On what observation date will Freddie Mac’s 30-year fixed mortgage rate first exceed 6.75%?

• My median prediction for Freddie Mac’s 30-year fixed mortgage rate to first exceed 6.75% is 8th July 2026.
• The current rate of 6.51% (as of May 21, 2026) is only 25 basis points below the required 6.76% resolution threshold.
• Upward pressure is driven by rising inflation, which reached 3.8% in April 2026, and geopolitical tensions in the Middle East causing energy price spikes.
• A hawkish shift in the Federal Reserve, led by new Chair Kevin Warsh (sworn in May 22, 2026), has increased market expectations for interest rate hikes.
• Daily mortgage trackers and Treasury yield peaks of 4.67% in mid-May suggest the underlying market is already hovering near the target threshold.
• Key upcoming data releases, including the June 10th CPI report and the June 17th FOMC decision, are expected to be the primary catalysts for a breach in early July.

Key figures

Historical context

• Mortgage rates peaked at 7.79% in late 2023 before trending downward.
• In early 2025, rates again crossed the 7% mark during initial trade tensions.
• The 2026 low of 5.98% was reached in late February, but rates have risen nearly 53 basis points in the three months since.
• Historical PMMS data (since 1971) shows a mean rate of 7.69%, placing the current 6.51% significantly below long-term historical norms.
• The 99th percentile for a one-week upward move in PMMS rates is +31 basis points, indicating that a +25 bp jump to hit the 6.76% threshold is rare but historically precedented.
• The 10-year Treasury yield hit its 2026 closing low of 3.961% on 27th February 2026, marking the start of the current upward regime.

Tailwinds

• Rising inflation expectations (TE proprietary forecast predicts 3.9% CPI in June and July).
• New Fed Chair Kevin Warsh is viewed as ‘hawkish’ and may allow yields to rise to combat persistent inflation.
• 10-year Treasury yield technical momentum is ‘bullish’ (Golden Cross regime).
• Geopolitical instability (Iran conflict) and 25% tariffs on Canada/Mexico are driving energy and goods prices higher.
• Daily mortgage proxies (Lender Price, Optimal Blue) are already frequently reporting rates between 6.70% and 6.80%.

Headwinds

• Recent minor retreat in the 10-year Treasury yield from 4.67% to 4.49% in late May could provide short-term breathing room.
• Potential cooling of energy prices if a ceasefire or diplomatic resolution is reached in the Iran conflict.
• Historical tendency for the Freddie Mac PMMS to lag behind and ‘smooth’ daily market spikes reported by other trackers.
• Mortgage spreads (currently ~1.91%) could compress if the market perceives current high yields as temporary.
• A significant weakening in labor market data (e.g., jobless claims rising sharply) could force a more dovish Fed stance despite inflation.

Detailed reasoning

My analysis indicates a high likelihood of Freddie Mac’s 30-year fixed mortgage rate exceeding 6.75% (reaching 6.76% or higher) by the summer of 2026. As of May 21, 2026, the Primary Mortgage Market Survey (PMMS) reported an average rate of 6.51%. This leaves a 25-basis-point gap to the target threshold. While the PMMS typically lags daily market movements, daily proxies such as Lender Price and Bankrate already recorded rates between 6.70% and 6.80% in late May, suggesting that the broader market is already testing the threshold.

The primary driver of this forecast is the upward trajectory of the 10-year Treasury yield, which serves as the benchmark for mortgage pricing. The 10-year yield reached a peak of 4.67% on May 19 before a minor technical retreat. Historical data shows a median spread of approximately 216 basis points between the 10-year Treasury and the 30-year mortgage rate. At a yield of 4.60%, the implied mortgage rate is approximately 6.76%, exactly at the resolution threshold.

Macroeconomic conditions support sustained or rising yields. Inflation (CPI) surged to 3.8% in April 2026, and forecasts from Trading Economics suggest it will remain elevated at 3.9% through June and July. Geopolitical tensions, specifically the conflict involving Iran and resulting energy price shocks, are exerting additional upward pressure on inflation. Furthermore, the Federal Reserve transitioned to a new Chair, Kevin Warsh, on May 22, 2026. Warsh is historically viewed as ‘hawkish’ regarding interest rates and skeptical of Fed intervention in mortgage debt markets, leading markets to price in a higher probability of rate hikes (up to 62.5% by December) rather than cuts.

I have modeled the prediction around the weekly PMMS release dates (Thursdays). The highest hazard rates are concentrated in mid-June and early July. The June 11th and June 18th releases are particularly critical as they follow the release of May CPI data (June 10) and the FOMC policy decision (June 17). If inflation remains sticky or the Fed signals a policy firming, a sharp upward move in yields is expected. Given that a 25-basis-point weekly move is within the 99th percentile of historical PMMS changes (+31 bps), a quick breach is statistically plausible. However, the median date of 8th July 2026 accounts for the ‘choppy’ nature of weekly changes and the possibility of a more gradual multi-week ascent. While there is a risk that rates could stabilize if energy prices retreat, the current momentum and fundamental drivers strongly favor a breach in early Q3 2026.

Key uncertainties

• Whether the Federal Reserve will actually follow through with rate hikes in July or December 2026.
• The degree of ‘smoothing’ in the Freddie Mac PMMS methodology relative to more volatile daily trackers.
• Impact of the 14-day revision window specified in the resolution criteria, which could delay the ‘official’ date of a breach.
• Duration and intensity of energy price shocks from the Middle East conflict.
• Potential for ‘bond vigilantism’ to drive yields higher regardless of official Federal Reserve policy.

Conclusion

• The median forecast date is 8th July 2026, driven by persistent inflation and hawkish Federal Reserve expectations under new leadership.
• The 25-basis-point gap is narrow enough to be bridged by a single tail-event week or a series of consistent upward moves in Treasury yields.
• Decision-makers should closely monitor the June 10th CPI release and the June 17th FOMC decision as the most likely catalysts for the breach.

Question: will the weekly Freddie Mac PMMS 30-year fixed rate first equal or exceed 6.76% on a Date-column observation by the resolution cutoff? As of the forecast start (2026-05-27) the official PMMS was 6.51% on 2026-05-21—about 25 bps below the strict 6.76% trigger (note 6.75% would not qualify) but showing recent upward momentum (+15 bps week-over-week). Forecasters agree the series is close enough that a June crossing is plausible—historical post-2022 analogues show moves from the mid-6s to above 6.76% within 1–4 weekly observations—but they caution that single-week whipsaws are common and the five-week drift model likely overfits. High-frequency proxies (Mortgage News Daily, MBA) are mixed: MND briefly hit 6.75 then eased after May 19, which lowers the likelihood of the immediate 2026-05-28 print and argues for concentrating probability in mid-to-late June. Numerical/ensemble estimates range from roughly ~55% to ~77% chance of a qualifying observation by early August, with most mass in June and with a nontrivial tail for no qualifying print by the cutoff; forecasters differ mainly on how much to discount recent momentum given the softening daily proxies. Finally, remember resolution mechanics: PMMS weekly Date-column values determine outcome, and initial prints can be revised within a 14-day window.

Forecast and Reasoning

Forecast: Probability distribution:

• 10.00% chance of value below 2026-06-18 04:22:46 UTC
• 20.00% chance of value below 2026-07-08 21:46:03 UTC
• 40.00% chance of value below 2026-08-05 22:55:31 UTC
• 60.00% chance of value below 2026-08-12 12:00:00 UTC
• 80.00% chance of value below 2026-08-12 12:00:00 UTC
• 90.00% chance of value below 2026-08-12 12:00:00 UTC

Forecast Reasoning

Baseline (Freddie Mac PMMS / FRED MORTGAGE30US) is 6.51% on the latest observation date 2026-05-21. The contract triggers only on an observation date with 6.76% or higher, i.e., about +25 bp from the latest level.

Key dynamics: PMMS is effectively weekly (often Thursdays, but the archive’s listed “Date” governs), and the last clearly in-window “Date” is likely the early-August release (around 2026-08-06) because 2026-08-12 is a Wednesday. Over the last ~3 months rates rose sharply (5.98% on 2026-02-26 to 6.51% on 2026-05-21), showing the level is reachable, but moving another 25 bp within ~10–11 observations still plausibly requires either renewed Treasury/yield pressure or another macro/geopolitical shock; absent that, stabilization/mean reversion makes a miss-by-2026-08-12 a substantial possibility.

