I'm joyful that you mentioned that. Stochastic are wonderful in a range bound market. Agree. But, as you cannot predict ahead of time if a market is going to be range bound, it is useless. Obviously that's not what I meant when I said they don't work.
Are you actually arguing that an indicator is made to give signals according to its formula and not to help traders pick profitable entries and exits? So if I made an indicator that was previous low price plus current bar range squared divided by number of bars printed so far this day, that would "work?"
By the time a typical indicator tells you something is happening that might influence your decision to buy or sell, it's too late, assuming variable market conditions. You have to compensate by being exceptional at everything else. Where's the argument here?
Very interested in technique on this. Can you elaborate? As a second point - I would like to submit evidence against the validity of indicators. (I'm a new student of these and have no firm bias either way. I'm still learning, so I when I see something, i point it out for discussion). I looked at stochastics on a ticker after seeing your post. On a 30 minute period, it was overbought, on a 1 day period, it was oversold. How can an indicator be considered valid when it offers contradictory output depending on the time frame? Of course, this wouldn't disprove all indicators do not work, but it greatly diminishes their utility, since you would be suggested to buy or sell depending on how far back you pull. Unless you look at different holding periods and different stop settings, one of these outputs is flawed. So what am I missing? thanks.
There will always be time frame conflict whether seen within the context of price derivatives or simply as price action. One's plan has to decide on the time frame(s) relative to the rules of trade. As far as "oversold, overbought", by themselves they are largely meaningless as a sto can stay OB/OS for a long time in a trending market. New traders have a tendency to look for easy simple rules. Once you know what you are seeing and what you are doing, then you can look to simplify as much as possible but until that point, nothing is simple.
All right - as has been said, it varies in application. However, the creator of the indicator gave specific rules about 80/20 OB/OS. If those guidelines don't work, then how can the indicator be useful.
George Lane had a lot more to say than that....BTW Viagra was meant to be a blood pressure treatment. Ya never know what uses a tool might have.
It helps to have a faster stochastic to compare to. Try 14,3,1 and 5,3,2 as a starting point. It won't be clear until one logs crossovers, divergence, convergence, entwined, etc,... and incr, decr, per zone per indicator. Watch when the 5,3,2 crosses over the 50% line for entries. Watch when the 14,3,1 exits the 80/20 line for exits to those entries. There are periods of synchronizations where the above is clear. One line of the pair within the stoch is for early entries. As for early entries, since they are early, they frequently reverse. Waiting for confirmation, since they are waiting, miss out on the beginning of a move. As some point in your observations, they will bring you to looking inside the market on the DOM and T&S - who making and who's taking liquidity.
Just because I cannot does not mean others don't. I think the dude who wrote the book Pit Bull made a fortune off of stochastics back in the 70s. I am sure there are people still using it successfully.