Essays
The number we took out.
A watch bought in 2007 for what a small car cost, now worth roughly the same nominal figure, has not held its value. It has lost about a third of it. Inflation did that quietly, and no price chart shows it.
Showing that correction seems obviously useful. It is arithmetically sound, it is the honest number, and it corrects a bias almost every collector carries. We built it, shipped it, and then removed it.
A correct number answering the wrong question
The problem was not the arithmetic. It was what the figure implied about the person reading it.
A real-terms gain is a statement about an investment position. It says: measured against what your money could otherwise have done, here is how you did. That framing is appropriate for an asset held to produce a return. It is the wrong framing for an object somebody bought because their father wore one, or because it marked a job they were proud of, and intends never to sell.
Put a real-terms loss beside a watch like that and the application has made an argument nobody asked it to make: that the piece was a poor decision. It was not a decision of that kind at all. The number is accurate and the implication is false, which is the most dangerous combination a metric can have — a reader has no way to reject it, because the maths is right.
The test that decided it
Every figure in a record should answer a question its owner would actually ask. "What is this worth now?" is such a question. "What would I get if I sold?" is another. "Did this outperform inflation over eighteen years?" is a question an accountant asks about a portfolio, and a collection is not a portfolio, however much its spreadsheet resembles one.
A metric that fails that test does not get refined, softened, or hidden behind a preference. It gets removed, because a number on a screen is read as a claim whether or not it is labeled as one.
What survived, and why
Inflation itself is still there, doing a different job. Restating a model's launch price in today's money is genuinely useful context: it tells you what the watch cost its first owner in terms you can feel, and it often reframes what looks like a modern price rise as roughly flat.
That is a fact about the watch. The version we removed was a claim about the owner. The distinction sounds small and is the whole of it: inflation describes the object, never the person's financial position.
Where the software fits — and where it does not
Watch Vault Archive converts historical prices using published national inflation data, and names both ends of the conversion under every figure so the arithmetic can be checked against any public table. What it does not do is tell you whether a watch was a good investment. It holds no market data, cannot value anything itself, and every valuation in a record is the owner's own estimate unless independently appraised.
One thing worth doing today
Take the oldest purchase price in your records and find what that sum would be in today's money. Any public inflation calculator will do it in a few seconds.
Then decide, deliberately, whether that changes anything about the watch — or only about how you were thinking about it.