Synthesizing the analysts: Forecasters 1 and 4 converge on ~40% in-window with the median in the open-after tail; Forecaster 2 is notably more aggressive (median in mid-July, implying >50% in-window); Forecaster 3 appears overconfident and also mishandles the open-tail representation in parts of their submission. Weighting the more defensible “trend but not guaranteed” view, I center on ~40% chance of first crossing occurring on/before the last likely eligible observation date (≈2026-08-06), with the conditional (given it happens) timing concentrated in mid/late July.

An amateur forecast often (i) linearly extrapolates the recent +15 bp week and concludes a quick crossing, and/or (ii) anchors percentiles at the 2026-08-12 bound instead of placing mass past the bound to represent the “no qualifying observation date by then” outcome. This forecast improves by explicitly (a) tying possible resolution dates to PMMS observation dates, (b) treating the +25 bp gap as plausible but not “nearly certain” over ~10–11 observations, and (c) allocating substantial probability to the open-after tail via percentiles strictly after 2026-08-12.

Forecast rationale (numeric):

— Iteration 1 — Across the forecasts, the core reasoning is that Freddie Mac’s 30-year fixed rate is already close to the 6.75% trigger, so only a modest additional rise is needed for the first qualifying observation. The main drivers cited are:

• Current proximity to the threshold: The PMMS is around 6.51%, leaving only about 24–25 bps to clear.
• Recent upward momentum: All models note the rate has been rising, with the latest weekly move reinforcing the possibility of an imminent cross.
• Treasury yield and spread sensitivity: A move higher in the 10-year Treasury yield is treated as the main transmission mechanism, with mortgage spreads also relevant.
• Near-term macro catalysts: Upcoming economic data releases and the June FOMC meeting are seen as potential triggers for higher yields and, by extension, mortgage rates.
• Weekly observation structure: Because the PMMS is a weekly average, the event depends on a sustained move rather than a one-day spike, which slightly delays the most likely qualifying date.
• Wide uncertainty bands: If the threshold is not crossed during this near-threshold episode, the next crossing could be much later, so tails extend well beyond the near-term window.

Areas of consensus

There is broad agreement that:

• The threshold is close enough that a crossing in the next few weekly observations is plausible.
• If it happens soon, it is most likely to be in late July or early August 2026.
• The most important short-term risk factor is higher rates driven by Treasury yields.

Main disagreement

The forecasts differ mainly on whether the crossing happens before the mid-August boundary:

• Some put roughly 55–60% probability on a crossing by early August, with a median around August 6.
• Others judge the odds of crossing before Aug. 12 as lower, assigning most probability to after that date.
• This difference reflects varying weight placed on recent momentum versus the possibility that yields ease and inflation cools.

Overall synthesis

The collective view is that the mortgage rate is near enough to the threshold that a crossing is a live near-term possibility, but not assured. The most likely timing, if it occurs soon, is late July to early August 2026, while the distribution remains wide because a decline in yields could postpone the first qualifying observation substantially.

— Iteration 2 — The forecasts share a similar structure: they start from Freddie Mac’s PMMS rate being around 6.51%, note that it needs roughly another 25 bps to clear 6.75%, and then weigh whether recent upward momentum will continue long enough to trigger a crossing in the next few weekly observations.

Main factors driving the reasoning

• Current level is close to the threshold: only a modest rise is needed, so a crossing is plausible without a major shock.
• Recent momentum is upward: the rate has risen over recent weeks, which supports a near-term crossing case.
• Treasury yields are not strongly supportive right now: mixed or slightly lower 10-year Treasury trends could slow or interrupt the rise in mortgage rates.
• Weekly PMMS averaging matters: because Freddie Mac’s rate is a weekly survey average, brief daily spikes are less important than sustained moves.
• Macro uncertainty remains high: inflation, Fed policy, and broader risk sentiment could push rates either above the threshold quickly or back down for a long time.

Areas of agreement

• If the rate does cross soon, the most likely window is summer 2026, especially late July to early August.
• The event is viewed as sensitive to continued rate pressure; a sustained rise in Treasury/MBS yields would make a crossing more likely.
• There is a long tail of uncertainty if the threshold is not reached quickly.

Main disagreement

• One view puts more than half the probability on a crossing by the early-August observation, implying a near-term event is slightly more likely.
• Another view is more cautious, assigning only about 30% probability within the deadline window and putting the median outcome after the cutoff, with the possibility that the rate does not exceed 6.75% until much later.

Unified takeaway

Overall, the reasoning suggests that the threshold is most likely to be tested in early August 2026 if it happens soon, but there is meaningful disagreement about whether it will happen before the deadline at all. The core tension is between near-term upward momentum and recently softer Treasury support, with the weekly averaging of PMMS making the exact crossing date difficult to pin down.

— Iteration 3 — Across the forecasts, the main reasoning is that Freddie Mac’s 30-year fixed mortgage rate is currently close enough to the 6.75% threshold that a crossing looks plausible in the near term, but not inevitable.

Shared reasoning patterns

• Current level is close to the trigger: The PMMS rate is around 6.51%, leaving only about 25 bp to clear the “first exceed 6.75%” threshold, which effectively means reaching 6.76%.
• Recent trend is upward, but modest: One model emphasizes a recent pace of roughly +7 bp per week, supporting a possible crossing in July or early August if the trend continues.
• Short-term market moves matter, but smoothing delays the official print: Even though daily mortgage rates briefly touched 6.75%, the PMMS is a weekly average, so a one-day spike is not enough unless it lifts the weekly average above the cutoff.
• Treasury yields are the main driver: The forecasts link the mortgage rate path to 10-year Treasury yields and broader MBS/Treasury conditions. Recent easing in yields pulled expectations back somewhat.
• Incoming inflation data is a key catalyst: A rebound in yields, possibly driven by inflation surprises, is seen as the most likely way to produce a crossing soon.

Areas of consensus

• A summer 2026 crossing is plausible.
• The most likely window is mid-July to early August.
• There is meaningful uncertainty, with a non-trivial chance the rate does not exceed the threshold by the August deadline.

Main differences in emphasis

• Timing:
  • Some place the median around late July / early August.
  • Another leans slightly earlier, around mid-July.
• Confidence:
  • One forecast assigns a slight majority to a crossing by early August.
  • Another is more cautious, estimating only 35–40% probability before the deadline and assigning substantial weight to much later dates.
• Tail risk:
  • The most conservative view allows for a long tail into late 2026 or beyond, reflecting the possibility that mortgage rates stabilize or drift lower and the threshold is not reached soon.

Overall synthesis

The collective view is that the mortgage rate is close enough to the threshold that a crossing in summer 2026 is the central scenario, but the weekly PMMS methodology and the recent cooling in Treasury yields reduce confidence in a near-immediate breach. The forecasts cluster around mid-July through early August, while still leaving substantial probability on either a delayed crossing or no crossing before the deadline.

Macroeconomic Upward Pressures Forecasters note that reaching the target threshold requires a moderate, historically plausible upward shift in current mortgage rates. The primary drivers for this potential increase are sticky inflation—fueled by energy shocks and geopolitical tensions—and a decidedly hawkish repricing in the bond market. With near-term Federal Reserve rate cuts largely priced out and expectations shifting toward potential rate hikes, Treasury yields face significant upward momentum. The most likely catalysts for a sudden mortgage rate spike are upcoming summer macroeconomic data releases, particularly Consumer Price Index (CPI) reports, Nonfarm Payrolls (NFP), and Federal Open Market Committee (FOMC) meetings. An upside surprise in any of these could quickly propel rates above the threshold.

Structural Dampening Forces Balancing the hawkish macro environment are several structural headwinds preventing a runaway spike. First, the methodology of the Primary Mortgage Market Survey (PMMS) relies on a multi-day rolling average, which inherently dilutes and smooths out brief, single-day market spikes. Second, ongoing interventions by Government Sponsored Enterprises (GSEs), such as large-scale purchases of Mortgage-Backed Securities (MBS), are actively compressing the spread between Treasury yields and consumer mortgage rates. This intervention acts as a heavy anchor, meaning Treasury yields would need to surge exceptionally high to drag mortgage rates across the target line. Additionally, consensus forecasts from major housing and mortgage authorities anticipate a general cooling or stabilization of rates over the summer.

