🔥 The Heat Sheet

AI-powered stock intelligence · StockArkalytics.com

ISSUE 013

APR 18, 2026

You're reading a free issue of The Heat Sheet

Get it in your inbox every week — AI-powered signals, volume leaders, unusual options flow.

Ising, PDT, and the Avis Squeeze

NVDA's quantum bomb lit the fuse. The SEC gave retail traders back their day trades. Short sellers ate $2.5B in Avis losses in ten sessions. 13 names hit the scanner, 8 more flashed bullish divergence. Here is what the tape is telling us — and who is actually worth watching into next week.

The most intense story on our scanner this week. XNDU is a Toronto-based photonic quantum play that only started trading on NASDAQ and TSX on March 27, 2026 via its SPAC merger with Crane Harbor Acquisition Corp. That transaction brought in ~$302M of gross proceeds per Investorideas. Three weeks later, NVDA dropped Ising — and Xanadu, as a photonic qubit shop whose PennyLane software is used by roughly half of all quantum developers worldwide, was positioned for maximum leverage to the narrative.

What happened on the tape: per Investing.com, +77.5% on April 15. Another ~33% surge on April 16 closed around $33.59 on ~9.7M shares. IBTimes Australia tags the stock at +194% YTD with a market cap of ~$7.5B.

The counterweight: FY2025 revenue of ~C$3.45M against a C$58.49M net loss per IBTimes. Invezz flagged XNDU as a potential short candidate, citing the disconnect between $4.6M in revenue and a >4x-since-mid-March stock move. Morningstar rates it at a 893% premium to fair value.

Photonics · Newly Public (SPAC) · PennyLane Software · Narrative Stock · High Volatility

The sector's institutional darling and the closest thing quantum has to a bellwether. Per The Motley Fool, IONQ closed April 15 at $43.25, +20.95% on the session. TipRanks reported +19.2% on World Quantum Day itself.

The fundamentals actually support the rally better than XNDU's do. IONQ's 2025 revenue hit $130M, up 202% YoY per the company. 2026 guidance is $225-245M, implying ~81% YoY growth. Per TradingKey, IonQ ended 2025 with ~$3.3B in cash — more than enough runway to weather another year of widening losses (the 2025 GAAP net loss was $510.4M, and 2026 EBITDA losses are guided to widen to $310-330M).

TipRanks shows 9 Buy ratings and 3 Hold over the trailing three months, average price target $65.91 — implying ~84% upside from pre-rally levels. The consensus is a "Strong Buy."

Trapped-Ion · Revenue Scaling · Institutional Favorite · Cash-Rich

The annealing-specific play had arguably the cleanest technical move of the week. Per The Motley Fool, QBTS closed April 15 at $20.81, +22.63% on the day — its biggest single-session gain of 2026. The real tell was the volume: 90.2M shares, roughly 227% above the three-month average. That is institutional participation, not a retail flash mob.

The Motley Fool notes D-Wave has grown 105% since its 2020 IPO, which belies just how much time the stock has spent underwater. The counter-narrative: Schwab Equity Ratings still had QBTS at an "F" rating (Strongly Underperform, 97th percentile) as of March 20, 2026, pre-rally. Ising does not change the underlying revenue trajectory, but it does buy the sector months of sentiment runway.

Annealing · Volume Spike · Beta 3+ · Schwab F-Rating

Per The Motley Fool, RGTI closed April 15 at $19.11, +13.28% on the day — TipRanks had it at +12.4% on World Quantum Day. The superconducting qubit shop had a unit physically on display at the NVIDIA GTC booth in Washington DC in October 2025, which matters more than it sounds: Ising's architecture is modality-agnostic, but the relationships NVDA chose to cultivate on-stage are the ones most likely to see early real-world co-engineering.

Of our four quantum names, RGTI has the tightest relationship to both the modality NVDA cares most about (superconducting, because that's what IBM, Google, and Microsoft are also building) and the partnership track record with NVDA. That makes it the asymmetric bet within the theme — less hype than XNDU, more NVDA-adjacent than IONQ or QBTS.

Superconducting · NVDA GTC Partner · Mid-Cap Quantum

💡 Hot Take — Quantum Wave 2.0

NVDA monetizing quantum without building a quantum computer is the most NVDA move imaginable. Every qubit shop now needs their calibration and decoding tooling, which means every qubit shop is now more GPU-bound than they were two weeks ago. The rally is real. The revenue for most of these names still isn't.

Our ranking: IONQ (best fundamentals, scale cash, consensus Strong Buy) → RGTI (best partnership optics, cleanest NVDA-adjacency) → QBTS (strongest technical setup with 227%-of-average volume) → XNDU (purest momentum trade, highest volatility, most extreme valuation mismatch at 893% premium per Morningstar). Position size accordingly.

