reading Weak/Markdown. A name running hot against a sector that's already rolling over is one thing; a name running hot *while being the entire reason its sector looks strong* is another - that's a crowded trade with real squeeze risk if the underlying commodity or crack-spread story keeps running, not a quiet fade. The sector score isn't a footnote to the stock-level screen. It's telling you which of these ten names are fighting their own sector's trend and which are riding it.

The same leverage question can be checked a second, independent way: Stocklake's `forensic_scores` block runs the published [Altman Z-Score](https://doi.org/10.1111/j.1540-6261.1968.tb00843.x), [Piotroski F-Score](https://doi.org/10.2307/2672906), and [Beneish M-Score](https://doi.org/10.2469/faj.v55.n5.2296) formulas straight off each company's own filed statements - a name sitting in Altman's distress zone with a weak Piotroski band is the accounting-side version of the same story a Debt/Equity ratio and a completed exhaustion signal are telling on the technical side. We wrote up how those three formulas actually work, and the five places they mislead, in [a separate post](https://www.stocklake.dev/blog/forensic-accounting-scores-api).

Given that screen, how much cushion do I actually have in my account right now?

Assistant

Recent

Sector rotation check

Leverage screen

Account risk check

get_account_summary({})

Comfortable buying power headroom, and account leverage sitting right around 1.0x - no forced-selling risk anywhere close. That's worth knowing on its own, independent of anything in the screen above: you have room before you have to decide anything.

That's the actual shape of a portfolio risk review: not “what should I trade,” but “where is risk concentrated in the market right now, is it broad or a handful of levered names carrying it, and how much room do I have before any of that touches me.” Three genuinely different data sources - a sector regime pipeline, a company's own filed balance sheet plus its technical chart, and a real brokerage account - answering one thread of questions instead of three separate ones.

## What this replaces

Done by hand, the same morning looks like this: a chart platform for RSI, Bollinger position, DeMark, and Elliott Wave; a separate fundamentals lookup (or a 10-K) for Debt/Equity on each name; a sector-rotation service, if you subscribe to one, to know whether a stock's strength is genuine or just riding its sector; and a tab-switch to your broker's own app to check buying power and leverage before any of it becomes relevant to your own account. Four or five surfaces, none of which know about each other, and the same ticker re-typed into each one. Wired into Claude through Stocklake Pro - and, if you use one, a brokerage connector - it's the same conversation: plain questions, cross-referenced answers, and nothing you have to reconcile by hand across tabs.

$20/month

· 5,000 calls/day · all 20 tools, including sector intelligence, fundamentals, and forensic scores

7-day trial - every Pro tool is usable before your card is charged, cancel any time during the trial window.

## Say it twice: this is not a trade idea

Every number above is real, sourced from Stocklake's own pipeline the week this was written - and every number above is also a snapshot that will be stale by the time you read it. Debt/Equity updates on a company's own filing cadence, not daily; a completed exhaustion signal describes a chart pattern, not a guaranteed reversal; and a sector that's Strong today can be Neutral by Friday. None of the ten names in that table is a recommendation to buy, sell, or short anything, and nothing here should be read as a signal that a reversal is imminent on any of them. What you do with a risk map like this - tighten a stop on something you already hold, size a hedge, or just know where the crowded, levered corners of the tape are - is entirely your own call, and one worth making with a licensed advisor if real money is involved. Stocklake's job stops at giving you the data to make it with.

## Frequently asked questions

Is a leverage-and-momentum screen like this a trading recommendation?

No. It's a way of reading which recent gains are backed by leverage and technical exhaustion versus real improvement. Stocklake surfaces data - fundamentals, technicals, sector context - and doesn't tell you what to buy, sell, or short. What you do with a screen like this, if anything, is your own decision.

What data actually goes into a screen like this?

Three layers, cross-referenced: sector and macro context (is a whole industry running hot, and how concentrated is that strength), fundamentals (Debt/Equity and other balance-sheet ratios read straight from a company's own filed statements), and technical exhaustion signals (RSI, Bollinger Band position, DeMark TD Sequential, and Elliott Wave completion). None of these alone tells the full story; together they show whether a rally is broad and earned or narrow and leveraged.

How often do sector scores and Debt/Equity ratios actually update?

Sector scores refresh roughly every four hours through Stocklake's market intelligence pipeline - often enough to catch a real regime shift, not so often you're chasing noise. A company's Debt/Equity and the rest of its balance sheet only change when it actually files a new period, so Stocklake recomputes those the same day a new filing lands rather than on a fixed schedule. A stock's sector context can move within the same afternoon; its leverage ratio generally won't.

What's the difference between a technical signal that's complete versus one that's still building?

A DeMark TD Sequential reading has two phases: a Setup (nine bars building toward a possible turn) and a Countdown (up to thirteen bars confirming it). “Sell setup complete, countdown 8 of 13” means the early warning fired but the confirming phase hasn't finished. A completed countdown means both phases ran their full course. This screen only counted the second kind as tripping the technical-exhaustion condition - a setup on its own is common enough that treating it as equivalent would flag far more names than the pattern actually supports.

Can a screen like this be run on any sector, not just Energy and Technology?

Yes. Every sector Stocklake tracks gets the same `sector_score`, cycle stage, and concentration read - Energy and Technology just happened to be the two pulling in opposite directions the morning this post is based on. Swap the sector filter and the same three-part screen runs against whichever industry is showing the most extreme concentration that day.