Timing and Outlook Due to the tension between hawkish inflation risks and strong structural dampening forces, forecasters agree there is a substantial likelihood that the rate will fail to reach the target before the late-summer deadline, which would result in an out-of-bounds resolution. If a breach does occur, it is widely expected to happen strictly in the wake of major inflation, employment, or central bank policy surprises in the early-to-mid summer months.

SUMMARY

Question: On what observation date will Freddie Mac’s 30-year fixed mortgage rate first exceed 6.75%? Final Prediction: Probability distribution:

• 10.00% chance of value below 2026-06-13 08:00:00 UTC
• 20.00% chance of value below 2026-06-22 16:00:00 UTC
• 40.00% chance of value below 2026-07-23 00:00:00 UTC
• 60.00% chance of value below 2026-10-22 00:00:00 UTC
• 80.00% chance of value below 2027-09-06 16:00:00 UTC
• 90.00% chance of value below 2028-09-07 00:00:00 UTC

Total Cost: extra_metadata_in_explanation is disabled Time Spent: extra_metadata_in_explanation is disabled LLMs: extra_metadata_in_explanation is disabled Bot Name: extra_metadata_in_explanation is disabled

Report 1 Summary

Forecasts

Forecaster 1: Probability distribution:

• 10.00% chance of value below 2026-06-18 00:00:00 UTC
• 20.00% chance of value below 2026-07-02 00:00:00 UTC
• 40.00% chance of value below 2026-07-30 00:00:00 UTC
• 60.00% chance of value below 2026-09-17 00:00:00 UTC
• 80.00% chance of value below 2027-06-17 00:00:00 UTC
• 90.00% chance of value below 2028-05-18 00:00:00 UTC

Forecaster 2: Probability distribution:

• 10.00% chance of value below 2026-06-11 00:00:00 UTC
• 20.00% chance of value below 2026-06-18 00:00:00 UTC
• 40.00% chance of value below 2026-07-23 00:00:00 UTC
• 60.00% chance of value below 2026-10-29 00:00:00 UTC
• 80.00% chance of value below 2027-06-17 00:00:00 UTC
• 90.00% chance of value below 2028-03-16 00:00:00 UTC

Forecaster 3: Probability distribution:

• 10.00% chance of value below 2026-06-11 00:00:00 UTC
• 20.00% chance of value below 2026-06-18 00:00:00 UTC
• 40.00% chance of value below 2026-07-16 00:00:00 UTC
• 60.00% chance of value below 2026-11-19 00:00:00 UTC
• 80.00% chance of value below 2028-02-17 00:00:00 UTC
• 90.00% chance of value below 2029-06-21 00:00:00 UTC

Research Summary

The research finds that Freddie Mac’s official Primary Mortgage Market Survey (PMMS) showed a 30-year fixed rate of 6.51% for the observation week published May 21, 2026, up from 6.36% the prior week, creating a 0.25 percentage-point gap to the target threshold of 6.76%. Some non-Freddie Mac sources and daily-rate providers reported intraday or alternate measures near 6.7–6.75% around May 19, 2026, but those differ from Freddie Mac’s weekly PMMS (the question’s required data source). Mortgage rates have recently moved sharply (examples given: ~+15 basis points week-over-week and larger multi-week swings), and historical weekly volatility typically ranges ±0.10–0.15% though larger moves occur in stressed periods.

Key drivers identified across the research include rising long-term Treasury yields (10-year and 30-year), elevated inflation readings (eg, April CPI noted as higher than recent years), geopolitical shocks (the Iran conflict and oil-price spikes), and a perceived shift in Fed policy under new leadership toward more hawkish outcomes. Prediction-market activity (notably Kalshi, which resolves on Freddie Mac PMMS) showed traders increasing probabilities for higher mortgage-rate thresholds in mid-May (including a jump in odds of >7% on May 18–19), while other markets (Polymarket) had low liquidity. The PMMS publishes weekly (Thursday) averages with a 14-day revision window, and the research emphasizes that only a Freddie Mac PMMS observation ≥6.76% on or before the August 12, 2026 deadline would meet the question’s criterion.

Sources used (as cited in the research):

• Freddie Mac PMMS: https://www.freddiemac.com/pmms
• CNBC: https://www.cnbc.com
• Kalshi: https://kalshi.com
• Polymarket: https://polymarket.com
• Mortgage News Daily: https://www.mortgagenewsdaily.com
• Mortgage Bankers Association (MBA): https://www.mba.org
• U.S. Bureau of Labor Statistics (CPI): https://www.bls.gov
• Federal Reserve / Treasury yields data (eg, FRED): https://fred.stlouisfed.org and https://www.treasury.gov
• Local/regional reporting referenced (example): https://www.cumberlink.com

RESEARCH

Report 1 Research

I’ll help you gather comprehensive information to assist with this forecasting question about Freddie Mac’s 30-year fixed mortgage rates. Let me conduct several searches to find current rates, trends, driving factors, and relevant base rates.

Forecast — Post 350

Modal: Aug 12, 2026 to Oct 28, 2026 (21.0%) • partial_shift • 11m15s

Interpretation

Reading: strict

The resolution criteria are highly prescriptive: a specific numerical threshold (strictly ≥6.76%, not 6.75%), a single named data source (Freddie Mac PMMS archive), a defined 14-day revision window, explicit annulment and fallback conditions, and a requirement for daily-precision date verifiability — all of which demand a narrow, literal reading with no room for charitable interpretation.

Time bounds

• observation_window: All PMMS observation dates from now through 2026-08-12T12:00:00+00:00
• resolution_date: 2026-08-12

Edge cases identified

1. A rate of exactly 6.75% does not qualify; the threshold is strictly 6.76% or higher — resolvers must check the exact reported value, not round.
2. The 14-day revision window means a rate initially reported below 6.76% could be revised upward into qualifying territory, or vice versa — the operative value is whatever is published within 14 days of initial publication.
3. If Freddie Mac retracts or declares a data point invalid within the 14-day window without providing a replacement, that observation date is ignored entirely.
4. PMMS releases typically fall on Thursdays, but the resolver uses the exact ‘Date’ column entry in the archive, not the inferred Thursday — a mismatch could matter if Freddie Mac posts a non-Thursday date.
5. If no qualifying observation date occurs on or before the deadline, resolution is ‘>2026-08-12T12:00:00+00:00’, not a null or annulment — this is a distinct non-annulment outcome.
6. The question could be annulled (not resolved as the fallback) if a qualifying rate is reported but the observation date lacks daily precision — an unusual edge case if the archive format changes.
7. Multiple PMMS releases could exceed 6.76% before the deadline; only the FIRST such observation date matters for resolution.

Research (3/3 variants, shared evidence pool)

Total evidence registered (shared pool): ?

Research Brief

Evidence confidence: medium

Scenario 1: Early threshold crossing — PMMS reaches 6.76%+ by late June 2026 [medium evidence]

Conditions favoring

The 10-year Treasury yield is already at 4.56% (above the 4.50% threshold analysts flagged) and recently touched 4.67%; the mortgage-Treasury spread would need to stay near current ~195bp levels or widen. A continued fiscal-driven bond selloff, persistent inflation, or delayed Fed action could push the 10-year to 4.83–4.88%, mechanically lifting PMMS above 6.76%. Daily trackers already showed 6.75% on May 19. The hazard model’s p10 date is ~early June 2026.

Conditions against

PMMS weekly surveys are backward-looking averages; even if daily rates spike, the weekly average may be cushioned. The 10-year pulled back from 4.67% to 4.56% within days, showing volatility. The gap from current 6.51% to 6.76% is 25bp, requiring a sustained (not just intraday) rise.

Scenario 2: Mid-period threshold crossing — PMMS reaches 6.76%+ in July 2026 [high evidence]

Conditions favoring

The survival model’s median crossing is ~5.8 weeks from late May (≈ early-to-mid July 2026), consistent with a gradual continuation of the current rate-rising trend. The 10-year Treasury trending toward 4.80%+ as fiscal concerns persist would provide the necessary lift. The p25–p75 range of the survival model spans roughly June through late July.

Conditions against

Treasury yields have shown high volatility and could reverse if the Fed signals rate cuts or if economic data weakens. Mainstream forecasts still project declining rates in H2 2026.

