Automated strategy
Insider Purchases
+110.63% since January 4, 2021 · AEX +63.81% over the same days
Return since start+110.63%AEX +63.81% over the same days
Value on Sep 22, 2026USD 210,628from USD 100,000 · 1,469 trading days
Positions opened131from 131 entries the rule produced · 10 still open
- This rule’s backtested portfolio
- AEX, from the same amount on the same day
- Both lines run from USD 100,000 on the day of the first entry; a shorter period shows the part of both that falls inside it. The reference stands next to the portfolio and is never subtracted from it.
About
the rule, in wordsEvery Monday this strategy scores every listed company by adding up the Insider Score of each open-market purchase its insiders reported in the preceding year, after weighting each one by how old it is: a filing counts fully on the day it appears, half after 45 days, a quarter after 90, and about six per cent after half a year. It holds the ten highest in equal weights. A company that enters the ranking is bought on the next trading day; one that drops out is sold; one that stays is left alone. The ranking is measured on notification dates only, and only on filings that were public before that Monday.
It starts at USD 100,000 on January 4, 2021 — the day the rule first had something to act on — and the AEX stands next to it from the same amount on the same day. Each trade costs 0.1% one way, 0.2% round trip.
What it does not do
- The score is our own model, not a rating anyone else recognises. It weighs the transaction, the insider and the company; it is calculated once per filing, on the day that filing became public, and it never looks at what the share price did afterwards.
- It is long-only. There is no short leg and no leverage; a company leaving the ranking is sold, never sold short.
- The backtest is in-sample. The model was designed and calibrated on the same filings the curve runs over, so it says what this rule would have produced on data we already had — it is not a forecast, and it is not an out-of-sample test.
- The notification date is mostly an assumption. The Dutch regulator does not publish when a filing arrived, so for almost every row we use the transaction date plus three working days: the latest moment the law allows. That is deliberately late, so the strategy never acts earlier than it could have.
- It is one register. These are filings by insiders at companies supervised by the Dutch AFM — six in ten of the positions trade in Amsterdam, the rest elsewhere in Europe and beyond. It is not a European index and not a global one.
- Prices are converted. The portfolio is measured in dollars, so a euro, krone, zloty, rand or pound price is converted at the ECB rate of that day — which means part of the return is the currency, not the company.
The 14 key figures
as of Sep 22, 2026Return, 1 day+0.41%AEX +0.35%
Return, 30 days−3.57%AEX −1.76%
Return, 1 year+14.32%AEX +15.93%
Annualized return+13.93%AEX +9.02%A year's worth of return over four months would be an extrapolation, so it stays empty until there is a year.
Maximum drawdown−39.03%AEX −35.45%The deepest fall from a peak to the low that followed it.
Beta0.78How much the portfolio moved when the index next to it moved. 1 means it moved along one for one.
Alpha, annualized+7.83%Jensen’s alpha: the part of the return that beta does not explain — not a subtraction of the index.
Sharpe ratio0.62Return per unit of its own volatility, with a risk-free rate of zero.
Winning positions58.0%The share of all positions that ended above what they cost. A position at exactly zero counts in the denominator, not in the numerator.
Average win / average loss+24.52% / −14.39%Per position, and independent of how much went into it.
Volatility, annualized25.3%AEX 17.8%
Information ratio0.26The difference with the index per unit of how much that difference itself moved.
Treynor ratio0.20Return per unit of beta.
Transactions executed252Entries plus exits — the reported lines that actually moved the portfolio.
Every figure here is a calculation over filed forms, made afterwards, under assumptions that are written down: what the portfolio starts with, how much goes into each position, what closes one, and what a trade costs. A backtest says what a rule would have produced on filings that were already public. The definitions are on the methodology page.