SEATTLE BUYING WINDOW VERIFICATION As of October 8, 2026 FINDING October ranks first in a six-factor monthly reconstruction of Seattle-Tacoma-Bellevue conditions for 2018-2025 excluding 2020. It scores 68.99 versus September's 68.36. This small gap supports a September-October window more strongly than one unique month. October is first in 8 of 10 scoring/period specifications and second in the other two. In individual years it is first in 3 of 7 years and in the top three in 6 of 7. The actual Realtor.com 2026 report names October 18-24. October 19-25 and the earlier quoted Seattle comparison figures belonged to the 2025 report. METHOD Geography: CBSA 42660, Seattle-Tacoma-Bellevue, containing King, Snohomish and Pierce counties. All residential core listings, including single-family and condo/townhomes, per Realtor.com definitions. 84 month observations. Score each year independently so dollar and inventory changes between years do not dominate. More active listings, fresh listings, days on market and price reductions score higher. Lower asking prices and views per property score higher. Each metric uses 100*(x-min)/(max-min) within a calendar year, reversed where lower is favorable. Average the six scores equally. Average the resulting scores by calendar month with equal weight for each year. Demand: compound the published month-over-month changes in views per property within each year. January=100 is a scale anchor; it is not an invented viewer count. This preserves within-year relative demand. Do not use the national-relative views column as absolute local demand. The public report describes 0-100 scores but does not fully specify scaling, calendar alignment or detrending. This is an explicit reconstruction, not an exact replication. Sensitivity: percentile-rank scaling, excluding 2025 from calibration, 2021-2025, 2023-2025, excluding entire 2022/2023 years with source quality flags, and leaving each year out individually. Source flags: core flags occur May-December 2022 and January-April 2023; hotness also flags February 2023. The baseline includes these observations with a sensitivity excluding those entire years. In that sensitivity min-max ranks September first and October second; rank scaling ranks October first. Redfin cross-check: five-factor proxy for two distinct metropolitan divisions, 2021-2025, using non-seasonally-adjusted values. Components are active listings, fresh listings, lower new-list asking price, longer days to contract and lower pending-sales/active-listings ratio. It omits price reductions and uses a demand proxy rather than Realtor.com views. It ranks October second for Seattle and first for Tacoma. Each Redfin observation is a rolling 28-day period, updated weekly, so it cannot isolate October 18-24. Overlapping windows are correlated. Results are averaged within calendar month of period end. Both publishers draw on MLS listings, so this is an independent publisher check, not wholly independent underlying data. CURRENT EVIDENCE AND LIMITS Core data through September 2026. Demand through August 2026. Redfin regional rolling windows through September 27, updated October 7. NWMLS's October 6 release reports three-county active inventory of 14,294 (+31.2% year over year) and pending sales down 16.4%. These regional facts corroborate favorable conditions but do not independently rank seasons. Listing medians reflect the mix of homes offered for sale. A 4.9% lower median does not show that the same house sells 4.9% cheaper. Rates are excluded to match the seasonal system. Actual payment affordability requires current financing terms. Public histories can be revised. This packet preserves the exact local extracts used. HOW TO VERIFY THE EXACT WEEK 1. Obtain Realtor.com's raw weekly panel and full score specification for CBSA 42660 from economics@realtor.com. Request 2018-2025 excluding 2020, all six weekly series, quality flags, boundaries, week start/end, active-listing definition, price reduction denominator, score scaling, year weighting, detrending and tie handling. Request current 2026 weekly data too. No message has been sent. 2. Reproduce all 52 seasonal weekly scores, and compare October 18-24 to every week, showing neighboring scores and a top-week band. Test equal-year weighting, percentile scaling, flagged-year exclusions and leave-one-year-out results. Freeze historic calibration before evaluating 2026; do not normalize a nine-month partial year as a complete year. 3. Independently reconstruct weekly listing supply and competition from NWMLS for King, Snohomish and Pierce counties. Required inputs: stable property identifier and listing identifier, property type, county/ZIP, initial and cumulative listing date, each status change timestamp, every price change with old/new price, original and current ask, pending date, closing date, sold price, concessions if available, relisting links, new construction/resale indicator, beds, baths, square feet and lot size. Historical active inventory requires status/price event histories or dated snapshots; a current active-list export cannot reconstruct prior supply. MLS showings per listing can be a labeled demand proxy, not an exact replacement for Realtor.com's audience. 4. Deduplicate repeated listings using property identity and retain cumulative market exposure across relistings. Score detached homes and condos separately; stratify by price range and submarket. For prices, compare matched properties or quality-adjusted asking-price measures as a sensitivity to sales/listing mix. 5. Judge purchase opportunities by offer/contract week. Closed-sale months lag negotiations. Follow contract cohorts to eventual sold/original-list ratios, concessions and failed contracts, with adequate maturation. Avoid treating withdrawals as completed buyer bargains. 6. For 2026, compare the target week against observed weeks as data arrives. A full-year winner can only be declared once the remaining weeks have occurred and reporting revisions mature. If October is near the maximum or other specifications favor September/November, report the wider favorable window instead of a unique best week. SOURCE LINKS https://www.realtor.com/research/best-time-to-buy-2026/ https://www.realtor.com/research/data/ https://econdata.s3-us-west-2.amazonaws.com/Reports/Core/RDC_Inventory_Core_Metrics_Metro_History.csv https://econdata.s3-us-west-2.amazonaws.com/Reports/Hotness/RDC_Inventory_Hotness_Metrics_Metro_History.csv https://www.redfin.com/news/data-center/downloads/ https://redfin-public-data.s3.us-west-2.amazonaws.com/redfin_data_center/housing_market/weekly/all_metros.csv https://www.redfin.com/news/data-center/methodology/ https://www.nwmls.com/home-inventory-reaches-highest-level-since-2014-expanding-buyer-choices/ CONTENTS ReadMe.txt: definitions, method, limitations and exact-week verification process. seasonal_profile.csv: six-factor monthly score results. monthly_scores.csv: each historical month and each score. results.json: full results and robustness checks. work/: local raw regional source extracts and reproducible analysis. seattle_buying_seasonality.png: chart. To rerun the preserved snapshot, run work/verify_seasonality.py from the extracted packet with Python, pandas and numpy. It writes intermediate results into work/. This does not automatically refresh upstream sources or create a scheduled monitor.