Scenario 3: No threshold crossing before Aug 12, 2026 — resolves as >2026-08-12 [medium evidence]

Conditions favoring

Mainstream analyst forecasts (Fannie Mae, Morgan Stanley, MBA, Bankrate) project 30-year rates declining toward 5.75%–6.3% by end-2026. A Fed rate cut (signaled as possible per source 34) or a Treasury yield reversal could pull rates back well below 6.76%. The survival model’s p90 date (~week 15.7) is beyond the resolution deadline, meaning ~22% probability of non-resolution. The random-walk p90 at the final week is 6.82% but p75 is only 6.67%, suggesting substantial probability of staying below threshold.

Conditions against

The current rate trajectory (6.51%, up 15bp in one week) is sharply upward; Treasury yields at multi-year highs; fiscal pressures are structural, not transitory; daily trackers already near 6.75%.

Scenario 4: Late-window crossing — PMMS reaches 6.76%+ in August 2026 (close to deadline) [low evidence]

Conditions favoring

Rates continue rising but more gradually; the random-walk model’s p90 final-week distribution is 6.82% but the threshold crossing could occur in the final weeks. If fiscal/inflation pressures persist but do not accelerate, rates could slowly grind higher and reach 6.76% only in late July or early August.

Conditions against

The survival model’s p90 crossing date already falls beyond the resolution window; this scenario requires rates to rise 25bp but over a longer timeframe than the median forecast. July–August seasonality could see rate volatility; the deadline is Aug 12 at noon UTC.

Key uncertainties

• Whether the 10-year Treasury yield will continue rising toward 4.83–4.88% (needed for PMMS to reach 6.76%) or reverse course — this is the primary mechanical driver
• Federal Reserve policy path: a rate cut or dovish signal could arrest the mortgage rate rise before it reaches 6.76%
• Whether the ~195bp mortgage-Treasury spread will widen or compress (spread compression would require even higher Treasury yields to reach the threshold)
• Timing of PMMS survey averaging: even if daily rates briefly touch 6.76%, the weekly average may not reach that level in a given Thursday release
• Calibration of the hazard/survival model: the assumed hazard rate of 0.12/week is uncertain and the direction of trend could shift quickly given Treasury volatility
• Whether fiscal concerns (deficit, ‘One Big Beautiful Bill’) will continue driving bond yields higher or be offset by other macro factors
• The 14-day revision window has no formal Freddie Mac counterpart — in practice PMMS figures are rarely revised, but an unusual data quality issue could change the qualifying threshold picture

Forecast variants (3/3)

Variant 0: inside_view (inside_view_v1)

Inside View Causal Analysis

Current state → mechanism → outcome

The causal chain is: 10-year Treasury yields (currently 4.56%, peaked at 4.67% on May 19) + ~195bp mortgage-Treasury spread → PMMS 30-year fixed rate needs to reach 6.76% (25bp above current 6.51%). The mechanical trigger: 10-yr needs to sustain ~4.83–4.88% for a full survey week to produce a qualifying PMMS reading.

Key causal factors driving the inside view:

1. Upward momentum: PMMS jumped 15bp in a single week (6.36% → 6.51%, per source 1). MND daily tracker showed 6.75% on May 19 (source 5). The 10-yr has risen ~54bp in 3 months (source 24). The 30-yr Treasury hit a 19-year high of 5.2% (source 21) — structural fiscal pressure from the “One Big Beautiful Bill Act” and $1.9T deficit (source 21).

2. Survival model (high evidence strength per brief): P(event by Aug 12) ≈ 77.7%; median crossing ≈ 5.8 weeks from May 27 = ~July 2, 2026; p10 ≈ early June; p75 ≈ late July; p90 beyond deadline. This is the dominant quantitative signal.

3. Random-walk model: p90 of final-week distribution = 6.82% (above threshold); p75 = 6.67% (below). Implies significant probability mass above the threshold by Aug 12.

4. Scenarios by evidence strength:

   • “Mid-period crossing — July 2026” tagged high evidence strength → weighted heavily
   • “Early crossing — by late June 2026” tagged medium → moderate weight
   • “No crossing before Aug 12” tagged medium → moderate weight (~22% from survival model)
   • “Late window — August 2026” tagged low → light weight

Distribution construction:

• The survival model gives P(event by deadline) ≈ 77.7%, so ~22% mass on “never” (bin_8 and beyond, but primarily the “>Aug 12” bins).
• The median crossing is ~July 2, 2026 (week 5.8 from May 27). PMMS releases on Thursdays only. Week 5.8 from May 27 ≈ July 2, 2026 — falls in Jun 25–Jul 4 (bin_3).
• p10 ≈ early June → bins 0-1
• p25 → mid-to-late June → bin_2
• p75 ≈ late July → bin_5-6
• Given the upward rate environment and that the threshold is only 25bp away with rates already at a weekly inflection, I’ll spread mass as follows:

Bin allocations (only Thursdays qualify):

• May 27–Jun 5 (bin_0): ~5% — very early, only 1-2 Thursdays, but p10 is near here
• Jun 5–Jun 15 (bin_1): ~8% — early June, within p10-p25 range
• Jun 15–Jun 25 (bin_2): ~12% — mid-June, approaching the median
• Jun 25–Jul 4 (bin_3): ~18% — near the survival model median (~July 2)
• Jul 4–Jul 14 (bin_4): ~15% — just past median, still in high-probability zone
• Jul 14–Jul 24 (bin_5): ~8% — approaching p75
• Jul 24–Aug 2 (bin_6): ~5% — p75 zone
• Aug 2–Aug 12 (bin_7): ~4% — near deadline, low but nonzero
• Aug 12–Oct 28 (bin_8): Resolves as >Aug 12 if not by deadline; survival model ~22% non-resolution. Since this bin straddles the deadline and can only capture the “>Aug 12” fallback outcome… wait. The bin structure shows bin_8 as “Aug 12 to Oct 28” — but the question resolves as “>2026-08-12” if no qualifying date by then. The fallback outcome would map to bins 8-12. The survival model gives ~22% probability of no crossing by Aug 12, which should mostly go to the “>Aug 12” fallback (bin_8) and beyond.

Actually, re-reading: if no qualifying date by Aug 12, resolution = “>2026-08-12T12:00:00”. This would fall in bin_8 or later. But if it happens after the deadline, it would be bin_8 through bin_12 depending on when. However, the question is asking for first exceedance — if it doesn’t happen by Aug 12, the resolution is literally “>Aug 12” which is an output, not a future date. So all “no qualifying date” mass should go to bin_8 as a “beyond deadline” bucket.

Let me reconsider: The survival model says ~22% probability of no resolution. The remaining ~78% gets distributed across the bins before Aug 12. Let me refine:

Across the 11 weeks of eligible Thursdays (Jun 4, 11, 18, 25, Jul 2, 9, 16, 23, 30, Aug 6), the survival model’s conditional weekly probabilities given the Weibull shape factor concentrate mass around week 5-6 (early July).

Final allocation:

• bin_0: 4% (Jun 5 is the only Thursday in this window)
• bin_1: 8% (Jun 5-Jun 15: June 5 and June 12 Thursdays - wait, Jun 5 is in bin_0 or bin_1? bin_0 is “May 27 to Jun 05” — Thursday June 4 or June 5 might be on the boundary)

PMMS is weekly Thursday. From May 27: May 28, June 4, June 11, June 18, June 25, July 2, July 9, July 16, July 23, July 30, August 6 are the relevant Thursdays.

• May 27–Jun 5 (bin_0): May 28 and Jun 4 — but Jun 5 is the upper bound; Jun 4 falls in bin_0
• Jun 5–Jun 15 (bin_1): Jun 11 Thursday
• Jun 15–Jun 25 (bin_2): Jun 18 Thursday
• Jun 25–Jul 4 (bin_3): Jun 25 and Jul 2 — both fall in bin_3
• Jul 4–Jul 14 (bin_4): Jul 9 Thursday
• Jul 14–Jul 24 (bin_5): Jul 16, Jul 23 Thursdays
• Jul 24–Aug 2 (bin_6): Jul 30 Thursday
• Aug 2–Aug 12 (bin_7): Aug 6 Thursday
• bins 8-12: no qualifying date by Aug 12 → fallback

Survival model medians align with ~July 2 (bin_3). p10 ~June 4 (bin_0). p75 ~ late July (bin_5-6). ~22% on no crossing (bins 8-12, mostly bin_8 for “>Aug 12” resolution).

