Every hunch deserves a test before it gets your cash.
You already trade, and the ideas never stop coming. The expensive part is finding out which ones were wrong by funding them. Stax turns an idea into a full trading plan in minutes, replays it across years of real prices, and tells you what would have happened before a dollar moves.
I think defence stocks run this year. Test it before I put real cash in.
I would test it on LMT, RTX, NOC, GD and BA, the five biggest US defence companies. One question first: sell a stock the moment its loss reaches 10%, or hold it to the monthly review?
Sell at 10%. I do not want to sit and watch it fall.
Tested over 5 real years
simulated, not a forecast- Practice account
- $10,000 grew to $12,180
- Worst stretch
- down $2,140
- Trades
- 84, fees included
Weakest through 2022: the plan sold four times in three months and still finished that year down.
Built for the pace ideas arrive at
From hunch to rules in one sitting
Describe the idea the way you would to a friend: which stocks, when to get in, what makes you get out. Stax writes it as exact rules, so the idea becomes something you can test instead of something you argue about.
History does the arguing
The replay runs the idea through years of real prices, with fees counted and tomorrow hidden. A weak idea shows itself in seconds, and finding out costs you nothing at all.
Versions compete, the market judges
Put two versions of the same idea in separate practice slots: one that sells a stock once its loss reaches 10%, another that holds until 20%. They trade the same market on the same days, so the results tell you which rule you should have been using.
Your idea, as you said it
“Solid companies that pay dividends. Sell anything that drops hard.”
$10,000 became
$12,940
worst stretch
-$1,870
trades, all logged
96
The idea held up, but selling at a 10% loss cost more than it saved in two of the five years. Worth running a second version that waits for 20% beside this one.
It follows your rules on the days you would not.
Most trading losses are not bad ideas. They are good ideas abandoned halfway through a losing stretch, or doubled up at the worst moment. A running plan trades the rules you wrote on a calm day, every day, and writes down what it did and why.
When you disagree with a trade, the record is right there: the rule that fired, the price it filled at, and what it did to the account. You can change the rule and test the new version the same night. The old record stays, so you always know what the idea really did.
Room to run several ideas at once
- 20
- practice plans can run at once on Pro
- $0
- to find out an idea would have lost
- 100%
- of trades logged with the rule that fired
Side-by-side paper results show which version handled the recent market best. Treat them as a comparison, not a forecast.
Questions people ask
How fast can I go from idea to result?
Minutes. Describing the idea takes a few sentences, the rules come back for you to approve, and the replay over years of prices finishes in seconds. Testing a second version of the same idea costs you a few minutes more.
Can I run several variations of the same idea?
Yes, that is the intended workflow. Deploy the variations to separate practice slots and watch them trade the same market. The free tier gives you six slots; Pro raises it to twenty for $19.99 a month.
What about trading with a real account?
When a plan has earned it, the top tier connects one plan to one real brokerage account. It is a deliberate step, kept separate from practice on purpose.
Will it stop me from overriding my own rules?
The plan trades its rules exactly as written, so the trade you would have forced at midnight never happens. If you have genuinely changed your mind, edit the rule, test the new version, and start it again. That extra step is deliberate: it keeps a 2am decision from undoing a rule you set on a calm day.
Your next idea can be tested tonight.
Describe the idea in a few sentences and the replay tells you how it would have done, with every trade and the rule behind it listed.