An edge is being well capitilized, from that position and using a methodical stop out system one can do very well. It doesn't require a sophisticated system, many simple systems work. Hence one reason new traders struggle, they start off undercapitilized and often the fear of losing their meagre money sees them abort a position at the wrong moment.
17) Follow the herd―just don't be the last one in line. 18) Do not bottom-fish or pick the top. Trends usually aren't over until you're stopped out. 19) Profit/loss is determined not by where you enter but where you exit. 20) Entry anticipates the future. Exit reflects on the past. . . .
Intuition is something that the trader has yet to find a way to put it into words (and/or numbers), and include it as part of his mechanical trading process. Everything can be placed into an algo.
I was referring to trading algo. Your link is describing driverless algo. It's still work in progress and they have yet to input a driver's "intuition" into the algo. That's why there are issues. The examples quoted from the article: A person standing in a crosswalk looking at their phone is probably distracted, for example, so the driver knows to yield. A group of kids playing around on the sidewalk could tumble into the street, so caution is advised. A motorcyclist at an intersection touches their feet to the ground, so it's probably safe to proceed. The AI developer needs to have all these into the algo and more that are not listed.
Trading is harder than driving a car. Ergo. If you see a person from behind, in a crowd, at a distance... based on context and heuristics, you can very quickly assess with high accuracy if you know that person. Without ever even accessing your mental database of all the people known to you, and without the required data to even perform an accurate search of your mental database. A computer can't do that.
I agree with the driving part. lol Recognising someone from behind does the dig into your memory even if for a split second. The computer can store that action/posture/movement in it's database , do a query and match it. Every problem in the world that can't be solved is just the lack of a creative solution or probably a current technological incapability.
In my opinion hundreds if the success rate is high, possibly thousands if it is low. Some edges do disappear quickly over time.