With the standard US mortgage now costing 7.45% — the highest since April 2024 — affordability for ordinary buyers is deteriorating fast, and new-home demand is the first casualty. Wealth manager Barry Ritholtz argues rising yields are not a blip but a 'r

Verdict: 7.45% mortgage rates argue for the short, but the trigger hasn't fired

**Verdict: the thesis has a live catalyst, but the trade is not yet on.** The strongest argument for the short is the macro setup: the 30-year mortgage hit 7.45% on September 24 per CNBC — the highest since April 2024 — and the idea, citing Ritholtz's Bloomberg argument, treats high yields as a durable 'return to normal' rather than a blip. The strongest argument against is that the freshest fundamentals contradict a demand collapse: DHI's June-quarter revenue rose 22.1% to $9.23B with free cash flow swinging from -$449.7M to +$397.3M, and Lennar insiders were net open-market buyers of roughly $348.8M per the June 30 ownership reports, while the backtest's own edge decayed to a 9.6% loss over the last twelve months despite a 36.8% total return over five years. No robust alternative parameter setup was established, so the fixed rules stand as published. The verdict flips if any of three things happen: RSI (14) on DHI, LEN or PHM confirms at or below 45 with the other entry conditions met (act on the short), a mortgage-rate pullback or a second straight positive cash-flow quarter at PHM materializes (abandon it), or the September 30 ownership filings show Lennar's insider buying continuing (treat the short as crowded against informed money).

Trade now: the short setup is close, but momentum has not confirmed yet

The builders are already trading below their 50-day trend lines and MACD readings are negative across all three names, so the trend part of this short thesis (per the idea's argument that builder stocks lag the mortgage-rate move) is largely in place: DHI closed at $141.51 versus its 50-day average near $144.45, LEN at $82.15 versus $83.59, and PHM at $119.85 versus $123.75. But the setup is not triggered. Each entry requires the 14-day RSI at or below 45, and the live readings are DHI 52.3, LEN 55.4, and PHM 49.1 — none there yet. DHI also needs its ADX reading above 20 (it is 3.8), while LEN (31.0) and PHM (28.0) already clear that bar. What "wait" means concretely: do not chase the short here. Watch for the 14-day RSI on each name to reach 45 or below — LEN is 10.4 points away, DHI 7.3 points, PHM 4.1 points. Once triggered, the plan's discipline is fixed: a stop at a 2.4% loss from entry and a take-profit at a 4.9% gain, a built-in reward-to-risk of roughly 2-to-1, with no single position above 25% of capital. The completed backtest supports the discipline but not urgency: over the five-year window the rule set produced 56 trades on DHI at a 44.6% win rate with a 36.8% total return and a 27.8% peak drawdown — but over the most recent twelve months it lost 9.6%, a reminder that this pair worked far better in the earlier rate-shock regime than in the last year. No robust alternative setup was established through parameter testing, so the published rules are the plan. Bottom line: thesis intact, entry conditions not yet met. Set alerts at RSI 45 on DHI, LEN and PHM (and ADX 20 on DHI) and act only on the trigger, not on the headline.

Rate shock, margin fade, and a short setup the data already scored

The macro setup is the cleanest part of this thesis. Per CNBC, the 30-year fixed mortgage jumped to 7.45% on September 24 — the highest since April 2024 — and per the Bloomberg segment, Barry Ritholtz frames higher yields as a 'return to normal' driven by tariff and energy inflation, not a temporary blip. If mortgage costs stay elevated, the affordability shock lands directly on new-home demand, the single biggest revenue driver for DHI, LEN and PHM. The fundamentals already show the stress the thesis predicts, even before the newest rate move. DHI's gross margin has slid from the 28–33% range of 2021–2022 to 21.7% in the quarter ended September 2025, and LEN's quarterly net margin has compressed from over 16% in 2021 to 3.8% in the quarter ended May 2026. LEN's latest full-year revenue fell 5.8% year over year and PHM's fell 6.3%. Cash generation is deteriorating where it matters: LEN posted negative operating cash flow of $433.5M and negative free cash flow in the quarter ended February 2026, and again in the quarter ended May 2026. PHM's free cash flow swung from +$134.4M to -$12.9M between the quarters ended March and June 2026 — builders are funding land and incentives into a softening market. The realized backtest statistics give the trade a scored history. Over the 60-month window the setup produced 56 trades with a 44.6% win rate, a 36.8% total return and a 27.8% maximum drawdown. The shorter windows matter for current conditions: over the most recent 24 months the setup ran 33 trades with a 45.5% win rate but a -15.0% return and a 16.5% drawdown, and over the last 12 months it ran 19 trades with a 36.8% win rate and a -9.6% return. That recent regime has been hostile to the setup — which is precisely the question a bearish reader should weigh against the fresh rate shock. Note that parameter-sensitivity work did not establish a robust setup (the evaluation exceeded its time budget), so these results reflect one fixed rule set, not a tuned one. Insider posture adds a modest signal. As of the June 30, 2026 ownership reports (deadline since passed), PulteGroup insiders were net sellers of open-market stock, to the tune of roughly $1.3M — small in dollar terms but directionally consistent with management not adding…

Scores

• Conviction score breakdown: 49
• Thesis support: 65
• Trade readiness: 40
• Risk quality: 55
• Backtest evidence: 45
• Fundamentals trend: 40

Watch items

• DHI — RSI (14)
• DHI — ADX (14)
• LEN — RSI (14)
• PHM — RSI (14)
• DHI — 30-year mortgage rate
• PHM — Insider open-market activity
• LEN — Insider open-market activity
• DHI — Price vs 50-day EMA