Allocation:

• bin_0: 6% (May 28 and Jun 4 Thursdays; near p10)
• bin_1: 9% (Jun 11)
• bin_2: 11% (Jun 18)
• bin_3: 20% (Jun 25 and Jul 2; median is Jul 2)
• bin_4: 13% (Jul 9; just past median)
• bin_5: 9% (Jul 16, Jul 23)
• bin_6: 5% (Jul 30)
• bin_7: 4% (Aug 6)
• bin_8: 18% (“>Aug 12” fallback from ~22% non-resolution, slightly discounted as some of that could resolve in later bins)
• bin_9: 2%
• bin_10: 1%
• bin_11: 1%
• bin_12: 1%

Sum: 6+9+11+20+13+9+5+4+18+2+1+1+1 = 100 ✓

Variant 1: outside_view (outside_view_v1)

Reference class: time-to-first-PMMS exceedance given the current upward PMMS trajectory (the brief’s time-to-event/survival analysis is the best-matching reference class). Base rate: 77.7% chance the PMMS 30-yr rate reaches ≥6.76% on or before 2026-08-12 (the brief’s survival model: P(event by Aug 12) ≈ 77.7%, median crossing ≈ early-to-mid July; p10 ≈ early June; p75 ≈ late July) — I take this as the base rate because the brief flagged this survival projection as a strong quantitative piece of evidence (source 33 in the brief). Adjustment: I make no large shift from that base rate (no ±15+ pp change) because the brief treats the survival model as the primary quantitative anchor and provides supporting market facts (current PMMS 6.51% on May 21, 2026; gap ≈25bp to threshold; 10-year Treasury recent high 4.67% and mortgage–Treasury spread ≈195bp) that make the model directly applicable (sources cited in the brief: PMMS=6.51% on May 21; gap ≈25bp; survival model numbers). I do modestly widen the temporal spread because the brief’s confidence_in_evidence is only “medium” and because mainstream analyst forecasts and the possibility of Fed action introduce meaningful downside risk (brief cites Fannie Mae/Morgan Stanley forecasts and Fed-cut possibility). Distribution logic and mapping to bins: the brief’s survival-median (early–mid July) gets the largest single-bin allocation (bin_4 Jul04–Jul14 = 20.8%) and adjacent July bins (bin_5 Jul14–Jul24 and bin_3 Jun25–Jul04) receive substantial mass to reflect the p25–p75 spread (bin_3 = 14.4%, bin_5 = 14.4%). The brief’s p10 (≈early June) justifies nonzero mass in the earlier June bins (bin_1, bin_2) and a small early tail (bin_0). A modest tail into late July/early August weeks (bin_6, bin_7) reflects the survival-model p75 and random-walk p90 behaviour; total probability on or before Aug 12 sums to ~77.7% (the survival-model base rate). The remaining 22.3% (non-event by the deadline) is placed after the deadline across bin_8–bin_12 with most of that mass in the near-post-deadline bin (bin_8 Aug12–Oct28 = 16.0%) and small probabilities for later first-crossing windows (bins 9–12). Scenario-weighting rationale: per the brief, the “Mid-period threshold crossing — July 2026” scenario had the strongest evidence tag (high) and so receives the largest share of probability concentrated in early–mid July (bin_4) and nearby July bins; the “Early crossing” and “No crossing by Aug 12” scenarios were tagged medium and receive moderate probability (distributed across June and the post-deadline bins respectively); the “Late-window crossing (near deadline/Aug)” scenario was tagged low and receives a small share in the late-July/early-Aug bins. Data/brief citations used inline: most-recent PMMS 6.51% on May 21, 2026 and gap ≈25bp (brief facts), 10-yr yield behaviour and mortgage–Treasury spread (~4.67% peak, ~195bp spread), and the survival model statistics (P≈77.7%, median early–mid July, p10 early June, p75 late July) — all come from the supplied brief and were used to form the base-rate anchor and the bin-level temporal shaping. Reading: strict (I treated the brief’s resolution rules literally and used the brief’s survival projection as the primary reference-class anchor). Confidence/uncertainty note: the brief labels evidence confidence as medium, so I maintained the survival-model base probability while widening temporal dispersion modestly to reflect macro/Treasury volatility and the nontrivial chance of policy/data-driven reversals (hence non-negligible mass in June and post-deadline bins).

Variant 2: contrarian_v2 (contrarian_v2)

Adversarial reading

The bulk of the evidence, anchored by the high-evidence-strength survival model (Weibull hazard 0.12/week, median crossing ~5.8 weeks from late May), pulls toward a July crossing. The model’s p10–p75 window maps to early June through late July, with 77.7% probability of crossing by the Aug 12 deadline. Daily trackers already touched 6.75% on May 19 and the 10-year Treasury is at 4.56% (above the 4.50% analyst threshold), supporting mechanical upside.

Counter-evidence found (or absence)

Searched for counter-evidence; nothing substantive found; consensus appears robust. The medium-evidence reversal scenario (Fed cuts, Treasury pullback) and low-evidence late-window scenario rest on pre-surge analyst forecasts that are already outdated relative to the May 19–22 Treasury spike and fiscal drivers. Historical precedent of 33 consecutive weeks above threshold in 2024–25 also favors continuation once the trend is established.

Distribution implications

The high-evidence July scenario receives the largest mass (bins 3–5 together ~0.50). Early bins (0–2) receive modest but non-zero mass consistent with the model’s p10. Late-July and early-August bins receive smaller shares reflecting the p75 tail. The never-happens terminal bins receive only the model-implied ~22% residual, not inflated by lower-evidence bearish forecasts.

Consolidate

Strategy: logit_mean • Drafts: 3/3

Reviewers (3/3)

anchoring → shift_lower (medium magnitude, medium confidence) anchoring_v1

The draft rationale explicitly anchors its p50 at the survival model’s median crossing of “~5.8 weeks from May 27 ≈ July 2, 2026” and places the single largest bin mass there (bin_3 + bin_4 together ≈ 39%). However, the survival model’s hazard rate of 0.12/week was acknowledged as “uncertain” and flagged as potentially overstating the likelihood “if rates stall.” The rationale’s treatment of this as the dominant quantitative anchor — despite the survival model being a computed estimate with no external validation — risks anchoring the p50 at the optimistic (earliest plausible modal) end of the distribution. Critically, the mainstream analyst forecasts (Fannie Mae 5.7%, Morgan Stanley 5.75%, MBA/Bankrate/Norada all sub-6.5%) collectively represent a substantial body of evidence that rates won’t reach 6.76% at all; the rationale qualitatively dismisses them as “pre-surge” and outdated, but rates have a long history of mean-reverting from short spikes. Furthermore, the “no crossing by Aug 12” bin (bin_8) receives only ~3.8–18% in the draft variants, but a flat-prior reading of the evidence — where pre-surge institutional forecasts and Fed cut possibility collectively represent ~50% of the analyst community expecting sub-6.76% rates throughout 2026 — would assign substantially more mass to the never-happens outcome rather than treating 77.7% as an anchor.

Flagged concerns

• Optimistic anchoring on survival model median: The draft places its modal mass (bin_3 Jun25–Jul4 and bin_4 Jul4–Jul14 together ~39%) at the survival model’s median crossing of ~July 2, treating that computed estimate as the primary anchor. The survival model’s 0.12/week hazard was explicitly flagged as potentially overstating likelihood ‘if rates stall,’ but the draft rationale only qualitatively discounts this (‘NOTE: This hazard rate is uncertain’) without redistributing mass away from the modal early-July peak.
• Underweighting of no-crossing outcome: The draft assigns only ~17–22% combined mass to bin_8 and beyond (the ‘>Aug 12’ fallback), despite multiple mainstream institutional forecasts (Fannie Mae 5.7%, Morgan Stanley 5.75%, Bankrate below 6%, MBA/Norada 6.3–6.5%) projecting rates well below the 6.76% threshold throughout 2026. These forecasters are dismissed as ‘pre-surge and outdated,’ but a 25bp further rise from 6.51% — sustained over a full weekly survey average — has meaningful probability of not occurring given Treasury yield mean-reversion (10-yr pulled back from 4.67% to 4.56% within days of the spike).
• Optimistic treatment of daily tracker proximity to threshold: The rationale emphasizes that MND daily tracker showed 6.75% on May 19 as evidence of imminence. But this was a daily peak that immediately retreated (6.61–6.67% by May 22–26), and the PMMS weekly survey uses a full-week average of application data — not daily peaks. The draft’s early bins (bin_0–bin_1) may be slightly elevated due to over-reading this daily spike as a leading indicator of a near-term PMMS crossing.

ceiling → redistribute (low magnitude, high confidence) ceiling_v1

The key structural ceiling the forecaster partially but incompletely integrates is the PMMS weekly granularity constraint: the question resolves only on a Thursday PMMS publication date, and there are exactly 11 eligible Thursdays between May 28 and Aug 6, 2026. The brief’s survival model gives P(event by Aug 12) ≈ 77.7%, implying ~22% terminal “no event” mass. However, the forecaster then distributes that 22% across bins 8–12 (post-deadline bins), implying the event could happen after Aug 12 and still matter for resolution. This is structurally incorrect: the resolution criteria state explicitly that if no qualifying date occurs on or before 2026-08-12T12:00:00 UTC, the question resolves as “>2026-08-12T12:00:00+00:00” — a single terminal outcome. All ~22% “no event by deadline” mass should collapse into the closest post-deadline bin serving as the fallback bucket (bin_8), with bins 9–12 receiving essentially zero weight. The draft assigns 3.8% to bin_8 but spreads ~5.5% across bins 9–12, misallocating the terminal non-resolution mass across irrelevant post-resolution bins.

