If you have been reading housing headlines before listing this fall, you have probably noticed they do not agree. That is not your imagination, and it is not a market turning. It is a data-quality problem.

Here is what this post covers: the three different July 2026 median prices published for the same market, why the gap exists, why the days-on-market numbers are even further apart, and the pricing method I use with sellers in Henrietta, West Henrietta and the rest of south Monroe County when the public numbers cannot be reconciled. No slogans. Just the steps.

The same month, three different numbers

For July 2026, three consumer real estate sites published a median sale price for the Rochester market:

That is a $56,000 spread on the same statistic for the same month. The top figure is roughly 28 percent above the bottom one.

Sales volume is no better. One source reported sales up 78.6 percent. Another printed 643 homes sold, down from 707. Those two sentences cannot both describe the same set of closings.

So when a neighbor tells you prices are up sharply and a coworker tells you the market is softening, they may both be quoting a real published number. They are just quoting different publishers.

Why the numbers disagree

I cannot audit anyone else's methodology from the outside. What I can tell you is what usually causes gaps this wide:

• Different geography behind the same word. "Rochester" can mean the city, the metro area, a county, or a set of ZIP codes. City-only data and metro data produce very different medians.
• Different date logic. Some feeds count closings in the month. Some count pendings. Some count listings that were active in that month.
• Different property mix. Include condos and multi-family, or exclude them, and the median moves.
• Lag and revision. Recorded sales trickle in for weeks. A number published early in the month gets restated later, and not every site restates.

Here is a related trap I ran into while pulling data. Searching for Monroe County housing inventory returns Monroe County, Pennsylvania, with a $318,500 figure and 2.8 months of supply, plus the Town of Monroe in Orange County, New York. The Pennsylvania numbers look plausible enough to slip into a conversation unchallenged. If you are doing your own reading, check which Monroe County you are looking at before you trust a Monroe County median sale price.

One more, because it shows how loose some of these feeds are: Movoto lists Henrietta's school district as "Monroe County R-1." That is a district in Missouri. The district here is Rush-Henrietta, and if that detail matters to your listing, confirm it with the district or NYSED rather than a listing portal.

The days-on-market numbers are the bigger tell

Price disagreement is one thing. Speed disagreement is where the reporting really falls apart.

• Houzeo, July 2026: median 51 days on market.
• Movoto, July 2026: 13 days.
• Zillow: roughly 8 days to pending.

Fifty-one days and eight days are not variations on a theme. They are describing different things, probably days from list to close versus days from list to accepted offer, without labeling the difference.

It gets stranger. That same Houzeo page for July 2026 reports a sale-to-list price ratio of 120.57 percent. It also publishes a rule saying 45 to 70 days on market indicates a balanced market. So the page tells you the market is balanced while also telling you homes closed at more than 120 percent of list price. Those two statements do not sit together comfortably. A market where the typical sale clears asking by that margin is not a balanced market by any working definition.

I am not pointing this out to score points. I am pointing it out because sellers read one line from a page like that, anchor on it, and then make a pricing decision. If you had only read the balanced-market line, you would price differently than if you had only read the 120.57 percent line.

What I do trust from the July 2026 data

A few figures held up when I checked them, and they matter more than the median anyway.

0.33 months of supply, July 2026. That is the number worth internalizing. Months of supply is inventory divided by monthly sales pace: how long it would take to sell everything currently listed at the current rate. A third of a month is extremely tight. Compare that to the 2.8 months reported for Monroe County, Pennsylvania, and you can see how differently those two places behave.

Homes are separating into two groups. A July 13, 2026 post from Elysian Homes described the market returning to more traditional conditions. Their newer August 5, 2026 post describes something more useful for you: a widening split between homes that sell quickly and homes that sit. That matches what I see going through houses in Henrietta and the surrounding towns. The condition, the pricing and the presentation decide which group a house lands in. The county median has almost nothing to do with it.

I want to be straight about what I do not have. I did not retrieve Greater Rochester Association of Realtors data for this post, so nothing here is sourced to GRAR. I saw a claim that new listings fell 43.87 percent and a claim about mortgage rates near 6.6 percent, and I could not verify either, so I am not printing them as fact. I also could not find any current sold data specific to Pittsford, Brighton, Mendon, Rush, Honeoye Falls or the Finger Lakes towns. Henrietta has town-level figures for median sale price, days on market and active listings, but its sold data stops at June 2026. Months of inventory, new listings and sale-to-list ratio do not exist at town level in anything I found. I am not going to substitute county numbers and call them Henrietta numbers.

How to set a list price when the sources conflict

The Rochester median home price for July 2026, whichever version you pick, tells you nothing about your house. Medians describe a pile of transactions. Buyers make offers on one address. Here is the sequence I actually work through.

1. Ignore the medians entirely. They are a headline input, not a pricing input. No seller has ever been paid the median.
2. Pull closed sales, not listings. Closings from the last 90 days, in your school district and your immediate area, in your style and size range. Listing prices are asks. Closings are facts.
3. Adjust for what the photos do not show. Roof age, furnace and hot water tank, windows, kitchen and bath vintage, basement condition, whether the driveway drains toward the house. This is the part that requires walking through both your house and the comparables.
4. Read what did not sell. Expired and withdrawn listings nearby tell you exactly where buyers stopped following. That ceiling is more informative than any average.
5. Check the pending set. Pendings show current demand better than closings, which reflect offers made weeks earlier.
6. Pick the price that generates the first weekend of showings. With supply as thin as 0.33 months in July 2026, the strategic question is not how high you can print a number. It is whether you land inside the search filter where buyers are looking. Priced a thousand dollars above a round bracket, you disappear from a lot of searches.
7. Write down your response plan before you list. What happens if there are no showings by day five. What happens if there are showings and no offers by day ten. Deciding in advance keeps the decision unemotional.

What this means for a fall listing

Conflicting home price data is going to keep appearing in your feed, and it will keep pointing in opposite directions, because the underlying feeds define their terms differently and revise on different schedules. Do not use them to time your listing. Use closed comparables, your own house's condition, and honest arithmetic about what a buyer will owe monthly at today's rates.

The split described in the August 5, 2026 report is the practical takeaway. Some houses go fast. Some sit. The gap between those two outcomes is mostly things you control: price, condition, access for showings, and how the house presents in the first ten photos.

If you are thinking about listing this fall in Henrietta, West Henrietta, Pittsford, Brighton, Mendon, Rush or Honeoye Falls, I will walk your house with you, pull the closed and expired comparables for your specific street, and show you the math behind the number I recommend. You will see the same data I see, including the parts that argue against my own suggestion.

Book a time with me here and we will start with your house instead of a headline.

About this data

The figures in this post were compiled from publicly available sources including Houzeo, Movoto, Zillow, FRED (Federal Reserve Bank of St. Louis) and RochesterFirst, along with other public market data. Real estate numbers change quickly, and these were accurate as of August 2026. For current figures on a specific home, street, or town, ask me directly rather than relying on a published average.