SeeFundsee funds clearly 🔍
Saved Sign in

Methodology

SeeFund · last updated 2026-09-30 · every number on this site comes from one pipeline, described here

The point of this page is that you can check us. If a number on a fund page disagrees with this description, the page is wrong and we want to know — see reporting an error.

What the numbers mean

Every fund page answers the same question: what did this fund actually do, using one window, one method, and the fund's own history. There are no ratings, no scores out of ten, and no forecast.

FieldDefinition
Total returnEnd value ÷ start value − 1 over the window, with dividends and distributions reinvested and the fund's own expense ratio already deducted.
Annualised (CAGR)The constant yearly rate that turns the start value into the end value over the same window: (1 + total)1/years − 1.
VolatilityAnnualised standard deviation of daily returns (√252 scaling).
SharpeAnnualised return ÷ annualised volatility with a risk-free rate of zero. It is a ratio, not a grade — a higher number is not a recommendation.
Deepest drawdownThe largest peak-to-trough fall inside the window, with the two dates it happened between. Recovery time is shown where we can measure it.
Every calendar yearYear-by-year returns, computed from the same daily series — compounding them reproduces the total return (we test this).

The window

Unless you change it in the tool, the window is the last 10 years, or the fund's whole history if it is shorter than that. The data date is printed on every page, because a static page can go stale and you should be able to see exactly how old the numbers are.

Our market data starts in 2006. Where a page says “max”, it means “as far back as our data goes”, not “since inception” — for funds older than 2006 the first decade of their life is simply not in our data, and we would rather say that plainly than imply a longer record.

Where the data comes from

Prices, adjusted prices, dividends and splits come from Tiingo under a paid licence that permits public display, and the attribution appears in the footer of every page. We compute returns ourselves from that data: the numbers on this site are our calculation, not a value copied from a vendor screen. That is also why you will sometimes see a fund's return here differ slightly from another site — different dividend treatment, different window, or a vendor error somewhere. When two independent sources disagree we assume we might be the ones who are wrong until we have checked.

What we deliberately do not do

  • No forecasts, targets or “expected” returns. Everything is labelled as history.
  • No buy/sell signals and no advice. SeeFund is a self-analysis tool; it never tells you what to own. Terms of Use.
  • No fund “scores”. Metrics are reported with their definitions so you can weigh them yourself.
  • No hidden data. The same engine drives the pages and the interactive tool, so a static page and the app cannot drift apart.

Known limitations

  • Coverage is a screened set, not every fund that exists. A fund needs roughly three years of usable history before it gets a page. That keeps empty shells out, but it also means very new funds are missing. Coverage is listed on the Funds hub.
  • Some fields are blank on purpose. Where we do not have a fee, a yield or a prospectus detail from a source we trust, the row is left out rather than filled with a guess. An empty cell is honest; a made-up number is not.
  • Leveraged and inverse products are excluded from the page set: their daily reset makes a “returns since 2016” curve misleading without a long explanation we would rather not bury a page in.
  • Past performance does not predict future results. A drawdown of −33% means an investor who bought the peak watched a third of the value disappear; it says nothing about the next one.

Reporting an error

If a figure looks wrong, email support@seefund.app with the page URL and the fund. Data errors are the one bug class we treat as urgent, and corrections ship with the next daily rebuild.

See the method applied to a fund you own: open the Backtest tool — same engine, your own window. Or start with the fund pages.