I made my trading system grade itself against just buying the index

A rising equity curve is the most persuasive lie in trading. Your account is up, so the thing you built must be good. Except the market was up more, and if you had done nothing at all, put the money in a broad index fund and gone to the beach, you would be richer and you would have slept better. A green line only means something next to the green line you would have had for free.

So the last piece of Aegis, and the one that changed how I read everything else, is a benchmark that refuses to let the system grade on a curve.

What it measures against

On the day the experiment started, Aegis takes the same amount of money each account was seeded with, imagines putting all of it into the index and holding, and tracks that shadow position every day after. Every account is then plotted not in dollars but as a percentage return since day one, sitting right next to the index doing nothing. The dashboard leads with a single line: how many of the accounts are beating buy and hold, and by how much.

It is a small thing to build and it is brutal to look at, which is the point. It takes the flattering framing away. The question stops being "is my account up" and becomes "is my account up more than the free option that took no effort and no risk of me getting it wrong."

Recording how, not just how much

Return alone is still not enough, because two accounts can end the month at the same number having taken wildly different risks to get there. So alongside the outcome, Aegis records the path of every trade. For each position it keeps the furthest it went in my favour and the furthest it went against me before it closed.

That one addition makes losses legible. A trade that went up eight percent and then gave it all back to a stop is a completely different failure from one that went straight down from the open. The first is an exit that was too rigid. The second is an entry that was wrong. Without the path you cannot tell them apart, and every stop out looks like the same shrug. With it, each closed trade gets tagged by its shape, and the shapes add up into a diagnosis.

The numbers I did not want

Here is where the honesty tax comes due.

Over the run so far the index has climbed several points. The accounts are roughly flat. Put those together and my system is behind the do nothing option by about five points. Measured against the only benchmark that matters, the clever thing is losing to the lazy thing.

And the path data surfaced something worse than a return shortfall. The stop that is supposed to cap a loss at eight percent has, on gap days, slipped past it and closed nearer twelve. The floor I designed is not the floor I am actually standing on. For a system whose entire pitch is safety before cleverness, that is the finding that stings, and it is one I would never have seen from the equity curve alone. It only showed up because I was recording how each trade got where it got.

Beat it substantially and safely, or do not bother

That benchmark reset the bar for the whole project. The plan was never to beat the index by a hair. A hair is noise, and a hair is not worth the complexity, the eight accounts, the nightly jobs, the reconciling, any of it. The system is only worth carrying forward if an account can beat the index substantially and safely: a return that is clearly ahead, and a smaller drawdown getting there, not just a marginally higher final number.

Substantially is easy to check. Safely is the harder half, and it is the half I have barely tested, because the whole run has happened in a rising market. Stops and drawdown controls are cheap to admire when nothing is falling. The real exam is the first real decline, and I am holding the experiment open to the end of the year specifically to sit one and see whether the safety was real or just untested.

What I would tell you to steal

Not the strategy. The strategy is behind the index and I have said so all the way through this series, which is the whole brand. What is worth stealing is the framing.

Grade against the free option, always. Put the boring benchmark right at the top of the page where it can embarrass you, because the moment it is one click away instead of front and centre, you will quietly stop clicking. And record how your wins and losses happened, not just that they happened, because the shape of a loss is where the fix lives. The equity curve tells you the score. The path tells you the game. If you only keep one, keep the path.