Flagged concerns

• Terminal bin misallocation: The resolution criteria state that if no qualifying PMMS observation occurs by 2026-08-12T12:00:00 UTC, the question resolves as ‘>2026-08-12T12:00:00+00:00’ — a single fixed terminal outcome. This outcome maps to the fallback resolution, which should be concentrated in the first post-deadline bin acting as the fallback bucket (bin_8). The draft assigns only 3.76% to bin_8 while spreading ~5.5% across bins 9–12, but since the resolution is a single fixed string (not a future crossing date), bins 9–12 should receive near-zero mass and all non-resolution probability should pile into bin_8.
• Weekly observation frequency ceiling: There are exactly 11 eligible PMMS Thursdays (May 28, Jun 4, 11, 18, 25, Jul 2, 9, 16, 23, 30, Aug 6). The forecaster correctly identifies this but partially double-counts by allowing bin_0 to capture both May 28 and Jun 4 — both plausible but the bin_0 upper bound is Jun 5, so Jun 4 is the only Thursday fully inside bin_0. The mass assigned to ~5.5% (bin_0) is reasonable but should be understood as representing at most 2 Thursdays.

math → redistribute (medium magnitude, medium confidence) math_v1

The draft distribution shows internal math coherence overall, but exhibits one notable issue: probabilities on the “never happens” outcome (bins 8–12) total only ~7.1%, whereas the forecaster’s own rationale cites the survival model’s ~22% probability of non-resolution by the Aug 12 deadline. The discrepancy suggests either the survival model’s 77.7% crossing probability was not fully mapped to the published bins, or an unexplained downward adjustment was applied. Specifically: bins 8–12 sum to 0.037630 + 0.009541 + 0.017816 + 0.012560 + 0.014403 ≈ 0.092 (9.2%), leaving ~13 percentage points unaccounted for against the stated 22% non-resolution baseline.

Flagged concerns

• Probability mass inconsistency: non-resolution tail: Rationale states survival model gives P(event by Aug 12, 2026) ≈ 77.7%, implying P(no event by Aug 12) ≈ 22.3%. However, bins 8–12 (all outcomes on or after Aug 12, representing ‘never happens’ or post-deadline resolution) sum to only ~9.2% (0.0376 + 0.0095 + 0.0178 + 0.0126 + 0.0144 = 0.0919). This is ~13 percentage points below the model-implied 22.3% non-resolution mass. Either the survival model’s non-resolution tail was substantially discounted without explicit justification, or the bin allocations do not fully reconcile with the stated base-rate anchor.
• Logical flow: stated confidence vs. distribution width: The rationale cites ‘medium’ evidence confidence and mentions widening the temporal spread, yet the p10–p90 implied by the draft runs from May 27–Jun 5, ~1.2 weeks from today (bin_0) to Aug 2–12, ~11 weeks from today (bin_7). This ~10-week spread is reasonable given Treasury volatility, but the phrasing ‘modestly widen’ understates the effective width. Minor: not an inconsistency, but a calibration note.
• Bin boundary matching: continuity check: The draft’s bin labels and contents appear to map correctly to the supplied bin_id keys (bin_0 through bin_12). No gaps or overlaps detected in the date ranges. CDF check: probabilities are all ≥0 and ≤1, and the sum across all 13 bins is 0.9999 ≈ 1.0 ✓.

Reconcile: partial_shift

Reviews consumed: 3

The ceiling reviewer correctly flagged a structural error: the resolution criteria specify that if no qualifying observation occurs by Aug 12, 2026, the question resolves as “>2026-08-12T12:00:00+00:00” — a single fixed terminal outcome that should collapse into bin_8, with bins 9–12 receiving near-zero mass. I accepted this fully and consolidated post-deadline mass into bin_8. The math reviewer flagged that the draft’s bins 8–12 total only ~9% versus the survival model’s ~22% non-resolution implication; I integrated this by raising bin_8 substantially to ~21%. The anchoring reviewer argued for shifting lower (more mass to no-crossing) citing mainstream analyst forecasts (Fannie Mae 5.7%, Morgan Stanley 5.75%) and Treasury mean-reversion; I partially accepted this but rejected the full shift_lower magnitude because those forecasts predate the May 2026 Treasury surge and the survival model (flagged as high-evidence) remains the strongest quantitative anchor at 77.7% crossing probability. The survival median (~July 2) keeps bin_3 and bin_4 as modal.

Rejected critiques

• [anchoring] Shift substantial mass to no-crossing outcome (bin_8 ~27%) based on mainstream analyst forecasts projecting sub-6.76% rates throughout 2026.

  • Why rejected: Those institutional forecasts predate the May 2026 Treasury surge and the PMMS jump from 6.36% to 6.51% in a single week. The survival model is flagged as high-evidence-strength and gives 77.7% crossing probability; fully discounting it would over-weight stale forecasts. I split the difference by raising bin_8 to ~21% rather than 27%.

• [math] Distribute the ~22% non-resolution mass across bins 8–12 to match survival model.

  • Why rejected: Partially accepted on magnitude (raising tail mass to 21%+) but rejected on allocation — the ceiling reviewer correctly noted that the ‘>Aug 12’ fallback is a single terminal outcome that should pile into bin_8, not spread across bins 9–12 which represent later first-crossing dates not resolution-relevant.

Final Distribution (date ranges)

Evidence Sources (46)

