What the scores deliberately do not cover
Every index is a set of choices about what to leave out. Most sites hide theirs. These are ours, with the reason for each, because a gap you know about is useful and a gap you don't is a trap.
There are things a reader reasonably expects a place-comparison site to score, that this one does not. Each omission is a decision, and in every case the decision is the same one: a number that cannot be produced honestly for every town in the index is worse than no number, because it will be read as fact and compared against other towns where it means something different.
Crime
This is the most requested and the most firmly excluded.
The FBI's crime reporting has its floor at the police agency. A city police department reports one set of figures for the whole city, so every neighbourhood inside that city — the quiet end and the difficult end — carries an identical number. Reporting is also voluntary in practice, with agencies moving in and out of the programme between years, so a town's apparent change can be an artefact of who filed.
Publishing that as a neighbourhood safety score would mean assigning a city's average to places that do not resemble it, and doing so with the authority of a grade. Instead, safety is one of the twelve questions put to residents, where it is at least an honest description of what people who live there report.
Which school your address attends
The national dataset of school attendance boundaries is a decade old and only about half of it represents real catchments; the rest is approximated. Attendance zones are also redrawn regularly in exactly the fast-growing suburbs people are most often moving to.
So this site does the two things it can do accurately: it names the districts that serve a town and the share of the town each covers, and it lists the public schools nearest the town centre with their distances. Which school a specific address attends is a question only the district's own lookup can answer, and it is worth answering before you make an offer rather than after.
Comparing schools across state lines
Covered at length in its own guide: states write their own tests and set their own pass marks, so a national proficiency score is confidently wrong. School proficiency here is always a percentile within the same state.
Weather, politics and "vibe"
Climate data is available and genuinely national, and it was still left out, because within a metro it is nearly constant — every suburb of one city shares its weather — and between metros it is a preference rather than a quality. A score that treats sunshine as better would be an opinion wearing a number's clothes.
Politics is the same argument with sharper edges. The data exists at county level, which is too coarse for suburbs and too charged to present as a grade.
The test for including something is not "can we get the data". It is "does the same number mean the same thing in every town in the index".
Property listings and accounts
Neither exists here. Listings are a different business with different incentives, and the moment a site earns money from moving people it stops being able to tell them a town is a bad fit. Accounts would let the site remember preferences, at the cost of collecting people — and the resident poll works perfectly well with an anonymous cookie.
What to do about the gaps
Use the index for what public data does well — cost, housing, schools at district level, commute, amenities, employment — and treat it as a shortlisting tool rather than a verdict. For crime, ask the local police department for neighbourhood-level figures and read a local forum. For schools, use the district's boundary lookup and visit. For the feel of a place, spend a Saturday and a Tuesday evening there.
A site that told you it could replace those steps would be lying to you.