The stock where the scanner and the signal agree. HOOD showed up on both our most-active scanner AND our bullish divergence radar in the same week. That kind of confluence does not happen often; when it does, it earns the ⭐.

The setup: Per Parameter, HOOD has bounced approximately 33% off March lows and traded recently around $71.67 — still 53% below its 52-week high of $153.86, down 23% YTD. Bernstein's Gautam Chhugani kept an Outperform rating with a $130 price target, implying ~50% potential upside. Bernstein's 2026 crypto revenue forecast runs 31% above consensus, projecting $1.1B in crypto revenue (+23% YoY), with prediction markets modeled to grow 286% to $586M.

Why the divergence matters: In technical terms, price has been making lower lows while momentum has stabilized — the classic Fischer Transform setup that our Live Options Trader Analysis flags. Combine that with the PDT-repeal tailwind and Q1 2026 earnings on April 28, and you have three catalysts stacked into the next 10 sessions.

Context: 2025 full-year revenue came in at $4.5B (+52% YoY) per TipRanks with adjusted EBITDA of $2.5B (+76%) at a record 56% margin. Full-year EPS of $2.05. Q4 2025 revenue of $1.3B (+27% YoY). Per company filings, Bitstamp (the institutional crypto exchange Robinhood acquired last June for $200M) has roughly doubled its volumes since close and now represents ~60% of total crypto transaction volume.

PDT Repeal Winner · Earnings Apr 28 · Bullish Divergence · Bernstein $130 PT

The higher-beta PDT play. Per Invezz, Webull jumped 13% on Wednesday April 15 to its highest price since March 5, breaking out of a giant falling wedge pattern and pushing above the 50-day EMA with rising RSI and MACD. That is a textbook technical setup: descending, converging trendlines resolving to the upside right as a major fundamental catalyst lands.

Two catalysts in one week, not one: Beyond the PDT repeal, Webull separately announced the termination of a restrictive financing agreement that had weighed on sentiment for months. 24/7 Wall St. flagged both catalysts as contributing to BULL outpacing HOOD's move — BULL up 8% versus HOOD's 7% on the news session.

Fundamentals: Per company filings, 2025 revenue of $571M, up 46% YoY. Q4 2025 revenue of $165.2M, up 53%. Adjusted operating profit of $110M. Funded accounts grew from 4.66M to 5.03M, with 26M total registered users across 14 international markets. The forward P/E is ~21 per Invezz, versus HOOD at ~32 — BULL is the value option within the theme.

The analyst setup: MarketBeat tracks five analysts at Moderate Buy with a consensus $13 average target (targets range $9-$18) — implying 120%+ upside from current levels near $6. Rosenblatt Securities maintained Buy at a $12 PT, valuing the stock at 20x 2027 adjusted EBITDA.

The risk: BULL is down 91% year-over-year per Stocktwits. The stock traded as high as $62.90 12 months ago. That kind of chart does not reverse in one week — but it does create the conditions for an extended short squeeze if the technical breakout holds and the fundamental tailwind from PDT repeal turns into actual trading activity growth.

Small-Cap Broker · Falling Wedge Break · 26M Users · Deep Discount to HOOD

💡 Hot Take — Retail Trading Reboot

The PDT repeal is the most important structural regulatory change in retail trading in a decade. Every sub-$25K margin account in America just went from "can trade three times a week" to "can trade as often as they want, subject to margin." Bernstein's $240B-to-$1T prediction-market forecast is the frame.

HOOD is the quality play (scale, brand, the Bitstamp acquisition driving institutional crypto, cross-signal confluence with our divergence radar, earnings April 28). BULL is the leveraged call option (higher beta, wedge breakout, deep discount to HOOD on forward P/E, 120%+ average analyst upside, and the potential for a violent squeeze if the chart confirms). Own both if you have the risk budget. Own HOOD if you have to pick one.

What it is: An Amarillo, Texas-based REIT (per Yahoo Finance) building grid-independent, gigawatt-scale power campuses specifically for AI-centric tenants. Only 35 employees per TradingView, but with deep Texas grid connections and long-term leases as the business model.

The tape: FRMI bounced from a 52-week and all-time low of $4.47 on April 9 back above $6 by April 16, per StocksToTrade. StockTitan notes the 2026/03/30 selloff came on the same session that crude prices ripped higher on Hormuz tensions — which was odd, because higher oil should normally support energy equities. That disconnect is the setup: FRMI is not trading as an oil-price play; it's trading as a pure AI-power-thesis narrative stock.

Fundamentals (pre-revenue): Per StockTitan, TTM net income of -$486.4M and TTM EPS of -$1.13, operating cash flow -$34.2M, and operating income -$177.8M. Balance sheet per TimothySykes: ~$408M cash against ~$110M long-term debt with ~$1.10B equity. Heavy cash burn, but real runway. Market cap ~$3.04B per Investing.com.