1. Freddie Mac PMMS - Mortgage Rates Average 6.51% — Freddie Mac PMMS reported the 30-year fixed-rate mortgage at 6.51% as of May 21, 2026, up from 6.36% the prior week — a 15 basis point jump in one week.
2. Freddie Mac Mortgage Market Survey Archive — Freddie Mac PMMS archive shows recent weekly 30-yr fixed rates: May 21=6.51%, May 14=6.36%, May 7=6.37%, March 26=6.38% (also noted 6.22% the week prior to March 26).
3. Mortgage Rates Drop Below 6% for the First Time in 3.5 Years - Freddie Mac — Freddie Mac PMMS reported a 30-year FRM rate of 6.76% approximately one year ago from May 2026 (i.e., around May 2025), per the Freddie Mac news release comparing to “a year ago at this time.”
4. Freddie Mac PMMS Archive — Recent PMMS weekly readings in 2026: May 21: 6.51%, May 14: 6.36%, May 7: 6.37%, Feb 19: 6.01% (a local low). A year ago (May 2025) the 30-year FRM averaged ~6.85% per Freddie Mac’s own year-ago comparison.
5. Mortgage News Daily - 30 Year Fixed Mortgage Rates — Mortgage News Daily (daily rate tracker) shows the 30-year fixed at 6.61% on May 26, 6.65% on May 22, 6.67% on May 20, and 6.75% on May 19, 2026 — indicating rates are very close to the 6.76% threshold in daily tracking (though not yet reflected in PMMS).
6. FRED MORTGAGE30US series, 1-year history — FRED MORTGAGE30US data (PMMS-based) over the past year: max was 6.89% (May 29, 2025), min was 5.98% (Feb 26, 2026), current (May 21, 2026) is 6.51%. The 90th percentile over 52 weekly observations is 6.75% and the overall mean is 6.36%. Standard deviation is ~0.25 percentage points per week on average (but this is cross-sectional std, not weekly change volatility).
7. FRED - 30-Year Fixed Rate Mortgage Average — FRED indicates the next PMMS release date is May 28, 2026, which will be the next weekly observation date after May 21, 2026’s reading of 6.51%.
8. FRED - MORTGAGE30US series history — FRED 52-week PMMS history (May 2025–May 2026): The 30-yr rate peaked at 6.89% (May 29, 2025), was above 6.76% from late May through late June 2025 (6.89, 6.85, 6.84, 6.81, 6.77), hit a low of 5.98% in Feb 2026, then rose to 6.51% by May 21, 2026. The p90 over this period is 6.75%, suggesting ~10% of observations exceeded 6.75%.
9. FRED MORTGAGE30US series — PMMS 30-year rate history (Dec 2025–May 2026): reached a low of 5.98% on Feb 26, 2026, then rose sharply — 6.38% (Mar 26), 6.46% (Apr 2), 6.37% (Apr 9), 6.23% (Apr 23), 6.37% (May 7), 6.36% (May 14), 6.51% (May 21). The 6-month std dev is 0.141%, and the max in this period is 6.51%.
10. Morgan Stanley Mortgage Rates Forecast 2026 / Yahoo Finance — Morgan Stanley forecasts mortgage rates dropping to ~5.75% in 2026; Fannie Mae’s April Housing Forecast puts the 30-year fixed at 6.1% by end of 2026 — both are well below the 6.76% threshold.
11. Norada Real Estate Mortgage Rates Forecast May–July 2026 — Norada Real Estate forecast (May–Jul 2026): expects 30-year rates to stay in the 6.3%–6.5% range in May 2026, with a modest chance of easing in June–July 2026 — not projecting a move above 6.75%.
12. level_forecast tool output — Level forecast for the 30-year PMMS rate over 11 weeks (to ~Aug 12, 2026), using a random-walk model calibrated on 52 weekly observations: mean=6.51%, p10=6.24%, p25=6.37%, p75=6.65%, p90=6.78%, p95=6.86%, std=0.21%. The threshold of 6.76% sits near the 90th percentile of the projected final-week distribution.
13. FRED DGS10 and Yahoo Finance — The 10-year Treasury yield was 4.56% on May 22, 2026 (FRED). The mortgage-to-Treasury spread was approximately 1.88-1.93 percentage points in mid-May 2026 (e.g., 10-yr at 4.48% → PMMS at 6.36% = 1.88% spread).
14. CNBC / Reuters bond rout May 2026 — 10-year Treasury yield surged to 4.67% on May 19, 2026 (highest in over a year), driven by global bond rout on inflation fears; CNBC reported traders are now betting on NO rate cut in 2026 and a rate hike becoming more likely.
15. FRED DGS10 series — 10-year Treasury yield was 4.56% on May 22, 2026, up from 4.02% on Feb 26, 2026 — a rise of ~54 basis points in ~3 months; the current spread between the 10yr yield (4.56%) and the PMMS 30yr rate (6.51%) is ~195 basis points.
16. Derived from FRED DGS10 and PMMS data — To reach 6.76% in PMMS, the 10-year Treasury would need to reach approximately 4.83-4.88% at the current ~1.9% spread, or the spread itself would need to widen (currently at 4.56% + ~1.95% spread = ~6.51% PMMS). The 10-yr was at 4.67% on May 19, when MND daily rate hit 6.75%.
17. FRED MORTGAGE30US series, 2-year history — From 2-year PMMS history (May 2024–May 2026): The 30-year fixed rate exceeded 6.75% (i.e., was 6.76%+) on these PMMS dates: July 18, 2024 (6.77%), July 25, 2024 (6.78%), Nov 7, 2024 (6.79%), Nov 14, 2024 (6.78%), Nov 21, 2024 (6.84%), Nov 27, 2024 (6.81%), Dec 26, 2024 (6.85%), Jan 2, 2025 (6.91%), Jan 9, 2025 (6.93%), Jan 16, 2025 (7.04%), Jan 23, 2025 (6.96%), Jan 30, 2025 (6.95%), Feb 6, 2025 (6.89%), Feb 13, 2025 (6.87%), Feb 20, 2025 (6.85%), Feb 27, 2025 (6.76%), Apr 17, 2025 (6.83%), Apr 24, 2025 (6.81%), May 1, 2025 (6.76%), May 8, 2025 (6.76%), May 15, 2025 (6.81%), May 22, 2025 (6.86%), May 29, 2025 (6.89%), and many more through late June 2025. The rate dropped below 6.76% around July 3, 2025.
18. US News - Mortgage Rate Forecast 2026 — Mainstream forecasts for 2026 mortgage rates (made earlier in 2026) anticipated 30-yr rates staying 6.1%–6.3% (Fannie Mae, MBA), far below 6.76%. These forecasts were made before the recent surge above 6.5%.
19. FRED MORTGAGE30US series — The 30-year PMMS rate was at or above 6.76% continuously from approximately November 7, 2024 through late June 2025 (roughly 33 consecutive weeks), before declining sharply. It reached a peak of 7.04% in January 2025 (week of Jan 16, 2025).
20. The Mortgage Reports / Norada Real Estate — For much of 2025, the average 30-year mortgage rate hovered near 6.6%, about the same as 2024’s 6.7% average. The rate is currently (May 21, 2026) at 6.51%, down 35 basis points year-over-year from a year ago (~6.86%).
21. CRFB - Rising Interest Rates are Exploding the Debt — The 30-year Treasury yield hit 5.2% in late May 2026, its highest in nearly 19 years, according to CRFB (May 21, 2026). Rising fiscal deficits (CBO projects $1.9T deficit in FY2026) are driving Treasury yields higher, which is a key upstream driver of mortgage rates.
22. FRED MORTGAGE30US series — PMMS 30-year rate was at 6.76% on both Feb 27, 2025 and May 1, 2025, and 6.81% on May 15, 2025 — these are the most recent instances just above/at the 6.75% level, meaning exactly one year ago the rate was near this threshold.
23. CRFB - CBO Releases February 2026 Budget and Economic Outlook — CRFB noted the 10-year Treasury yield hit 4.7% in recent weeks (May 2026), driven partly by fiscal concerns from the “One Big Beautiful Bill Act.” The 10-yr at 4.7% is approximately consistent with a PMMS rate near 6.6-6.65% at current spreads.
24. FRED DGS10 series — The 10-year Treasury yield surged from ~4.46% on May 12 to 4.67% on May 19, 2026 (a 21 bp increase in one week), before pulling back to 4.56% by May 22. The 90-day DGS10 std dev is 0.14%, and the trend is sharply upward (+13% total).
25. WIO News / social media discussion — With the 10-year Treasury at ~4.60% in May 2026, market debate is focused on whether yields will rise to 5.00%, which would plausibly push mortgage rates to the 6.75%+ range given the historical ~195bp spread.
26. Bankrate / Fannie Mae April 2026 Housing Forecast — Bankrate’s 2026 forecast is for mortgage rates to potentially fall below 6%; Fannie Mae projects 6.1% by end-2026 — these consensus forecasts imply rates are unlikely to rise to 6.76%+ without a significant shock, representing a contrarian view against the upward rate scenario.
27. Forbes Mortgage Rates Forecast 2026 / Norada Real Estate 90-Day Forecast — Fannie Mae’s March 2026 Housing Forecast projects that 30-year fixed mortgage rates will decline to 5.7% by end of 2026. Most analyst forecasts expect rates to stay in the 6–7% range with a modest downward bias toward year-end 2026. One forecast for May–July 2026 gave a range of 6.37%–6.46% for the 30-year fixed.
28. FRED DGS10 series, 90-day history — The 10-year Treasury yield rose sharply in mid-May 2026: it was 4.47% on May 14, jumped to 4.59% on May 15, reached 4.67% on May 19 (a 90-day high), and stood at 4.56% on May 22. This sharp Treasury move is consistent with the PMMS jump from 6.36% to 6.51% in the May 14→May 21 PMMS survey.
29. level_forecast tool calculation — Random walk projection from current 6.51% over 11 remaining weekly PMMS releases (to Aug 12, 2026): median stays at 6.51%; p90 = 6.82% (above threshold), p75 = 6.67% (below threshold); p10 = 6.20%. To reach 6.76%, rates need to rise ~25bp from current level — achievable but not the central case.
30. survival_forecast tool calculation — Survival model (Weibull shape 1.2, hazard ~0.08/week, 11 weeks to deadline): P(PMMS rate reaches 6.76%+ by Aug 12, 2026) ≈ 60%. Median week to exceedance is ~week 9 (mid-July 2026). This is calibrated to current trajectory from 6.51%, needing ~25bp more upside.
31. Trading Economics - US 30-Year Mortgage Rate — Trading Economics shows the May 28, 2026 PMMS reading as 6.51% (same as May 21), suggesting the May 28 release may not yet be available as of the data pull, or rates held steady.
32. Boston Fed Why Mortgage Rates Exceed Treasury Yields 2026 — Boston Fed (2026 publication) notes the 30-year fixed mortgage rate is ~6.5% with the 10-year Treasury at its current level; typical mortgage-Treasury spread is around 190–195bp currently. If Treasury yields reach 4.80%–5.00%, that would mechanically push mortgages to 6.75%–6.95% at current spreads.
33. survival_forecast tool output — Survival forecast for when the 30-year PMMS rate will first exceed 6.75% (reach 6.76%+): Using a hazard rate of ~0.12/week (Weibull shape 1.2), the model gives P(event by Aug 12, 2026) ≈ 77.7%, with median crossing ~5.8 weeks from now (≈early July 2026), p10 at ~1.2 weeks (≈early June), p75 at ~10.3 weeks (late July 2026), p90 at ~15.7 weeks (beyond resolution). NOTE: This hazard rate is uncertain and reflects the upward-trending environment but may overstate the likelihood if rates stall.
34. Scotsman Guide - Will a Fed Rate Cut Lower Mortgage Rates — Scotsman Guide reports a Fed rate cut is seen as “imminent” amid economic uncertainty, rising debt, and fears of a bond market shock — suggesting the Fed may actually cut rates to prevent financial instability, which would be bearish for mortgage rates rising further.
35. CBS News Mortgage Rate Forecast May 2026 — CBS News mortgage forecast for May 2026 stated: “If the 10-year Treasury breaks back above 4.50%, the 30-year mortgage rate will head straight back toward 6.75% or higher.” As of May 22, 2026, the 10-year was already at 4.56%—above 4.50%—suggesting professional analysts anticipated the rate approaching 6.75%+ under current Treasury conditions.
36. Freddie Mac PMMS May 21, 2026 press release — The year-ago comparison: a year ago (approximately May 21, 2025) the 30-year FRM averaged 6.86% per Freddie Mac’s May 21, 2026 release. This confirms the FRED series value of 6.86% for May 22, 2025 week.
37. Social media / PMMS archive references — Historical reference: PMMS showed the 30-year rate at 6.76% on November 27, 2024 (declining from 6.81% the week before), and again multiple times in early 2025 and May 2025. The rate has oscillated above and below 6.75% repeatedly in the 2024-2025 period, indicating this threshold is frequently crossed during rate-rising episodes.
38. Freddie Mac PMMS – Mortgage Rates — The most recent confirmed Freddie Mac PMMS 30-year fixed-rate mortgage observation is dated May 21, 2026, at 6.51%, up from 6.36% the prior week. No PMMS release for May 28, 2026 or any later date has been published as of the current search date.
39. 30-Year Fixed Rate Mortgage Average – FRED, St. Louis Fed — FRED series MORTGAGE30US shows the last observation as 2026-05-21 at 6.51%, with the next scheduled release date listed as May 28, 2026; no observation for May 28 or later appears in the series.
40. Freddie Mac PMMS Archive and News Releases (search coverage through May 2026) — A comprehensive search of Freddie Mac news releases, the official PMMS archive, and news databases found no retraction, correction, invalidation, or suspension notice for any PMMS weekly observation in the May–August 2026 period. The most recent PMMS observation (May 21, 2026, 30-year FRM = 6.51%) and prior weeks in the archive appear as normal, unrevised data points with no data-quality flags.
41. Freddie Mac PMMS Archive Weekly Data — The Freddie Mac PMMS archive shows the following unrevised 30-year FRM data for late April–May 2026: April 30 = 6.30%, May 7 = 6.37%, May 14 = 6.36%, May 21 = 6.51%. Each week’s data was released as a standard weekly press release with no indication of subsequent correction or invalidation.
42. Freddie Mac PMMS Archive — The Freddie Mac PMMS official archive and PMMS homepage both show the May 21, 2026 30-year FRM rate as 6.51% and the 15-year FRM as 5.85%; this is the most recent release available as of the latest data pull. No subsequent release (May 28 or later) appears in either the Freddie Mac archive or the FRED MORTGAGE30US series.
43. Freddie Mac Mortgage Market Survey Archive — Freddie Mac’s PMMS archive page carries a blanket disclaimer that all content “is subject to change without notice” and is provided on an “as is” basis, but there is no documented formal 14-day revision window or specific correction/retraction procedure published for the PMMS series on Freddie Mac’s website.
44. Freddie Mac PMMS Publication Calendar — The Freddie Mac PMMS publication calendar shows the May 28, 2026 release covers the survey week of May 21–27, 2026, and the June 4, 2026 release covers May 28–June 3, 2026. Neither of those releases nor any subsequent week in June 2026 shows a rate figure in any publicly accessible source as of the current data pull.
45. 30-Year Fixed Rate Mortgage Average in the United States – FRED — The FRED series page for MORTGAGE30US notes “All data are subject to revision” as a general disclaimer, but provides no information about any specific revision window length (e.g., 14 days), and historically PMMS weekly figures are rarely if ever revised after initial publication — the series shows the May 21 figure of 6.51% has not been altered.
46. Mortgage Rates – Freddie Mac PMMS FAQ — The enhanced (post-November 2022) PMMS methodology draws rates from actual Loan Product Advisor (LPA) mortgage applications submitted Thursday through Wednesday each week, with results published every Thursday at 12 p.m. ET. The methodology page and FAQs contain no mention of a formal 14-day revision window or correction/retraction notices for published weekly figures.

