CoreWeave's entire business model depends on borrowing huge sums now and earning AI revenue later — it just launched a $3 billion convertible bond issue plus a share-sale vehicle, meaning it must tap markets repeatedly. With the 10-year Treasury yield abo

Verdict: CoreWeave's funding squeeze is real, but the trend filter says wait

The thesis has real support: CoreWeave's Q1 2026 filing shows gross margin down 2.1 points to 65.5%, net margin worsening 6.9 points to -35.6%, and shares outstanding up 5.9% to 532 million — dilution arriving exactly as the funding-squeeze argument predicts — while Bloomberg reported on September 17, 2026 that the company kicked off a new fundraising push with 10-year yields above 5%. The strongest counterpoint is that the business is genuinely monetizing demand: Q1 2026 revenue more than doubled year over year to $2.1B and the quarter generated $3.0B of operating cash flow, so if yields fall the high-leverage penalty reverses quickly. On the trade itself, three of four entry conditions are live (price below the $86.74 20-day average, RSI at 36.6, MACD negative) but ADX at 0.95 is nowhere near the required 20, and the completed nine-month backtest earned just 0.3% over 9 trades with a 33% win rate and a 12.4% maximum drawdown — a thin edge, not a green light. Verdict: wait. Reassess if ADX rises above 20, or immediately if the stock reclaims the 20-day average, which is also the strategy's exit condition.

Trade now

CoreWeave closed at $79.88, below its 20-day average of $86.74, with RSI at 36.6 and the MACD histogram at -1.37 — three of the four entry conditions for this setup are already live. The missing piece is trend strength: ADX (14) is at 0.95 and the entry requires it above 20. Until that turns, the trade is a watch, not an entry; 'wait' here means leaving limit orders aside and checking ADX daily rather than chasing weakness in a flat tape. The completed nine-month backtest of this exact rule set on CRWV daily bars traded 9 times for a 0.3% total return, a 33% win rate, and a 12.4% maximum drawdown. That is a thin edge, and the frozen parameter set could not be re-evaluated across longer windows because market-data coverage was incomplete, so no robust alternative parameter setup was established. Treat the historical result as modest evidence, not a green light. Risk framing if the entry triggers: the fixed-risk stop is at -2.42% (about $77.95 from the last close), the take-profit condition is +8.0% (about $86.27), and a secondary invalidation sits at a close at or below the first support level at $79.02 — effectively at the current price, so expect that support exit to be the live brake on any long-side execution. The structure offers roughly 3.3:1 reward-to-risk, but ADX at 0.95 says the market isn't trending yet. Be patient.

A Business Built on Borrowed Billions, Priced in a 5% World

The bear thesis starts with the balance sheet arithmetic. CoreWeave finished fiscal 2025 with $14.7B of long-term debt against just $3.3B of equity, $46.0B of total liabilities on $49.3B of assets, and a current ratio of 0.46 — $16.4B of current liabilities against $7.5B of current assets. The company paid $1.2B in interest expense in FY2025 while posting a net loss of $1.2B and a slightly negative operating income of $46M. That is a capital structure that requires continuous market access, and per the Bloomberg segment from September 17, 2026, the company just kicked off another fundraising push — described by the idea as a $3B convertible bond issue plus a share-sale vehicle, exactly the pattern the thesis predicts. The same day's Bloomberg article on Treasury yields describes 5%+ 10-year yields as potentially the 'new normal,' which raises the cost of every incremental borrowing. The free cash flow trajectory makes the dependence on external capital concrete rather than rhetorical: FY2025 free cash flow was negative $7.3B, and the most recent quarter (ended March 31, 2026) burned $4.7B — worse than the $2.5B burned in the prior quarter and far worse than the $1.3B burn in the year-ago quarter. Capex of $10.3B against $3.1B of operating cash flow in FY2025 means the gap is funded by debt and equity issuance, not operations. Meanwhile the share count rose from 502.4M to 532.0M in the latest quarter, a 5.9% increase — dilution is already showing up in the numbers, as the thesis argues. The most…

Scores

• Conviction score breakdown: 50
• Thesis support: 75
• Trade readiness: 35
• Risk quality: 45
• Backtest evidence: 30
• Fundamentals trend: 65

Watch items

• CRWV — ADX (14)
• CRWV — Close vs SMA (20)
• CRWV — RSI (14)
• CRWV — Price vs first support
• CRWV — Treasury yield regime