I thought what she meant was that edges go away when too many are trying to make money with them; sometimes they were never there. Regardless: Bot Destroyer in Traders of the New Era was clear that edges end quicker and quicker as time goes on. He is always on the search for new ones and in the notes to his chapter it was mentioned he backtests as part of his quest.
I would say that they end as quickly as they start if not based on true price movement that repeats with a very high degree of probability.
True, random correlations with increased pnl is not necessarily a true edge. You can find short-lived "edges" using anything, but what you'll find is that they start and end randomly, thus are not really very consistent.
how to be ahead of the crowd? enter where the masses are likely to have their stops placed. let them push prices in your favour. aka. momentum entries. then let the greenhorns chase price further in your favour if you have been the loser before, seeing price take off at your stop, you will understand that you were the trapped sucker. the next step is to understand what the hell happened, that made your stop the perfect entry once you spot such a pattern, pick up a book on classical chart patterns & identify which pattern you have spotted
Yep, an edge should not/would not/could not fail to work, the vast majority of the time, if based on price movement that has consistently been repeated since markets began.
It's all a probability-function, isn't it? I agree with both the comments above, and I suspect that we're all saying more or less the same thing, really, albeit in slightly different words ... They end as quickly as they start, if not based on true price-movement that repeats with a high degree of probability, and that's something that increasingly happens, and is increasingly observed and described and analyzed, as recent trends in IT and "the availability of trading" predicate that there are more participants, observers, commenters and analyzers? I think I'm saying "it was ever thus", really, but there are probably more people noticing it, and commenting on it, these days? (And of course I'd contend that that probably makes it all the more worthwhile to develop skills the herd hasn't developed!).
I am saying winning edges in US stock daytrading literally end quicker and quicker than they used to. Some of stock daytrading has revolved around regulatory glitches (SOES Bandits) that close quicker. Or Innefficiencis (following the short seller in low volume stocks, fishing for NYSE price improvement, etc). And Fall prey to a faster spread of knowledge as well as technalogical progress that moves faster- algos, automation, etc (scalping). ___ Fewer people are able to win daytrading as things toughen. Many of those who do win will find things tougher and their profits less. ___ Seven winning traders I knew from 15 years ago have totally quit or struggle. I am not saying things are so bad no one wins but for those that do, expect to have to learn new ways quicker than in the past. ____ * my comments are for US stock daytrading but would not be surprised if these ideas to some degree to other markets. ** Looking to the future, as AI and quatom computing become stronger and pair together...
So to trade successfully, you have to go with the herd, go in front of the herd, go behind the herd, go against the herd.....go on top of the herd maybe? I'm confuuuuused....