wooley_rhino • run c85615cd • $3.61 • 11m15s

The weekly application-based mortgage survey publishes on a fixed cadence, so only discrete weekly observations can trigger the threshold before the cutoff.

The cutoff window ends mid-August, leaving eleven eligible weekly observation dates to produce the first 6.76%+ print.

The reported 30‑year rate is about 25 basis points below the 6.76% trigger, so a crossing requires several above-average weekly gains or a single large jump.

Immediate upward pressure comes from Treasury yields, inflation persistence, Fed policy expectations, and mortgage-spread behavior; recent moves have been mixed.

A sustained rise in the 10-year toward the mid-4.6–4.8% band or an abrupt widening of mortgage spreads would materially increase the likelihood of a qualifying weekly observation.

By contrast, Treasury easing toward the low-4.3% area or spread compression would make a crossing within the window unlikely.

Recent weekly volatility under the current application-based survey is modest and large single-week moves of ~25 bps are uncommon, so base-rate evidence from similar starts dominates short-term odds.

Major remaining uncertainties are the small sample of comparable recent starts, the unobserved composition of application flows that feed the survey, and the potential for infrequent large Treasury or spread shocks.

Forecast Summary

Forecast: No median (P50) — probability of resolution within the evaluation window (May 27 – Aug 12, 2026) is approximately 30–40%, with the remaining 60–70% probability mass extending well beyond August 2026.

Rationale

• Current gap is substantial: Freddie Mac’s 30-year rate stands at 6.51% (May 21), requiring a +0.25 percentage point jump to reach 6.76%. This typically represents a 3-sigma weekly move.

• Treasury yields just reversed: The 10-year yield dropped from 4.67% to 4.49% over one week on Iran deal progress, widening the distance to the ~4.81% level needed to push mortgage rates to threshold. This near-term headwind reduces the probability of crossing within the 11-week window.

• Base rate and market pricing suggest 30–40% in-window probability: Historical data shows 26.9% of similar 11-week windows (starting from the 6.20–6.60% range) reached ≥6.76%. Kalshi’s full-year 2026 market implies ~45% for >6.8%, but conditioning on just this window and accounting for the recent DGS10 pullback yields ~30–40%.

• Geopolitical and inflation uncertainty dominates timing: An Iran deal finalization would likely suppress yields further, pushing resolution beyond the window. Conversely, deal collapse or a hot inflation print (CPI in June/July) could accelerate crossing to late June or mid-July. This uncertainty is too large to support a confident median within the window.

• Right-skewed distribution: If the threshold is crossed, it is more likely to occur earlier (on residual upward momentum) than later. However, the current downward yield trajectory makes early crossing less probable than it would have been one week ago.