Hi everyone I’m new to trading and I wanna know how do most traders determine accuracy? My 50 SMA on the daily showing this pair in a down trend, but the 15 min chart shows a potential up trend. So far I learned the daily, monthly, and weekly are the most accurate and smaller time frames are for entry.
I think you’re asking about “accuracy of trend direction”?
The key concept is that it’s normal for a currency-pair to be trending in one direction within a specified time-frame, and either not trending or even trending in a different direction on another time-frame, because a “trend” is something that - by definition - exists only within and with reference to a specific time-frame.
When the trend is the same both within the time-frame you’re trading from and within a couple of higher time-frames as well, exactly as you mention, that certainly increases the [I]reliability[/I] of trades taken in the direction of that trend (though never to 100%, of course).
Well, there’s a very big gap between the daily time-frame and the M15 time-frame.
If trading from the M15 time-frame, you’d perhaps want to “confirm the trend direction” from the H1 and H3 (or H4) direction, though. The daily one is of course another multiple above those.
This approach of “checking the next two widely-used higher time-frames” as a “[I]bias indicator[/I]” was explained and described in Marcel Link’s famous textbook “[I]High Probability Trading[/I]”, which is its most commonly recognised source, though of course a lot more has been written about it, too, since then (though most of it nowhere near as helpfully as in the original textbook on the subject, which tends to be the way with these things!).
In general, reliability of “signals” increases as time-frame increases, so you’ve certainly understood the underlying principle correctly.
Note, though, that the fact that higher time-frames have higher directional signal-reliability doesn’t necessarily make them more profitable, overall, to trade from or to consult, because using this kind of approach also greatly reduces the number of trades one makes. It can be better - with appropriate money management - to trade more often and with less overall reliability: it’s better to make an average of 15 pips profit, an average of 6 times per day, than it is to make an average of 30 pips profit, an average of twice per day.
I’m mentioning that as a fairly “random theoretical example” just to make the point that [I][U]reliability[/U][/I] and [I][U]profitability[/U][/I] are two different parameters. But I wouldn’t argue with what you’ve said above, at all. 
My own [B]main[/B] purpose in looking at higher time-frames isn’t actually to check that the trend-direction matches (though I do it for that reason as well): it’s simply to identify areas of recent support and resistance, and carry them over (as horizontal lines) onto the faster charts from which I’m trading.