Exactly. It's not clear the threshold is "a few million a year", but there is a line and if you cross it you will have your lunch eaten.
The way i see it, the more scalable a system is, the lower it is quality wise (eg lower Sharpe ratio, MAR ratio, Profit Factor or whatever metric you like to use) . eg. If you want a something with a Sharpe ratio above 2.0 it wont be nearly as scalable as a method with a Sharpe ratio below 1.0. The exception being Rentec Medallion, who have found a way of trading with a Sharpe ratio above 2.0 and can still make Billions a year with it
2 trades per day, 80% of success with a holding period of 1 hour. I am going to call bullshit on this one. Do you know how much profit would you turn into if you compound that return? Something insane.
Measuring the percentage of trades that are profitable is useless. It's very easy to get this number over 80%. Just set a very low stop-loss and take profits when you are a tiny fraction of a percent green on the trade. Presto. You don't need a trading strategy at all to do this. But doing so is useless and will drain your account with spreads and commissions.
Incorrect. Someone else is making that claim. I earn a full time income trading quantitatively but I'm certainly not here to convince you of that. You are free to disregard my advice if you feel the math is wrong.