Track record
Margin's track record
Margin scores every note against what actually happened in the print — wins and misses, on the record, with sources. As calls resolve, they land here so you can see whether the desk's confidence holds up over time.
In the meantime, individual notes show the per-note evaluation footer with the scoring rubric.
Resolved calls
Every note on a company that has since reported — its rubric score and, once we've confirmed the outcome, how the call held up. Each one listed on its own; no overall rate is claimed until enough have resolved.
- TSLA2026-07-22
Tesla goes into its July 22 print with the strongest volume-and-margin setup in over a year — Q2 was its best-ever second quarter (~480K+ deliveries) and Q1 gross margin had already recovered 478bp YoY to 21.1% — yet the number that matters is operating margin, which fell to 4.2% in Q1 as AI/autonomy opex rose 37% YoY. The binding question the print resolves: is the core auto business now healthy enough to fund the robotaxi/Optimus build-out, or does the spend outrun a margin recovery that itself leans on a shrinking regulatory-credit crutch? That question is being asked into a week of autonomy setbacks — fresh federal robotaxi safety scrutiny, a rejected recall bid, and shareholders publicly pressing Musk on missed autonomy targets — which is exactly the narrative that carries the multiple.
Rubric 3.6/5HedgedMissed — and owned - NFLX2026-07-16
Netflix goes into Thursday's print down 42.4% with the week's headlines — WSJ, Yahoo, Bloomberg — all reframing it as a slowing-growth, engagement-plateau story; but the last hard filing still shows +16% revenue growth and we have no current-quarter numbers to confirm the turn: does the July 16 guide validate the slowdown the narrative is already pricing, or expose it as a story the filings don't yet support?
Rubric 3.8/5HedgedRight - WFC2026-07-14
Q2 (July 14) is the first "unshackled" print worth judging on its own terms: with the Fed's asset cap gone, Wells is finally growing the balance sheet — avg assets +13% YoY to $2.17T, loans +10%, deposits +6% — but net interest margin is compressing hard (2.47% in Q1'26, down from 2.67% a year earlier). The binding question is whether volume can keep outrunning price: NII rose 5% YoY in Q1 purely because balance-sheet growth offset a 20bp margin squeeze. If Q2 NIM keeps sliding and loan growth cools, the "growth story" thins to a rate story the market has already discounted — which is why the stock trades at a persistent discount despite the regulatory overhang lifting.
Rubric 4.0/5StakedRight - JPM2026-07-14
Going into Monday's Q2 print, the binding question isn't whether JPM beats — Q1 already ran red-hot (net income $16.5bn, ROTCE 23%, NII +9%) — but whether management RAISES the ~$103bn full-year NII guide it set in April, and whether a likely one-time Visa share-exchange gain flatters an otherwise clean quarter. The bull case is that resilient rates, a Markets/IB tailwind, and benign credit push NII toward the high end; the bear case is that the guide was already conservative-by-design and the stock, sitting ~1% off its 12-month high, prices in perfection. Watch NII and the Card net charge-off trajectory against the 3.4% full-year guide: those two numbers, not headline EPS, resolve whether the setup deserves the premium.
Rubric 4.8/5HedgedRight - GS2026-07-14
Goldman goes into its July 14 Q2 print with the investment-banking recovery as its bull case — Advisory revenue nearly doubled YoY in Q1 on a rebound in completed M&A — but two things the reader should hold going in complicate the story: the IB fees backlog actually ticked *down* versus year-end 2025, and FICC was the Q1 soft spot (-10% YoY). The binding question the print resolves is whether the deal pipeline is converting into fees or stalling, and whether FICC normalizes now that Q2's equity backdrop recovered from Q1's -5% S&P quarter. Watch backlog commentary and the FICC line: a re-acceleration confirms the franchise is firing on all cylinders; a second soft FICC quarter plus a flat backlog would reframe the 14%+ ROE as advisory-cycle-dependent rather than structural.
Rubric 4.4/5StakedRight - C2026-07-14
Citi enters its July 14 Q2 print having run Q1 RoTCE (13.1%) two-plus points above its own ~10-11% full-year 2026 target — the binding question is no longer "is the turnaround real" but "is mid-teens returns a run-rate or a Q1 markets-seasonality peak?" A $30B buyback authorization launched in June says management is betting on durability; the Q2 line to watch is whether RoTCE holds double digits once Q1's trading tailwind fades and whether U.S. card spend (down 9% sequentially in Q1) is a wobble or a trend.
