onsemi Prices the Old Auto Cycle, Not the AI-Power Cash Reset

Summary: onsemi (Nasdaq: ON) closed at $80.40 at 16:00 EDT on August 3, 2026, equivalent to 2026-08-04 04:00 Singapore time, before its after-close second-quarter release. The reference price was 40.4% below the Yahoo Finance 52-week high of $134.92. The new filing reports Q2 revenue of $1.604 billion, non-GAAP gross margin of 39.3%, Q2 free cash flow of $425.4 million, and an expected 2026 doubling of AI data-center revenue. The setup is that the market may still be carrying the old automotive and industrial downcycle into a materially better cash-flow print.

This is not a clean AI narrative. First-half operating cash flow fell to $698.8 million from $786.6 million. The company repurchased $332 million of stock in Q2 while a new $1.5 billion zero-coupon convertible issue helped fund a portion of the first-half capital return. The pending all-stock Synaptics transaction is expected to leave Synaptics stockholders with approximately 12% of the combined company, and the treatment of Synaptics’ existing debt after closing remains unknown. The long case therefore requires an AI-power, margin, cash-conversion, debt, and share-count bridge.

This is a conditional U.S.-listed long, not an executable entry. The $80.40 close is a reference_price, not a verified post-release bid or offer. Live spread, depth, venue status, same-session exit liquidity, short interest, options, dealer flow, and fund flow were not independently verified. The signal is fail-closed with entry.price: null and execution.can_execute: false.

This run is limited to the U.S. market and long only. I scanned the current articles/2026-08/week-1/ folder and the repository by ticker, company, and mechanism. The June ON articles covered the Synaptics merger spread and the first announcement-day buyback interpretation. This article is materially different: it is triggered by the August 3 Q2 filing and tests end-market growth, AI data-center disclosure, free cash flow, debt maturities, convertible dilution, and post-deal share count. Creative search lanes included “AI data center revenue more than double working capital convertible notes,” “quarterly FCF quadrupled all-stock acquisition dilution,” “post-earnings semiconductor rebound cash conversion debt,” and “buyback authorization remaining zero-coupon notes due 2027.”

Why This Is the Best Opportunity Right Now

The strongest rebuttal is that the market is correct to discount the release. “AI data center” is the fastest-growing business, but the company does not disclose a standalone dollar figure for it. AI is included inside the broader “Other” end-market category. The Q2 cash-flow surge may be a one-quarter working-capital reversal, while first-half operating cash flow still fell. An all-stock acquisition can dilute existing holders before it proves its value, and the 0% notes create a separate conversion and warrant path.

The reason to keep the long case alive is narrower. The fresh filing arrived after the last regular-market price. It shows a company moving from a cyclical recovery story to a measurable cash and margin event: Q2 revenue grew 9%, Q2 free cash flow reached 26.5% of revenue by calculation, non-GAAP operating margin was 20.8%, and Q3 guidance calls for 40.0% to 42.0% non-GAAP gross margin. The market has not yet demonstrated how it values that combination. This is an inference from the filing and the pre-release quote, not proof that the next session will rise.

Opportunity Ranking

Rank	Candidate	Current reference and fresh evidence	Catalyst and more-than-5% path	Main reason to reject
1	onsemi, ON	$80.40 at the August 3 close, 40.4% below the 52-week high. The August 3 8-K reports Q2 FCF of $425.4M, AI data-center revenue expected to more than double in 2026, and Q3 gross-margin guidance above 40%	The first post-release session can reprice a pre-release quote by more than 5% in either direction. The next 10-Q can adjudicate AI mix, margin, cash flow, conversion, and buyback execution	Synaptics dilution, $804.9M of 0% notes due 2027, weak first-half operating cash flow, and unverified live execution data
2	Clorox, CLX	$98.26 at the August 3 close after FY26 results. FY27 guidance implies organic sales growth of 3.5% to 4.5% and adjusted EPS of $5.70 to $6.00	A clean organic-growth and margin recovery could support a more-than-5% re-rating	FY27 growth is heavily affected by the GOJO acquisition, FY26 operating cash flow fell about 38%, and the release was not a large price dislocation
3	Caterpillar, CAT	$830.03 at the August 3 close. Q2 results were scheduled for 05:30 CDT on August 4, later in the U.S. session	A margin, backlog, or cash-flow surprise can move a cyclical industrial by more than 5% before a new filing is fully absorbed	The result was pending, so the evidence base was a binary event calendar and older filings rather than a fresh operating disclosure
4	Palantir, PLTR	$125.65 after a fresh Q2 release showing 93% revenue growth, $1.220B adjusted FCF, and raised FY26 guidance	The result can still produce a more-than-5% move, but the catalyst is already public	Strong operations are offset by a full valuation and a weak long-side payoff asymmetry; the price is not presenting a clear cash-conversion discount

Selected opportunity: Conditional long ON common stock.

