Portfolio Stats
One strategy on its own tells you very little. Put several in a group, open the Stats tab, and HorizonAI runs them together as a single account: one balance, one equity curve, one worst drop. If you've told the group what you're aiming for (a profit target and the risk limits you can't breach) it also tells you the thing you actually want to know: whether you get there, and how long it takes.
Open it
- Go to My Scripts.
- Click a group in the left sidebar.
- Switch from Scripts to Stats.
Pin a backtest first
Every strategy in the group needs one backtest pinned: the run it should be known for. A strategy can have twenty backtests across different markets and timeframes; the pin is you saying this is the one that represents it.
Open a strategy, find the run you want in its backtest history, and pin it. Until a strategy has a pin, the Stats tab leaves it out and tells you so rather than quietly guessing.
What it shows you
The balance, up top. What your starting balance turned into with all of these running together, over the stretch of time where they were all running. If two strategies were tested over periods that barely overlap, we say so instead of putting them on one chart anyway.
Worst drop. The furthest the combined account ever fell from its high point, in money and in percent. If you set a max drawdown on the group, this is checked against it and flagged when it goes past.
How alike. The one most people never check. Two strategies can have different names, different code and different indicators and still be the same bet in disguise, both long-biased continuation on the same market, tested on the same year. When that happens you don't have two strategies, you have one strategy at double size. This tells you in plain English how much overlap there is, and names the two that are closest.
Teamwork. How much steadier the group is than the strategies that make it up. Above 1 means they're genuinely sharing the load. Around 1 means they're all moving together and you've gained nothing by adding more.
The chart. Your combined account as a bold line, with each strategy on its own as faint grey lines behind it, so you can see whether the group rides steadier than its parts. Your money is split evenly between them and evened up again each day.
The challenge question
This is the reason the page exists.
Set a profit target % and a starting balance on the group (plus a max drawdown and a max daily loss, if you're trading to prop-firm rules) and the Stats tab answers directly:
About seven weeks to make $10,000, and it gets there 68% of the time without breaking your limits on the way. Drops too far 9% of the time.
We get that by replaying your own trading days in thousands of different orders and seeing how often you reach the target before you breach a limit. It gives you three things at once: how long, how often it works, and what kills it when it doesn't: too deep a drawdown, one bad day, or simply running out of time.
The trade size slider
One control on the page: trade size. Drag it and everything above recalculates at that size.
Trading bigger multiplies the profits and the drops together, so it's the fastest way to see the real trade-off: a target you'd reach in three weeks at 3× that also blows your drawdown limit a third of the time. We suggest the biggest size that keeps the worst drop inside your limit with room to spare.
It assumes each strategy risks a share of your account. A strategy trading a fixed number of contracts won't scale like this.
If the page looks empty
It always says why rather than showing a blank panel:
- "This group is empty": add some strategies to it.
- "None of these has a pinned backtest yet": open one and pin the run it should be known for.
- "Tested over periods that barely overlap": the strategies were backtested over such different stretches of time that combining them would mean comparing different years. Re-run them over a common window.
Fine print
- Everything here is built from backtests: trades worked out against past prices, not trades that were placed. Putting several together always looks better than any one alone, so treat it as a guide rather than a promise.
- The window is the overlap of all the pinned runs. Adding one strategy tested only over the last three months shortens the window for the whole group, and the page names which strategy narrowed it.
- Deleting a group never deletes the strategies in it.
