Jamaica Plain, Roslindale and West Roxbury run into each other along the southwest edge of the city. Cross Washington Street or the far end of the Arboretum and you have changed neighborhoods without changing much else — same housing stock, same commute, often the same buyers touring all three on a Saturday.

So it would be reasonable to expect three similar market readings. That is not what the data shows. The three neighborhoods look far more alike than they look different — but inside each one, single-family and condo are running two separate markets.

What the Number Means

Market Temp is our 0–100 balance index. Fifty is balanced. Higher favors sellers, lower favors buyers.

It combines six things we track for every area and every property type: months of supply, the share of listings already under agreement, the sale-to-list ratio over the trailing year, how long a typical listing takes to accept an offer, the recent pace of closings, and the count of homes actively for sale. The scale is calibrated against the Greater Boston distribution rather than the national six-months-of-inventory convention, which was built for a very different market than this one.

One thing it deliberately does not do is forecast. It describes the conditions a buyer or seller is walking into this week.

Single-Family Is the Tight End of All Three Markets

Roslindale is the tightest of the nine at 67. Eleven single-family homes were listed for sale, against 1.7 months of supply — the lowest figure on the board. Thirty-nine percent of those listings were already under agreement, the typical one took seven and a half days to accept an offer, and over the trailing twelve months single-family sales there closed at 103.8 percent of list price.

West Roxbury reads 56 with a deeper bench: 33 homes listed, 2.2 months of supply, 31 percent under agreement, a median of 11 days to an accepted offer, and 101.8 percent of list across the year.

Jamaica Plain reads 51, the most balanced of the three. Fifteen homes listed, 2.6 months of supply, 25 percent under agreement, 18 days to an offer, and 101.2 percent of list.

The common thread is worth stating plainly. In all three neighborhoods, single-family homes have been closing above their asking price on average, and in none of them does supply reach three months.

Condos Carry the Choice

The condo readings run 46 in West Roxbury, 44 in Jamaica Plain and 38 in Roslindale.

Roslindale holds the sharpest contrast on the board, and it is the reason to look at property type before geography. In the same neighborhood, in the same week, there were 37 condos listed against 11 single-family homes. Condos showed 3.8 months of supply against 1.7. Sixteen percent of condo listings were under agreement against 39 percent of single-family listings. Buyers looking at condos there have roughly three times the selection and considerably less competition than buyers looking at houses a few streets away.

Jamaica Plain carries the largest inventory of any cell on the board: 70 condos actively listed, 3.1 months of supply, 22.5 days to a typical accepted offer, and 100.7 percent of list over the year. That is a market where a well-prepared buyer has room to consider options, and where a seller benefits from being priced against a genuinely long shelf.

West Roxbury sits between them at 2.7 months and 20 days, with condos closing at 99.1 percent of list across the year — slightly under asking, on average.

Multifamily Is Thin, and It Reads That Way

Jamaica Plain reads 28, West Roxbury 41, Roslindale 50, and all three rest on somewhere between five and seventeen active listings.

Jamaica Plain showed 17 multifamily listings against 5.1 months of supply, the longest on the board, with sales closing at 98.1 percent of list over the trailing year. West Roxbury had five listings and a 96.9 percent sale-to-list figure. Roslindale ran warmer, with nine listings, 3.2 months of supply and 102.3 percent of list.

None of the three had enough multifamily closings in the last ninety days to publish a median days-to-offer, so that figure is blank rather than estimated. These readings are directional, and we would rather say so than dress up a number that rests on a handful of transactions.

Where September Sits in the Year

Market Temp is a this-week reading, which makes it worth knowing what week it is. Closings in all three neighborhoods follow the same annual shape, and they have followed it two years running: a winter trough in February, a climb through spring, a peak in June and July, and a drop off that peak in August.

So a September reading is taken on the downslope of the year, not at its peak. That is true for all nine cells equally, which is why they are still fair to compare against each other — but it is worth holding in mind before reading any one of them as a permanent condition.

What This Means for You

How to Read These Numbers

Every figure comes from MLS Property Information Network listing and sales data that we collect, screen and map ourselves, with neighborhood boundaries we draw rather than MLS areas or ZIP codes. Listings are read from a daily archive of the MLS feed and are current as of September 23, 2026. Sale-to-list and pace figures cover the trailing twelve months.

Sales are screened for data errors before anything is computed, and a sale with one unreliable field still counts everywhere except the single measure that field would corrupt. Where a figure is blank, it is blank because the sales behind it are too few to carry meaning — three qualifying sales for a days-to-offer median, ten for a distribution. Nothing at the top of the market is trimmed, because a bidding war is a real transaction.

Market Temp is computed per property type and never as a blended figure across all three. There is no such thing as a single price or a single market for a neighborhood that contains both a two-family and a condo, and averaging them produces a number that describes neither.