Why this one now: It is the only candidate with a fresh after-close primary filing that simultaneously changes the revenue mix, margin, free-cash-flow, capital-return, debt-conversion, and acquisition-denominator evidence. The price reference predates the filing, so the next session supplies a dated market test.

Why This Can Jump Or Dump More Than 5% Soon

Why it can jump or dump more than 5% soon: The immediate trigger is the first U.S. session after the August 3 release. An upward move requires the market to accept the Q3 margin and AI-power evidence while treating the Synaptics issuance as manageable. A downward move requires only a focus on first-half cash-flow weakness, the 12% pro forma dilution, convertible supply, or the absence of a dollar AI disclosure. Both paths are observable within one to three sessions; the operating adjudicator is the next quarterly filing.

What Should Surprise the Reader

What should surprise the reader: Q2 free cash flow and first-half operating cash flow point in opposite directions. The company can report a very strong quarter while the half-year cash bridge is still weaker. The real question is not whether AI data-center revenue is growing. It is whether that growth survives end-market mix, gross-margin conversion, working capital, debt maturities, acquisition dilution, and actual share retirement.

The Setup

The August 3 Form 8-K and Exhibit 99.1 report Q2 revenue of $1.604 billion, up 9% year over year. GAAP gross margin was 38.4%, non-GAAP gross margin 39.3%, GAAP operating margin 16.1%, and non-GAAP operating margin 20.8%. GAAP diluted EPS was $0.56 and non-GAAP diluted EPS $0.74. The company reported Q2 free cash flow of $425.4 million, which quadrupled year over year, and said cash from operations increased 150% year over year. The Q2 Form 8-K and Exhibit 99.1 the earnings release exhibit

The end-market table is useful, but not as clean as the headline. Q2 automotive revenue was $781.3 million, up about 6.6% year over year. Industrial revenue was $422.7 million, up about 4.1%. “Other,” which includes computing including AI data center, consumer, networking, and communications, was $399.5 million, up about 21.3%. The filing does not separate AI data-center revenue from that wider category. This supports a real mix improvement, but it does not prove that the entire 21.3% increase is AI.

Management says AI data center remains onsemi’s fastest-growing business and expects its revenue to more than double in 2026. It also announced strategic AI data-center wins, an expanded role in the NVIDIA MGX ecosystem, and new GaN power products. These are company disclosures, not independent demand verification. The next filing must show whether the AI contribution is large enough to offset weakness elsewhere.

The Q3 outlook is a dated operating test: revenue of $1.650 billion to $1.750 billion, GAAP gross margin of 39.9% to 41.9%, non-GAAP gross margin of 40.0% to 42.0%, and non-GAAP diluted EPS of $0.81 to $0.93. The midpoint implies sequential revenue growth of about 6.0% from Q2 and a non-GAAP gross-margin midpoint of 41.0%, above the Q2 39.3%.

The balance sheet prevents a simple “AI growth plus buyback” conclusion:

Cash and cash equivalents were $3.5145 billion at July 3, 2026. Cash, cash equivalents, and restricted cash in the cash-flow statement were $3.5159 billion.
Gross debt maturities totaled $4.5049 billion, including $804.9 million of 0% notes due in 2027, $700.0 million in 2028, $1.5 billion in 2029, and $1.5 billion thereafter. These are principal maturities, not a net-cash calculation.
First-half operating cash flow was $698.8 million, down from $786.6 million. The company attributes the decrease primarily to unfavorable working-capital timing.
Q2 repurchases were $332 million. The cash-flow statement records $690.5 million of common-stock repurchases for the first half. The new share-repurchase program has a $6.0 billion ceiling through December 31, 2028, with approximately $5.3 billion remaining as of July 3. Authorization is potential demand, not an executed price floor.
In May, onsemi issued $1.5 billion of 0% notes due 2031. The 10-Q says $331.9 million of the net proceeds was used to repurchase 3.1 million shares. That is observed debt-funded capital return, not evidence that every repurchase is debt-funded.