The catalyst on the calendar: Project Matador groundbreaking on July 4, 2026. Per Investing.com, 9 analysts have it at Strong Buy with an average 12-month price target of $23.11 (range $8-$35) — implying significant upside from the current ~$5 range, but with a "Strong Sell" daily technical signal based on moving averages. The spread between the analyst fundamental view and the near-term technical view is the opportunity.

AI Power REIT · Private Grid Campus · Project Matador · Pre-Revenue

Returning guest. We flagged IREN in Issue 07 as part of the bitcoin-miner-to-HPC pivot thesis, and it keeps making our scanner. The setup is simple: IREN already has the land, grid interconnects, substations, and cooling infrastructure that hyperscalers need 24 months to permit and 36 months to build. That is the entire asset.

The differentiator vs. FRMI: IREN is operating today — revenue is real, kilowatts are flowing, and the question is mix shift between BTC mining revenue and HPC leasing revenue over the next 4-6 quarters. FRMI is a blank-sheet developer betting on 2027+ leases. Different risk profiles, same underlying macro thesis.

The dual-fuel thesis: IREN benefits whether Bitcoin or AI wins the power auction — and right now, both are bidding. That gives the stock a volatility profile that correlates only loosely with either underlying asset.

HPC Pivot · BTC + AI Dual-Fuel · Operating Today · Issue 07 Repeat

💡 Hot Take — AI Power Pipeline

We have been saying this since Issue 07: the AI trade is migrating from silicon to substations. That migration is not done — it's accelerating as Meta, Microsoft, and Google run out of places to put gigawatt-scale AI campuses in reasonable time frames.

FRMI is the developer play (optionality-heavy, catalyst-driven around Project Matador, pre-revenue). IREN is the operator play (revenue today, mix shift as the thesis). Own the developer for the asymmetric catalyst-driven upside; own the operator for the repeatable, cash-generating thesis. Don't own either if you can't stomach 30% drawdowns on Hormuz-related tape.

The trade of the week — possibly of the year. Per StocksToTrade: CAR closed near $99.90 on March 20. By April 14 it was at $411.56. By April 16 close: $448.98, with an intraday high just under $452. That is a more than 4x move in 17 trading sessions. GuruFocus flagged a 52-week peak of $450.00 with a market cap of $15.54B.

The mechanics: Per Bloomberg and S3 Partners, short interest is at 58% of free float — the highest in the past 10 years. Short sellers clocked in nearly $2.5 billion in mark-to-market losses over the 10 trading sessions through April 14. And despite that, bears are doubling down — S3's Ihor Dusaniwsky confirmed short interest was still climbing through the rally.

The narrative: A federal funding lapse left TSA officers working without pay in February. Security lines ballooned to 3-4.5 hours at major hubs per Yahoo Finance. Travelers abandoned flights for rental cars. Pentwater Capital Management disclosed a sizable stake before the rally took off. Result: textbook short-squeeze mechanics on top of a real (if temporary) demand catalyst.

The valuation disconnect: Per Bloomberg, only 1 of 8 analyst ratings is a Buy. Deutsche Bank's Chris Woronka downgraded to Hold with a $128 price target. Wall Street consensus target: $106.43. Current price: ~$448. That is over 3x the mean analyst target. Avis reported a $995M net loss last year and carries $25B+ in debt against ~$500M cash.

The warning: Per Vanda Research's Ashwin Bhakre, retail flows turned "aggressively negative" in April with daily outflows of $10-12M — some of the largest on record for the name. Miller Tabak's Matt Maley described the action bluntly: this is being caused by short-covering, not improving fundamentals. When the short-covering stops, the tape reverses. CAR fell as much as 19% on April 16 before paring losses per Bloomberg — a preview.

Short Squeeze · 58% SI · Retail Flows Negative · 3x Analyst Target

Chinese wealth-management and insurance brokerage mid-pivot into AI. Formerly Highest Performances Holdings / Puyi Inc. Per company filings, on March 31, 2026, MAAS closed its strategic acquisition of Huazhi Future (Chongqing) Technology — a computing-power and algorithm-solutions provider serving public security, firefighting, agriculture, and enterprise digital transformation in China. Management framed this as a pivot from "Scenario Operator" to "AI Industry Player."

The tape: Per IBTimes Australia, +17% on April 15, closing around $7.88, up >45% YTD in some tracking periods. But the stock has traded as low as $2.41 over the past year and remains well below its $14.00 52-week high. Morningstar (as of early March) rated MAAS at a 517% premium to fair value.

The risk: This is a Chinese small-cap in the middle of multiple reverse-merger-style acquisitions (also tea, drinking-water pipes, and new-energy tech). The narrative is moving faster than the financials. Treat as pure speculation.

