1Crude tankers are repricing on voyage distance, not on barrels
Ross Givens, in a video titled "Forget Oil... These 3 Shipping Stocks Are PRINTING Money," makes the case that the move in crude tankers has very little to do with how much oil is being produced and almost everything to do with how far it has to travel. Traffic through the Strait of Hormuz is running at roughly 20 ships a day against about 125 before the crisis, so Asian refiners are buying from the US Gulf, Brazil and Guyana instead — voyages about 2.6 times longer. A tanker is paid for ton-miles rather than barrels, and the Atlantic-to-Pacific share of very-large-crude-carrier movements went from 22% to 35% in eight weeks. Day rates tripled to around $280,000 against a break-even near $28,000–30,000, and on a largely fixed cost base almost the whole gap falls to the bottom line.
His four names are one trade expressed four ways. (Frontline) posted $2.51 of first-quarter earnings per share into the rate spike, with his targets at $60–80. (DHT Holdings) is the pure very-large-crude-carrier play near $20, paying a variable dividend yielding about 7.36%. (International Seaways) near $90 is diversified across crude and product tankers, which is why he calls it the safer of the three. (the SonicShares shipping ETF) is the basket version, flagged on a breakout setup.
The load-bearing part of the argument is durability rather than the rate itself: his claim is that refiners and captains will not trust the strait for years even if a ceasefire holds, so the rerouting persists after the shooting stops. He names his own reversal condition, which is the honest part — a genuine de-escalation between the US, Israel and Iran collapses ton-miles and takes the rates down with them — and says he thinks that unlikely. He also frames tankers as a rotation away from a crowded AI-semiconductor trade at lofty valuations, which is a relative argument, not a bearish one. One thing to weigh alongside it: the video doubles as a funnel for a paid subscription service, though the analysis underneath is specific enough to check.
The consequence is concentration. If the detour is structural, the day rate effectively is the earnings line for all four; if the strait is trusted again, the whole position unwinds at once, because this is one variable wearing four tickers.
2A memory quarter guided tight beyond 2027 got sold 5%
Micron's print was, on the numbers, strong: roughly $41.5bn of revenue, a market capitalisation approaching Meta's, and guidance that DRAM and NAND supply stays tight beyond calendar 2027. Ryan Detrick and Sonu Varghese, both of Carson Group, spent a stretch of The Compound's TCAF 248 treating it as confirmation that the memory cycle has years left to run, with the capital-spending read-through pointing at equipment names such as . fell about 5% on Friday regardless.
It was not alone, which is the part worth holding onto. Friday's semiconductor tape was broadly negative: off about 1.6%, roughly 3%, 3%, leading the losses on storage weakness, and down roughly 24% as the session's largest single loser. Alongside that, closed its worst week since the 2001 dot-com bust on concerns about how AI build-outs are being financed — in the same week Zacks upgraded it to Buy. Those two readings of Oracle are not reconcilable from the outside, and neither side of it was about demand.
So the selling ran on financing and positioning while the operating guidance ran the other way. Nothing in Friday's tape contradicted the supply-tightness claim, and nothing in the supply-tightness claim protected the share price. What would settle it is whether memory pricing actually holds through the next quarter — until then the two facts simply sit next to each other. (One caveat on the tape figures: the Yahoo Finance markets page returned a partial error state and served some month-old items alongside fresh ones, so the Friday moves here are the ones corroborated by the other feeds.)
3Intel's floor is being underwritten by the government, not by the business
Felix, on Felix & Friends, built a full video around and was candid that the operating business does not carry the thesis: his own scoring system rates it 13 out of 100, revenue growth is around 7%, and the foundry loses money. What he argues does carry it is ownership and strategy. The US government holds roughly 10% of the company through CHIPS Act support, the Trump trust holds both stock and bonds, and Paul Pelosi bought about $6m of Intel call options expiring in roughly a year. The operating case he puts underneath that is narrower than the AI headline suggests — data-centre and AI revenue up 22%, the 18A node (RibbonFET plus PowerVia) now in production, and a bet on cheaper inference and deployment processors rather than any attempt to take training workloads from .