The Synaptics transaction is also part of the denominator. The Q2 10-Q says each Synaptics share will convert into 1.350 ON shares, with Synaptics stockholders expected to own approximately 12% of the combined company. The deal is expected to close in mid-2027, subject to stockholder and regulatory approvals. On a simple current-denominator calculation, a 12% post-close ownership stake implies approximately 53 million new ON shares against 389.3 million filed common shares. This is an analytical bridge, not the final issued share count. The 10-Q also says the ultimate treatment of Synaptics’ existing debt after closing is unknown. ON’s Q2 Form 10-Q

The Market Price

The following are completed regular-session references. The latest price predates the August 3 after-close earnings release.

Measure	Value	Interpretation
Reference price	$80.40	Nasdaq regular-market close at 2026-08-04 04:00 SGT, equivalent to August 3 at 16:00 EDT
Prior completed close	$81.61	August 3 fell approximately 1.48% from July 31
August 3 high / low	$83.66 / $78.91	The pre-release session range
August 3 volume	14.42 million shares	About 1.55x the 20-session average calculated from the same chart series
52-week high / low	$134.92 / $44.56	The reference is approximately 40.4% below the high and 80.5% above the low
Filed common shares	389.312 million	Shares outstanding as of July 29, 2026
Filing-based reference equity value	About $31.30 billion	Filed common shares multiplied by $80.40; not a fully diluted valuation

The price path is not an intrinsic-value model. ON rose from a 52-week low of $44.56 to $80.40, so “down from the high” is not the same as “cheap.” It is a way to quantify how much of the prior cycle optimism has been removed. The useful disagreement is whether the market has carried the old automotive and industrial discount into a filing that now shows a stronger power and cash-flow path.

The Positioning

The observable positioning evidence is price and volume. The stock fell from $91.06 on July 21 to $80.40 on August 3, a decline of about 11.7%, and the August 3 volume was about 1.55 times the prior 20-session average. That shows a negative-flow backdrop before the release. It does not identify the marginal seller.

The company-sponsored flow is visible but cannot be treated as a floor. Q2 repurchases of $332 million and the remaining $5.3 billion authorization are documented. The new authorization can be modified, suspended, or terminated, and the 10-Q says the program remains subject to liquidity needs and other factors.

The following were not independently verified in this run:

Live short interest, days to cover, utilization, and borrow cost.
Bid-ask spread, order-book depth, venue status, and same-session exit liquidity.
Options open interest, implied volatility, skew, and dealer positioning.
Fund flow, dealer flow, and any post-release premarket price discovery.

The result is a 3/5 positioning score. The price and issuer repurchase facts are strong. The identity and leverage of the marginal market participant are unknown. Options are not an acceptable expression without a fresh liquid chain.

The Catalyst

The catalyst path has four steps:

August 4, 2026 U.S. session: The market responds to the August 3 after-close release. This is the immediate more-than-5% move test. A strong response should hold above the pre-release range while confirming the Q3 margin and revenue outlook. A weak response that breaks the August 3 low would show that the filing did not repair the old-cycle discount.
Q3 operating delivery: The next quarterly filing should reconcile revenue, AI data-center contribution, automotive and industrial mix, gross margin, operating cash flow, free cash flow, and diluted shares against the guidance.
Synaptics regulatory path: The parties filed HSR notifications on July 17. The 30-day waiting period expires at 23:59 Eastern on August 17, unless extended by a Second Request or ended earlier. This is a dated deal-risk catalyst, not a value guarantee.
Capital structure evidence: The next 10-Q should show 0% note conversion, warrant dilution, cash, debt, repurchase execution, and any updated treatment of Synaptics debt. The final S-4 and stockholder vote will be more important than the announcement headline.

The cheapest disconfirming sequence is the first post-release market reaction followed by the next filing that discloses AI mix, margin, cash conversion, debt, actual repurchases, diluted shares, and deal progress. Management’s phrase “fastest-growing business” is not enough by itself.

