AI data-center fears are punishing innocent bystanders — buy Galaxy Digital's undervalued Helios

Verdict: the divergence story is real, but the entry is weeks away — wait

The StoneX contagion thesis (per The Block, October 19, 2026) has genuine support: Galaxy's operating margin turned positive at 0.7% in the March quarter while CoreWeave's net margin deteriorated 6.9 points to -35.6%, and the pair's two-year correlation of just 0.09 backs the idea that Galaxy is not simply CoreWeave's proxy. The strongest point against is that the entry conditions are nowhere near firing — CoreWeave's daily RSI (14) of 14.6 sits 25.4 points below the 40 gate, and Galaxy's own $16.43 support reference has already been broken to the downside, with CoreWeave closing at $61.12, less than 0.5% above its $61.00 support. The rules produced no trades over 246 evaluated daily bars in the trailing 12 months, and no robust parameter setup was established because longer evaluation windows could not be computed on incomplete market data. CoreWeave's deterioration is real — a $740M quarterly net loss, free cash flow of -$4.71B, and share count up 5.9% in one quarter to 532M — so the

Trade now: both names are far from a live entry — wait for the reclaim

Nothing in this idea is actionable at today's prices. The strategy is a **Support Reclaim with Momentum Turn** long on CoreWeave (CRWV) and Galaxy Digital (GLXY), and both entry sets are far from triggering. CoreWeave closed at $61.12 with a daily RSI (14) of 14.6 — the entry needs RSI to cross above 40, so the oscillator is 25.4 points below its gate. The MACD line is at -5.92 and must cross above zero, which it has not done. Galaxy is in the same position: a close of $15.72, RSI of 13.1 (26.9 points below the trigger), and MACD at -1.82 below zero. "Wait" here means concretely: no position until both the support-reclaim price pattern (a low touching the first support level — $61.00 on CoreWeave, $16.43 on Galaxy — with the close back above it) and the two momentum conditions fire on the daily bars. If CoreWeave fills, the first invalidation is a daily close back below the $61.00 support reference, and the take-profit reference sits at the second resistance level, $63.00; for Galaxy the equivalent references are $16.43 support and $17.00 resistance. Those are tight geometric targets, so expect the position sizing (fixed risk of roughly 2.7% per position, capped at 25% of the book) to do most of the risk-control work. This is a watch-list setup by design, not a signal you are being asked to take on faith: the rules were evaluated on real daily bars and simply did not open an entry. One scope note that matters to the action plan: the historical evaluation of the frozen rule set could not be completed because market-data coverage for both charts was incomplete, and no robust parameter setup was established — so the levels above are the live, as-published configuration. The two stocks also move mostly independently (a correlation of just 0.09 over the past two years), which supports the idea's core claim that Galaxy is not simply CoreWeave's proxy. The broader context cuts both ways. CoreWeave's net margin worsened to -35.6% in the March quarter from -28.7%, and share count rose 5.9% quarter over quarter — real deterioration that bulls of the sector must acknowledge. Galaxy's operating margin actually turned positive at 0.66% from -2.16%, which is consistent with the StoneX argument the idea leans on: the two companies' fundamentals are diverging even as the market sells them together. That divergence is precisely what the wait-for-reclaim structure lets you measure.

Sector-wide panic may be mispricing the fundamentally distinct name

The thesis, per StoneX (as reported by The Block on October 19, 2026), is that the market is punishing Galaxy Digital (GLXY) for CoreWeave's (CRWV) debt problems after three shelved flotations and a failed Korean debut soured the whole AI data-center trade (per MarketWatch on October 20, 2026). The fundamentals partially support the distinction. In the quarter ended March 31, 2026, Galaxy's operating margin turned positive at 0.7%, up 2.8 percentage points from -2.2% the prior quarter, and its return on equity improved by 8.0 points to -11.9% from -19.9%. Its net margin also improved to -2.2% from -3.8%. That is a company whose reported trajectory in Q1 was getting better, not worse — consistent with the idea that Galaxy's sell-off is sentiment-driven rather than driven by its own results. CoreWeave, the alleged source of the contagion, does show real revenue momentum even as it burns cash: revenue climbed from $15.8M in FY2022 to $5.1B in FY2025, with Q1 2026 revenue of $2.08B versus $982M in Q1 2025 — more than doubling year over year. Operating cash flow reached $3.06B for FY2025. If the bull argument is that demand for AI data-center capacity is genuine, these revenue and cash-flow lines back it up. On the balance-sheet separation the thesis requires, Galaxy is the cleaner of the two: it holds $1.25B in cash against $2.49B of long-term debt, with a current ratio of 1.60. CoreWeave's current ratio is 0.46, with $16.4B in current liabilities against $7.5B in current assets — the leverage-stress profile StoneX says the market is wrongly applying to Galaxy. On execution mechanics: this is a watch-list setup, not an active signal. The entry rules were evaluated on real daily bars over the last 12 months (246 bars for CRWV) and never fired, so the strategy is waiting for a support reclaim accompanied by a momentum turn (a 14-period RSI crossing above 40 alongside a MACD line crossing above zero) before any position would open.…

Scores

• Conviction score breakdown: 38
• Thesis support: 62
• Trade readiness: 25
• Risk quality: 50
• Trigger proximity: 12
• Fundamentals trend: 42

Watch items

• CRWV — RSI (14) cross above 40
• CRWV — MACD (12,26,9) line
• CRWV — Daily close vs first support
• CRWV — Net margin trend
• CRWV — Shares outstanding
• GLXY — RSI (14) cross above 40
• GLXY — MACD (12,26,9) line
• GLXY — Daily close vs first support
• GLXY — Operating margin