Mortgage bond funds are bleeding cash as yields spike — short the long-bond proxies

Verdict: the mortgage-hedging feedback loop is real, but no trigger has fired yet

The verdict: this is a thesis worth watching and a trade worth skipping today. The strongest point for it is that both halves of the feedback loop — mortgage-hedging selling pressure and retail flight from mortgage bond funds — were independently reported on the idea's publication date (Bloomberg's outflow story and Reuters' hedging story, both October 12, 2026), and the shrinking payouts fit: REZ's 2026 distribution fell to $1.04 per share from $2.27 in 2025, and TBF's fell 31.7% to $0.37. The strongest point against is that the attached mechanical system is a contrarian LONG on the very assets the thesis is short — and its completed 9-month evaluation produced 9 trades, a 22.2% win rate, and a roughly -97.9% return — while no bearish ruleset was tested and no robust parameter setup was established because market-data gaps blocked the 12-, 24-, and 60-month windows. Neither leg is close to actionable either: REZ needs RSI to fall from 48.4 to 40 or lower (its price and trend conditions are met), and TBF needs both a close below its $25.90 50-day EMA and an RSI drop from 57.8. The verdict flips if yields stop rising — the idea's own stated invalidation — so the next reassessment trigger is a sustained move back above the 50-day EMAs ($91.16 on REZ, $25.90 on TBF); until a defined trigger fires, waiting costs nothing. **Conviction breakdown** (0-100, equal weights; the composite is computed server-side): Thesis support 65 — same-day sourced outflow and hedging headlines plus shrinking distributions. Trade readiness 25 — zero of the required entry conditions are met on either leg. Risk quality 30 — thin reward-to-risk near current levels and the hedge correlation is only mildly negative (-0.32 over 499 overlapping days). Backtest evidence 20 — one 9-month window, 9 trades, and it tested the opposite trade. Fundamentals trend 35 — payout cuts are real but issuer fundamentals are unavailable and TBF's distribution history is too noisy to extrapolate.

Trade now: two of three entry conditions are met on REZ — RSI is the holdout

The idea is a pure rate-shock trade, but the entry logic is a contrarian long on oversold downtrends in both REZ and TBF, so the actionable question today is not whether mortgage-bond funds are broken — the idea argues they are — but whether the tape is oversold enough to act. On REZ ($88.17), two of three entry conditions are already met: price sits $2.99 below its 50-day EMA ($91.16), and the trend-strength reading of 45.2 is comfortably above the required 30. The binding constraint is the RSI condition: it needs to be at or below 40 and currently reads 48.4, about 8.4 points away. On TBF ($26.68), the setup is further away: price is actually $0.78 *above* its 50-day EMA ($25.90), the trend condition is met, but RSI at 57.8 needs to fall roughly 17.8 points to reach the 40 threshold. Neither leg is triggered — this is a waiting setup, not an untriggered thesis. Once an entry fires, the risk framework is mechanical. A REZ position stops out on a daily close below the nearest support level ($87.84) or at a 2.0% unrealized loss, whichever comes first, and targets the nearest resistance ($88.92) or a 4.0% gain. From $88.17, the level-based stop is only about 0.4% away while the nearest resistance target is roughly 0.8% above — a thin, roughly 2:1 reward-to-risk before the fixed -2.0%/+4.0% brackets are considered, so sizing at the 2.0% fixed-risk setting with a 25% position cap matters more than conviction. TBF's bracket is wider: support at $25.54, resistance at $27.00. Be honest about what the completed 9-month backtest says: 9 trades, a 22.2% win rate, and a total return of about -1.0% with a maximum drawdown of 0.98% (exits were filled on daily bars, so stop and target quality is approximate). That profile — many small losses, frequent exits — is exactly what you'd expect from a fast-bracket contrarian rule, and it argues for strict adherence to the stop rather than discretionary averaging. Separately, no robust parameter setup was established: the optimization could not be evaluated due to incomplete market-data coverage, so the frozen rule set is what you get. Concretely, "wait" means: do nothing on TBF until price closes below $25.90 *and* RSI reaches 40 or lower; on REZ, watch for RSI to close below 40 while price stays under its 50-day EMA and trend strength stays above 30. If yields stabilize before those thresholds are hit — which is the idea's own stated invalidation scenario — the setup expires quietly and no trade is the right trade.

The rate-shock feedback loop is showing up in real flows

The idea's core mechanism — forced mortgage hedging compounding retail outflows — was directly echoed in the news on the same day this idea was published. Bloomberg reported on October 12, 2026 that US mortgage bond funds suffered one of the worst outflows since 2021, and Reuters reported the same morning that the US yield surge is triggering mortgage hedging that adds to bond-market pressure. Both legs of the feedback loop in the thesis — hedging-driven selling plus retail flight — are independently sourced, dated October 12, 2026. The dividend record…

Scores

• Conviction score breakdown: 35
• Thesis support: 65
• Trade readiness: 25
• Risk quality: 30
• Backtest evidence: 20
• Fundamentals trend: 35

Watch items

• REZ — RSI (14)
• REZ — ADX (14)
• REZ — Close vs 50-day EMA
• REZ — Daily close vs nearest support
• TBF — Close vs 50-day EMA
• TBF — RSI (14)
• TBF — Daily close vs nearest support
• TBF — Trailing 12-month distribution per share
• REZ — Trailing 12-month distribution per share