The Gap

What the tape may still be pricing	What the filings say	What remains unresolved
Automotive-cycle weakness is the consolidated earnings path	Automotive revenue grew about 6.6% in Q2 and industrial revenue grew about 4.1%	Whether those gains are durable or only a cyclical trough comparison
AI is a narrative without cash conversion	Q2 FCF was $425.4M and management expects AI data-center revenue to more than double in 2026	AI data-center revenue is not separately disclosed, and first-half operating cash flow fell
A buyback can absorb the denominator	Q2 repurchases were $332M and $5.3B remains authorized	The program is discretionary, and the issuer funded part of first-half repurchases from 0% note proceeds
The Synaptics deal is only a product expansion	The exchange ratio is 1.350 ON shares per Synaptics share and Synaptics holders are expected to own about 12% of the combined company	The final issued share count, closing date, regulatory path, and treatment of Synaptics debt
Convertible notes are low-cost financing	The 0% notes due 2027 had $804.9M of principal outstanding at July 3, and holders had conversion rights for Q3 under the filing’s test	Conversion, hedge, warrant, and cash-refinancing paths can affect the denominator and liquidity

Strongest Counterparty Argument

The strongest bear argument is that the market has already learned to discount semiconductor “AI” claims. The company’s only separately quantified AI statement is a growth rate, while the revenue table combines AI data center with consumer, networking, communications, and other end markets. Q2 free cash flow may reflect timing, not a new run rate, because first-half operating cash flow is lower. The company is also paying shareholders while carrying a current 0% note maturity and preparing to issue approximately 12% of the combined equity to Synaptics holders.

That argument is coherent. It would be stronger if the next filing showed Q3 revenue below the current guide, non-GAAP gross margin below 40%, no identifiable AI contribution, continued working-capital drag, or a materially worse share-count and debt bridge. Until then, the release creates a real but bounded disagreement. It does not justify treating ON as a net-cash or buyback-floor trade.

Canonical Rubric

Price: ON is 40.4% below the 52-week high, with 1.55x prior-20-session volume, but the $80.40 reference predates the release.
Positioning: Negative price flow and executed repurchases are visible; live short, borrow, options, dealer, fund-flow, spread, depth, and exit-liquidity data are not verified.
Catalyst: The August 4 session, Q3 guidance delivery, August 17 HSR milestone, and next share-count and cash-flow filing can close or widen the gap.
Reality: Q2 shows genuine power and cash-flow improvement, but AI is not separately disclosed, first-half operating cash flow is weaker, the acquisition dilutes the denominator, and current debt maturities remain.

The Payoff Map

The payoff map is a conditional price framework, not a valuation model. The top case does not require a return to $134.92. It requires the market to accept the Q3 margin path, assign some value to AI-power demand, and keep the share-count and debt bridge from worsening. The bottom case assumes that the market treats AI as an unverified mix story while cash flow and dilution risk compound.

Price Target and Probability Map

Scenario	Probability	Target / level	Return / payoff	Time horizon	Conditions required	Evidence quality
Top case	25%	$112	+39.3%	6 to 12 weeks	Q3 revenue and gross-margin guide hold or improve, AI data-center revenue becomes more auditable, cash conversion persists, and Synaptics or note dilution does not worsen the bridge	Medium
Base case	50%	$92	+14.4%	4 to 8 weeks	The post-release tape recovers part of the old-cycle discount, Q3 delivery is near the midpoint, and first-half cash-flow weakness does not deepen	Medium
Bottom case	25%	$55	-31.6%	4 to 12 weeks	Q3 guide breaks, AI contribution disappoints, automotive or industrial revenue rolls over, working capital remains weak, or conversion and deal dilution dominate	High
Invalidation / review level	n/a	$58 with a fundamental break	Review, not an automatic stop	Any time	A close below $58 together with a guide, cash-flow, gross-margin, debt, or share-count break; a price print alone is not proof	Medium

The probabilities total 100%. The judgmental probability-weighted target is $87.75, or approximately +9.1% versus the $80.40 reference price. This calculation excludes slippage, financing costs, taxes, time value, and any dilution not yet reflected in the filed denominator. It is not an intrinsic-value estimate or personalized financial advice.

Current market price / level: $80.40, ON common stock.

Timestamp: 2026-08-04 04:00 Singapore time, equivalent to August 3, 2026 at 16:00 EDT. This is the last regular-market close before the after-close earnings release.

Primary instrument: Nasdaq-listed ON common stock.

