Signals & Measurement
We Tested Six Reasons a House Sells. Five Were Noise.
Bottom line: we measured six things people treat as a reason a house is about to sell. Five of them did nothing. A late property-tax bill does not predict a sale. Neither does the value of the house, its age, or how many names are on the deed.
There is a rate at which a house changes hands whether anything happens to it or not. In St. Charles County, over six months, it runs around 2.5% to 2.8% depending on which homes you compare against. A signal is only worth something if it beats the rate its own comparison group already moves at.
Most of them do not.
The scoreboard
| What we tested | Effect on the odds | Verdict |
|---|---|---|
| A probate filing on the home | 5.1x | It fires |
| Falling behind on property taxes | 1.28 odds ratio | Ruled out |
| What the house is worth | flat | Ruled out |
| How old the house is | flat | Ruled out |
| How many names are on the title | 0.92–1.20x | Ruled out |
| Both co-owners having died | no separate inventory | Ruled out |
The one that fired: 14.1% of St. Charles homes with a probate filing recorded a real, priced sale within six months, against 2.80% for homes matched to them on price and length of ownership. And it survives the test that separates a real signal from a kind of house that simply turns over a lot — run the same measurement on the same homes starting one, two and three years before the filing, and they sold at or below the ordinary rate every time.
The one we most expected to hold did not. Deeper tax delinquency should mean more selling if financial distress drove sales. It runs backwards: 1.79 at 31 days past due, 1.28 at 60 days, 0.72 at six months. A ladder that descends as the trouble deepens is running the wrong way for a distress signal. We can say the effect is absent. We cannot tell you why, and we are not going to guess.
Why publish the misses
Each of these went through the same program as the one that worked. A research program that only ever discovers good news is not measuring anything.
The full report walks through all six — how the comparison group is built, the before-the-event test, what the five-times figure does not mean for a buyer, and the four things we still cannot tell you.