welcome to the forum
not a criticism, but what your system boils down to is that if yesterday’s median price is higher than that of the day before, it’s a buy signal, and vice versa
there’s a simple, attractive logic there
what matters is whether or not, when you run a quick backtest of this method over a decade’s data for any major currency pair, it demonstrates an edge, net of dealing costs
either it does, or it doesn’t - it’s objective and factual
the difficulty is that in order to run such a backtest, you need some defined trading parameters, not just a description of a “signal”
in other words, you need to define (probably by trial-and-error backtesting?) the targets, stop losses and trade management/duration as well
if you can do that, and demonstrate an edge, then you’d have a viable system
sorry to sound like a downer, but i think things like this have all been done hundreds of times before, and the truth is that although the idea is based on something attractive (i.e. you’re trading in the direction of the daily trend), it isn’t in itself going to be profitable over the long term
it’s absolutely essential, one way or another, to have the (very easily acquired) software and (very easily learned) skills to do this kind of backtesting, otherwise all you’re really doing is guessing, and in this game long experience teaches us all that we can’t develop a profitable system by guessing
one has to be able to prove or disprove it objectively, over a statistically significant data sample, and that means testing
without doing that it’s incredibly demanding of time and attention, and for something that may lead nowhere profitable
a week’s or a month’s figures are only “anecdotal”
either there’s an edge there, or there isn’t
you need to know which, to decide whether it justifies the time and effort put into the thing