Alternative expressions considered: Wait for the post-release market structure; use staged common-stock limits after the required checks; or use an options expression only if a fresh liquid chain, spread, implied volatility, and open interest are verified. Options are rejected for this run.

Confidence: Medium. The primary filing evidence is fresh and specific. The independent dollar size of AI demand, the next-session price reaction, the working-capital persistence, final share count, noteholder behavior, and executable liquidity remain uncertain.

What Could Go Wrong

“Other” end-market growth may not be AI data-center growth. The category includes computing, consumer, networking, communications, and other applications.
Q2 FCF may be timing-driven. First-half operating cash flow already fell by $87.8 million year over year.
Automotive remains the largest Q2 end market at $781.3 million. A renewed auto downturn can overwhelm AI growth.
The $6.0 billion buyback authorization is unexecuted capacity beyond the documented purchases. It can be suspended for liquidity or transaction reasons.
The $804.9 million 0% note maturity due in 2027 can create refinancing, conversion, hedge, and warrant pressure.
The all-stock Synaptics deal can issue approximately 53 million shares on the simple current-denominator bridge, while Synaptics debt treatment remains unknown.
The company reports one distributor at approximately 14% of Q2 revenue. A distributor correction or inventory adjustment could make reported demand look stronger than end demand.
The stock has already rallied 80.5% from the 52-week low. A 40.4% drawdown from the high is not a valuation floor.
The market can gap through staged controls after an earnings release. Spread, depth, venue status, and exit liquidity were not verified.
Live short interest, borrow, options, dealer-flow, fund-flow, and same-session premarket data are missing.

What Would Prove This Wrong

The thesis fails as an AI-power cash-reset long if the next operating disclosure shows Q3 revenue below the stated range, non-GAAP gross margin below the stated range, or no evidence that the “Other” growth is translating into repeatable power and cash economics.

The balance-sheet bridge fails if first-half cash-flow weakness persists while repurchases continue, if the 2027 notes require a cash response that consumes the liquidity cushion, or if the Synaptics exchange produces materially more dilution or debt than the current 12% and unknown-debt bridge implies.

A close below $58 together with one of those fundamental breaks is a review and exit condition. A price-only print below $58 is not standalone proof. Conversely, a price rally without better AI disclosure, cash conversion, and share-count evidence does not prove the thesis.

Best Trade Strategy

Direction: Conditional long, watch-only.
Preferred instrument: ON common stock on Nasdaq.
Common-stock stance: The $80.40 close is a reference price only. Consider staged, limit-only common stock only after a fresh post-release quote, bid-ask spread, order-book depth, venue status, volume, and exit-liquidity review. Do not use a market order into the first earnings gap.
Options stance: No options expression. A liquid chain, implied volatility, open interest, dealer positioning, and spread were not verified.
Take-profit map: Review the base case near $92 and the top case near $112. These are scenario levels, not guaranteed exits.
Stop or invalidation: Review near $58 only when paired with a guide, gross-margin, cash-flow, debt, or share-count break. A price-only stop is not enough to diagnose the thesis.
Timeline: Immediate reaction on August 4; operating review over 4 to 8 weeks; share-count, HSR, and transaction monitoring through the next filings and the August 17 waiting-period milestone.
Execution risks: Opening gap, high volatility, semiconductor factor de-risking, convertible hedge flow, acquisition headlines, wide spreads, shallow depth, halt risk, slippage, and the possibility that the market prices the filing before a staged order fills.
Do-not-trade conditions: No fresh same-session quote; wide spread; unreliable depth or exit liquidity; halted or disorderly trading; Q3 guide or deal terms materially worsen; the price is gapping through the intended risk control; or options are considered without a verified liquid chain.
Monitoring checklist: Q3 revenue and gross margin; AI data-center revenue or disclosed proxy; automotive and industrial growth; Q2-to-Q3 cash conversion; operating cash flow; free cash flow; capex; cash; total debt and 0% note conversion; actual share repurchases; remaining authorization; diluted shares; Synaptics S-4 and regulatory path; treatment of Synaptics debt; distributor inventory; and live market-structure data before any entry.

Bottom Line

ON’s fresh filing creates a specific disagreement. The last regular-market price still reflects a semiconductor company marked by the old auto and industrial cycle, while Q2 shows a materially better power, margin, and free-cash-flow print. That is the long case.