Rubric 4.4/5StakedRight - BAC2026-07-14
BAC goes into its July 14 Q2 print at a record share price on the back of a late-June NII guidance upgrade — but the number that actually matters is whether net interest income reaccelerates after Q1'26 FTE NII came in essentially flat sequentially ($15.9bn vs 4Q25's $15.9bn), snapping the multi-quarter run of clean sequential gains. The bull case says fixed-rate asset repricing plus an 11th straight quarter of deposit growth reignites the NII curve; the bear case is that the plateau is structural now that the rate tailwind has faded. Complicating the read: management leaned hard into buybacks ($7.2bn in Q1, $9.3bn total returned) and let CET1 slip to 11.2%, so a soft NII print with less capital cushion is a less forgiving setup than the record price implies.
Rubric 4.0/5HedgedRight - LEVI2026-07-08
Levi enters its July 8 Q2 print with the highest bar it has set in years: management already raised full-year sales, margin AND EPS guidance in April on the back of a 9% organic Q1, so the print won't be judged on "did it grow" but on whether the raise holds. The binding question is margin: gross margin actually slipped to 61.9% in Q1 as tariffs bit, and the company is spending into a "Behind Every Original"/FIFA-heavy advertising cycle — so the tension is whether accelerating DTC and pricing power can absorb both tariff cost and elevated ad spend without a guide-down. A wildcard sits underneath: Levi paid ~$80M of IEEPA tariffs that a February Supreme Court ruling made potentially refundable, an optionality that could flip the margin narrative if management quantifies it on the call.
Rubric 4.8/5HedgedRight - NKE2026-06-30
With the stock down 48.6% from its 12-month peak, can Elliott Hill's "Win Now" reset show enough North America wholesale traction and a credible FY2027 gross-margin recovery path to offset the ~$1.5B annualized tariff hit, deteriorating Greater China (-10% CC) and a collapsing Converse — or does tomorrow's print confirm the turnaround is still all cost and no inflection?
Rubric 3.6/5StakedMissed — but we said so - MU2026-06-24
Micron guided FQ3 to $33.5B in revenue and ~81% gross margins — numbers that would have been unthinkable a year ago — but the print on June 24 will test whether the AI-driven supply shortage that produced these figures is intensifying, plateauing, or showing its first cracks in non-data-center segments.
Rubric 3.6/5Right - FDX2026-06-23
FedEx's June 23 print is the last look at the old company — a consolidated, May-fiscal-year FedEx that still includes the Freight business spun off June 1. The binding question isn't FY2026 EPS (management already raised the bar to $19.30–$20.10 adjusted in March); it's the FIRST FY2027 guide for "new FedEx" — an Express-led, calendar-year, Freight-less entity carrying tariff and trade-policy uncertainty the company says it cannot yet quantify. The bull case is that >$1B of structural cost-out plus 5%+ domestic yields keep margins climbing even as Freight leaves; the bear case is that international export weakness, the IEEPA tariff reversal, and a grounded MD-11 fleet pressure a stub-period guide that the Street has no clean comparable for. Watch the FY2027 framework, not the trailing quarter.
Rubric 4.8/5Right - CCL2026-06-23
Carnival goes into its June 23 Q2 print with a record booked position but a guidance bar that has quietly stepped down: management is guiding Q2 net yields to just +2.0% constant-currency, versus the +2.7% it actually delivered in Q1 and the ~+5% it ran in 2025. The binding question for the print is whether that deceleration is genuine demand cooling or simply conservative tougher comps plus fuel drag — and whether the +150bp Q1 yield beat repeats. With ~85% of 2026 already on the books at "historically high prices," the demand setup is de-risked; the watch item is a softening North America occupancy (-1.3pts in Q1) that bears can read as the first crack in the pricing-over-volume strategy.
Rubric 4.4/5Right - KR2026-06-18
Kroger's Q1 print on June 18 will reveal whether the post-Ocado, post-Albertsons reset under new CEO Greg Foran is translating into sustainable margin expansion — or whether the ~130bp IRA headwind and a tough 3.2% comp are compressing the ID sales trajectory faster than eCommerce profitability gains can offset.
Rubric 4.4/5Verdict pending
Track record by confidence level
How often a call held up, grouped by how confident the note was. Well-calibrated confidence means the green shrinks as you move down the rows — high-confidence calls should hold more often than low-confidence ones.
How we count a miss
- Right
- The call held up in the print.
- Missed — but we said so
- The call was wrong, and the note had already flagged low confidence. Doubt surfaced before the print — the system working as intended.
- Missed — and owned
- A high-confidence call that didn't hold. The genuine miss, logged plainly and counted in full — no spin, no quiet removal.