China AI Pivot · Reverse Merger · Extreme Volatility · 517% Morningstar Premium

The AI interconnect silicon play. CRDO makes high-speed SerDes (serializer/deserializer) chips and Active Electrical Cables (AECs) that move data between GPUs inside hyperscaler datacenters. As rack densities scale from 72 GPUs to 576+ in NVL72 and NVL576 configurations, the interconnect silicon scales non-linearly. It is a pure pickaxe for the GPU rush — every incremental GPU deployed increases CRDO's TAM.

Two scanner hits this week align with broader datacenter infrastructure flows. The name benefits from every catalyst that benefits NVDA's hyperscaler customers — and doesn't get the valuation multiple NVDA carries.

AI Interconnect · SerDes Silicon · Hyperscaler Beta · Pickaxe Trade

Direct-to-consumer telehealth platform that's been turning into a serial category expander — weight management (GLP-1), dermatology, mental health, hormonal therapy. Scanner hits driven by continued volatility around category-launch news flow and margin leverage storytelling. HIMS is one of those names where the bull case and bear case can both stay alive for months because the underlying growth keeps surprising and the regulatory framework for telehealth keeps shifting.

DTC Health · GLP-1 Exposure · Category Expansion · High Beta

Zinc-based battery storage company. The thesis: grid-scale energy storage is the second-order beneficiary of both the AI-power buildout and the renewables pipeline, and zinc chemistry offers a safety and sourcing profile that lithium can't match. Per StocksToTrade, the scanner hits this week align with preliminary Q1 revenue and capacity-ramp commentary from management.

EOSE reads as the connective tissue between Theme 3 (AI Power) and the broader industrial grid-buildout trade. When FRMI eventually leases gigawatts of campus capacity, someone has to sell the batteries for the backup and time-shifting. Small cap, volatile, but the strategic logic is clean.

Zinc Battery Storage · Grid-Scale · Domestic Supply · AI Power Adjacent

💡 Hot Take — Squeeze & Spec

This theme is a reminder that volume and volatility do not equal investment thesis. CAR is one trade. MAAS is another. Neither is a multi-month position for anyone who hasn't already sized accordingly.

The takeaway: CRDO is the real fundamental name in this block (AI interconnect = pick-and-shovel play). EOSE is the thesis-driven small cap (grid storage = second-order AI power trade). CAR and MAAS are the trading instruments — fine for short-dated exposure, not fine for "buy and forget." HIMS is the category question mark everyone has an opinion on. Know which bucket you're in before you click.

★ Cross-Signal Pick of the Week: HOOD

When a name lands on our most-active scanner AND our divergence radar in the same week, that is not coincidence — that is confluence. HOOD's divergence setup lines up with a 33% bounce off the March lows, a structural regulatory tailwind (PDT repeal), and Q1 2026 earnings on April 28.

Technical entry windows, Fischer Transform chart, and options flow detail are on the platform. Read the full HOOD card in Theme 02 above for the fundamental context.

Enterprise AI software platform. The namesake ticker. Persistent divergence while the stock has traded sideways-to-down suggests enterprise-software bid is building under the surface — even as narrative attention rotates to infrastructure names.

SMB accounts-payable fintech. Beaten-up growth name with real revenue and recurring cash flows. Divergence setup often precedes a bounce in this kind of chart — high short interest, bombed-out sentiment, stable unit economics.

The "LinkedIn for doctors." High-margin networked healthcare play trading at a discount after a rough stretch. Divergence signal suggests accumulation underway.

Sports data + betting integration. Partnerships with NFL, NBA, Premier League put the company in the toll-booth position on a growing vertical. Divergence as the stock digests a long 2025 run.

Specialty insurance holdco often called "Baby Berkshire." When this name shows divergence, it's typically a high-quality entry — MKL trades by institutional book, not by meme.

Social platform + AI-data-licensing dual engine. Stock chopped through Q1 but the licensing economics keep improving. Divergence setup worth tracking into the next earnings window.

Programmatic ad-tech leader. Got punished through 2025 on competitive-positioning concerns. Divergence here coincides with a repositioning narrative on CTV and retail-media that may finally be getting traction.

🔒 Ultra Plan Exclusive

See the live Fischer Transform chart, exact divergence timestamps, and entry windows for all 8 names on the platform.

💡 How We Use This List

Divergence signals are not buy triggers. They're watchlist builders. The pattern says "momentum is leading price" — which often precedes a reversal, but not always, and not always on the timeline you want. We use divergence to surface candidates, then layer in fundamentals (like HOOD's earnings setup) and flows (like CRDO's options activity) before anything gets sized. The platform's unusual options flow is the next filter we'd apply here.

StockArkalytics

ARKA — A RAY OF LIGHT ON THE MARKET

Quick Links

Blog  ·  Pricing  ·  Dashboard