The counterweight is equally specific. AI is not separately quantified, first-half operating cash flow is weaker, the company used 0% note proceeds to fund part of its repurchase activity, an $804.9 million note maturity is current, and the Synaptics deal can add roughly 12% of the combined denominator. The right expression is therefore a conditional common-stock watchlist, not an AI slogan and not a net-cash floor.

The next decision should be made on the cash and denominator bridge. If Q3 delivery confirms margin and AI-power growth while cash conversion holds and dilution remains bounded, a move toward $92 to $112 is plausible. If the next filing shows narrative growth without cash or per-share delivery, the $55 downside case remains live. Until same-session execution checks are complete, the signal must remain fail-closed.

Research Quality Scorecard

Criterion	Score	Rationale
Market disagreement	5	A fresh after-close filing changes the operating and cash evidence while the reference price still reflects the pre-release cycle discount
Evidence base	5	The core claims use the August 3 SEC 8-K and Exhibit 99.1, the July 3 10-Q, and timestamped market data
Positioning and flows	3	Price, volume, and executed repurchases are visible; live short, borrow, options, dealer, fund-flow, spread, depth, and exit-liquidity data are missing
Catalyst path	5	The August 4 reaction, Q3 guidance, August 17 HSR milestone, and next 10-Q create observable closing mechanisms
Payoff architecture	4	The target map is asymmetric, but valuation, conversion, debt, and acquisition risks limit confidence
Invalidation discipline	5	The article separates price review from guide, cash-flow, debt, and share-count invalidation
Differentiated insight	5	The thesis separates AI mix, end markets, FCF, first-half OCF, buyback execution, current debt, and all-stock dilution
Client value	4	The bridge identifies what to monitor even if no trade is taken, while executable market data remains incomplete
Total	36 / 40	Publish as a conditional Watchlist long, not an executable trade

Sources and Data Audit

Source	Tier	Date / timestamp	Use
ON Semiconductor August 3, 2026 Form 8-K	Primary SEC	Filed August 3, 2026	Current report and Exhibit 99.1 incorporation
ON Semiconductor Q2 2026 earnings release, Exhibit 99.1	Primary company release	August 3, 2026	Revenue, margin, EPS, Q2 FCF, repurchases, AI data-center disclosure, and Q3 outlook
ON Semiconductor Q2 2026 Form 10-Q	Primary SEC	Filed August 3, 2026	End-market revenue, cash, debt maturities, working-capital cash flow, repurchases, notes, Synaptics exchange ratio, and transaction risks
ON market-data reference	Market data	Accessed August 4, 2026; regular-market timestamp 2026-08-03T20:00:00Z	Price, range, volume, 52-week levels, and historical volume series
Caterpillar Q2 2026 event notice	Primary company release	July 21, 2026	Candidate catalyst and scheduled reporting time
Clorox August 3, 2026 Form 8-K	Primary SEC	Filed August 3, 2026	Candidate-screen revenue, FY27 guidance, cash-flow, and GOJO integration evidence
Palantir August 3, 2026 Form 8-K	Primary SEC	Filed August 3, 2026	Candidate-screen growth, FCF, and valuation-asymmetry comparison

AI Illustration Prompt

Create a high-end institutional editorial illustration for The Mispricing Desk about a semiconductor cash-flow reset that must pass a denominator bridge. Show a dark graphite power-management wafer feeding two clearly separated rails: one rail labeled “AI data center” with a rising blue current, and another labeled “Automotive” with a slower amber current. In the center, place a transparent cash-flow bridge marked “Q2 FCF $425M” on one side and “H1 OCF lower” on the other. Beneath the bridge, show a precise share-count ledger with a 1.350x Synaptics exchange ratio, a quiet “12% combined ownership” marker, and a separate 0% note maturity card marked “$804.9M due 2027.” Add a restrained calendar card for “Q3 margin 40% to 42% guide” and a small document labeled “Next 10-Q: cash, shares, debt.” The mood should be analytical and unresolved, like a Bloomberg Markets or Financial Times cover, with charcoal, steel blue, muted amber, paper white, and a small controlled red accent for dilution risk. Avoid rockets, memes, casino imagery, generic candlestick charts, cartoon money, and AI clichés. Include a subtle but legible watermark reading “The Mispricing Desk.”