The caveat is the call. He states explicitly that this is not a buy yet: a watch-list candidate with an illustrative entry near $136, a stop at $115 or $97, a 1–5% position cap and an automated exit. Repeated without those numbers it becomes a different claim than the one he made. Two things sit awkwardly next to it — the video is heavily promotional, and its hook is a copy-the-politician framing, which is a reason to check the node roadmap rather than the disclosure. itself fell 3% on Friday in the broader chip selloff.
A floor supplied by policy is only as durable as the policy. That makes the 18A ramp and the persistence of the government stake the things this thesis lives or dies on, rather than the next quarterly print.
4The bull market entered the eight-longest list at half the length of the seven above it
Asked directly whether this is a bubble, Ryan Detrick on The Compound said no, and his answer was arithmetic rather than valuation. The run from October 2022 has just cracked the eight longest bull markets since the Second World War at about 3.5 years, and the seven ahead of it averaged around seven years — his line is that "bull markets last longer than you think." Note what that argument does and does not do: it addresses duration, not price.
The same session was defensive underneath. Gold rose 1.2% to about $4,096 with the VIX around 18.4, and gold came up 61 times in the recording — attributed to persistent client questions about the dollar's reserve status rather than to any single catalyst. The rest of the tape was rotation: small-caps with momentum, software firm, and Bitcoin exposure through versus regional banks described as the hottest current trade, with at an all-time high and up 12.6% as the day's top gainer on healthcare leadership.
The one uncomfortable number came out of a conversation about process rather than positions. Chris Mayer, who wrote 100 Baggers, spent an Excess Returns interview on how to find companies rather than on what he owns — how to identify high-return-on-capital reinvestment cultures with long runways, and whether AI erodes the moats those compounders depend on. (Constellation Software, listed in Toronto and outside a US-equity universe) was his culture exemplar rather than a recommendation. But SpaceX came up 63 times, cited at roughly 145 times revenue on a $2.6trn mark, discussed as a valuation case study and not buyable in public markets anyway. Duration and price are separate arguments, and Friday's material argued the first at length while only gesturing at the second.
What the sources recommended
Four buy calls came out of four videos, and all four are a single thesis from a single person rather than four independent reads.
The buys are , , and , all from Ross Givens and all the same Hormuz-rerouting trade set out above — held together with his own stated reversal condition, that a genuine de-escalation ends it. Take the caveat with the call: he is selling a subscription service in the same video, and three of the four names are direct operators of the same asset class, so they will move together.
The negative note of the day was not an exit. Givens argues the AI trade is crowded and semiconductor valuations are lofty, but he says it as a reason to rotate into tankers, not as a reason to sell chips — a relative call rather than a sell.
One name came close to a call without becoming one. Intel was presented as a watch-list candidate with an entry level, a stop and a size cap, and explicitly not triggered. Two others were discussed and pointedly not recommended: Micron, treated as a strategist's read on the memory cycle rather than a purchase, and Constellation Software, offered as an example of reinvestment culture and listed outside the US market in any case.
The next test of the tanker thesis is diplomatic, and it does not run on a market calendar
This was written over a closed weekend — all four videos went up on Friday 26 June, and the next US session is Monday 29 June. Tuesday 30 June is the last day of the second quarter and the first half, so quarter- and half-end rebalancing flows are the main scheduled event. There is very little else on the calendar: the Finnhub calendar shows nothing notable in the immediate day-ahead, and second-quarter earnings do not begin in earnest until around 9 July (PepsiCo) and 10 July (Delta and the banks).
That leaves the unscheduled items carrying the tape. The durability of the Iran–Israel truce is the single variable underneath the tanker trade, and it is the same variable behind the firm diesel refining economics Finnhub flagged. Whether the AI-financing wobble that gave Oracle its worst week since 2001 continues is the test of the second section. Gold at about $4,096 and a VIX near 18.4 say the defensive bid has not left yet. And there is now less advance warning built into the system than there used to be: the new Fed chair, Walsh, has moved to a far less transparent communication regime — a 130-word statement — trading telegraphing for flexibility, which means more of the repricing arrives on the day rather than ahead of it.
