△ ascent
Opening Ascent…
Opening Ascent…
Understand what supports your home’s price, explore your next move, and keep the evidence behind your decision. Start with an address. We’ll show you the recorded sales and the reasoning.
Type at least 3 characters of a street address.
No address in mind? Try the sample
Free, with no account. Covers houses and condos in Washington, DC, and houses and townhouses in Montgomery and Prince George’s counties, Maryland. Built from public records, not live MLS listings.
Optional private workspaces collect pricing, contacts, offers and deadlines. Public studios need no sign-in.
Demo: fictional records. Your account: your own evidence.
Move the price. Watch the reasoning change.
Near the weighted estimate of $963,010. 2 of 6 adjusted sales sit at or above this price.
Fictional practice evidence. 6 eligible fictional sales, 5.4 effective after weighting, largest weight 23.7%, and adjusted spread 5.6% of the estimate.
Decision Time Machine, miniature example. Compare this fictional calculation with a deliberately altered practice snapshot in which one fictional sale was entered $15,000 lower. Neither snapshot represents an actual market change.
Observed estimate difference: +$2,724. A changed calculation is not proof that any real home gained or lost value. Saved workspace analyses can be replayed against their archived input evidence.
Assumptions: $110/sq ft · $12,000/bedroom · $15,000/bath · $14,000/condition point · 0% monthly trend · 3 mi · 180 days. Weights favor nearer, newer, similar-size sales.
You do not need to be an agent to examine pricing evidence. Choose a covered market, or learn with a fictional example before exploring a real address.
Search eligible DC houses and rowhomes. Inspect recorded sales, adjustments and the validated settings behind the estimate.
Explore DC homes Two MD counties · Limited property fieldsExplore homes in Montgomery or Prince George’s County, with transparent limitations on property fields and accuracy.
Explore Maryland DC condos · Experimental estimatesInspect unit-level recorded sales. Condo testing missed acceptance criteria, so this studio does not offer a calibrated range.
Explore DC condos Fictional example · No sign-inPractice with a clearly fictional Arlington home. Compare prices and illustrative proceeds, then inspect the math.
Try the interactive sampleAscent is a decision aid, not an appraisal or live listing service. Future sale prices and renovation returns are not predicted. Read the published accuracy results and failures.
Public pricing works without an account. Configured private workspaces can connect saved analyses, contacts, offers and deadlines. Some market features require authorized provider data.
Open any sale to see its signed adjustments and weight, test which input moves the estimate most, and compare what each strategy or later move nets the seller.
Ask about the record on screen. See the facts, calculations, assumptions and unknowns behind the answer.
You review the draft. You make the decision.
DC-validated settings, tested on recorded sales held out from development.
Results apply to this market, period and configuration. They do not guarantee the price of your home. Some acceptance thresholds were met narrowly.
Maryland is available: Montgomery and Prince George’s counties. 8 of 9 accuracy criteria met — did not meet the acceptance criteria. Size, quality grade, location and sale timing only; bedrooms and bathrooms are unavailable. Virginia awaits permission.
DC condo pricing is available with a separate unit-level snapshot. Condo accuracy was tested separately and failed. See all published experiments.
These figures use the Montgomery & Prince George’s, MD (revised)-calibrated settings, which could value 94.1% of the holdout sales. With Ascent's general default settings, the median error on the same sales was 10.2%.
Revised experiment: consideration at least $50,000, no bedroom or bathroom inputs, construction quality adjustment capped at $75,000. Three previously unscored months, July–September 2026, with recording lag. Price-related bias still failed. This is a different cohort from the original 13.85% result; it is not a paired improvement estimate. The original failure remains published. Read both experiments. No calibrated range is released.
Maryland SDAT/MDP recorded sales of at least $50,000 (Public Domain public records). Rates were fitted on Jan 2022–Dec 2024, search filters chosen on Jan 2025–Sep 2025, the interval calibrated on Oct 2025–Jun 2026, then Jul 2026–Sep 2026 was scored once. Thresholds were fixed before scoring and have not moved. Report sha256-1c420bbcfa82… can be re-run with npm run validate:valuation -- --verify.
Also published: King County, WA, Jan 2015–May 2015, 10.6% median error on 6,975 sales, 5 of 9 criteria met. Results do not transfer between markets.
These figures use the Washington, DC-calibrated settings, which could value 93.3% of the holdout sales. With Ascent's general default settings, the median error on the same sales was 16.3%.
Met narrowly: the 95% intervals for median error 9.4%–10.6% and share within ±10% 48.1%–52.2% include values on the wrong side of the bar.
District of Columbia recorded sales (CC-BY-4.0 public records). Rates were fitted on Jul 2022–Jun 2024, search filters chosen on Jul 2024–Dec 2024, the interval calibrated on Jan 2025–Sep 2025, then Oct 2025–Sep 2026 was scored once. Thresholds were fixed before scoring and have not moved. Report sha256-d4acbe7087a2… can be re-run with npm run validate:valuation -- --verify.
Also published: King County, WA, Jan 2015–May 2015, 10.6% median error on 6,975 sales, 5 of 9 criteria met. Results do not transfer between markets.
What you can do, where the data comes from, and what stays yours.
No. Anyone can use public pricing studios without signing in. Professional verification and source permissions still control access to restricted licensed data.
Yes, for covered single-family homes in DC and in Montgomery and Prince George’s counties, Maryland. Maryland uses size, quality grade, location and sale timing only; bedroom and bathroom data are unavailable. Both markets publish their accuracy results. DC condos have a separate unvalidated studio; their ranges are uncalibrated. Virginia awaits written permission. Multi-unit buildings are excluded.
No. Public pricing uses bounded public-record snapshots, not real-time listings. Recorded sales can lag, and current competition, concessions and interior condition may be missing.
Not today. You can test assumptions and illustrative costs, but those scenarios are not forecasts of buyer response, renovation return or a sale price in six months.
Recorded closed sales, adjusted for property differences and weighted by distance, recency and size. You can inspect every adjustment and change the assumptions. AI does not set the price.
Open a printable pricing report or download the decision as JSON. Studio drafts stay in this browser on this device. A configured private workspace adds saved analyses, clients, offers and transaction records.
No. This is a decision aid. Public records can lag and omit interior condition, concessions and listing history. Review the evidence before using a draft with a client. No sale price or outcome is guaranteed.
Keep clients, pricing snapshots, offer comparisons and deadlines together. Ask Ascent explains the records and drafts text for review. It cannot send messages, accept offers or sign contracts. New accounts start empty; the workspace demo is fictional.
Start with comparable evidence. Keep a report you can question, share and explain.