It's not a matter of believing the formula. I completely understand the formula. It's a matter of Zorro having determined the risk incorrectly at least on those two trades , and now trying to figure out why, especially if it happens again. Other than the things we already looked at, is there anything else that could possibly result in Zorro incorrectly determining the risk? That way I can investigate and look into that should it happen again.

You've learned with the test script how to calculate the risk and print all parameters leading to it. Why don't you do this with your other script also? If you understood the formula in the test script, you just need to copy the print statement and apply exactly the same formula in your trading script.

Your own risk calculation was wrong, by factor 10 and more, because you were using a wrong formula and a wrong lot size - I told you that several times.

Yeah, I'm already doing that with the test script.

Regarding my risk calculation: 1 micro lot (i.e. 0,.01 standard lot) opened with the broker on AUDUSD is $0.10 per pip. When the ATR is 34.5, then 10*ATR is 345 pips, which at $0.10 per pip comes to $34.50 of risk.

And in fact the strategy is trading the same micro lot size that it used in the simulation, so the strategy is trading the correct lot size.

But as I've said many times, this in no way explains the huge discrepancy in the USDCHF trade, which was a factor of 19.5.

Just do exactly the same as you did with the test script.

Run your trade script in "Test" mode. Check the logfile and compare the displayed risk with the risk and the parameters from the formula. In the next step trade it on your demo account. Again compare the displayed risk with the risk and the parameters from the formula. You can then see all discrepancies. Your calculation above looks ok to me, but I cannot verify your parameters from here. If you get result A and the computer gets result B and you are sure that you're using the right formula, then obviously one of the parameters must be different.