I know what you mean but I think it is wrong to actually call it random. Random prices would jump all overthe place without any sequential movement at all. I think it is more realistic to call it erratic and unpredictable because it certainly is unpredictable in many ways. The frustration and confusion arising from trying to identify a move is mainly due to the fact that there are so many players in the markets with so many different objectives and horizons. This means that at any one time there is always an imbalance between the immediate supply and demand which wiggles the price and fudges any underlying trend.
One could describe it as micro-focusing on just the ripples on the sea surface in order to establish which way the tide is actually going! 
I would agree with you about the problem with indicators, although this is not due to randomness. It is because they are driven by mathematical formulas that only work when prices moves in a set pattern of speed and volatility. But markets don’t move like that and therefore indicators work occasionally and then they don’t, and then we adjust the parameters to make them work again and then soon they don’t work again. In other words we tend to curve-fit indicators to match the current moves - which then change…
Price Action analysis is a bit different but is still subjective in nature. Many swear that PA is much better than indicators but it is strange, then, that regulated brokers who are obliged to state what percentage of retail clients lose money still report figures between 70-80% even though PA has been actively promoted for years now. If it was so much better than indicators wouldn’t we be seeing some overall improvement in retail trader performance? I am very curious about the stability of these statistics on losing traders!
I think the problem here is that there are probably few such major trends and one has to wait for a very long time before being able to conclude that the trend has ended. I think this is a difficult timeframe to trade on its own if one is going to wait for a month or so before accepting the trend - and then it is too far from the normal daily/4-hour/1-hour environment of most traders to say where and when it has ended.
As I mentioned in my last post here, I don’t think we can talk about predicting because we can never know what is going to happen next and that will be based on fresh info compared with the info that has driven movements so far. But we can talk about “assumption”, and I think that is what all our trading is based on.
It is difficult to comment on your particular situation without knowing more about what you base your entry and exit decisions on. But it is certainly true that people enter trades at varying points as a move gathers momentum depending on their strategy and timeframes and the large institutional participants will add to positions progressively over long periods of time (such as your monthly trends!).
And so it is very valid to make the assumption that once a move starts it will continue for a certain time and distance in the same direction. Others will be entering and exiting along the way but the overall pressure will be trend-based for as long as the current price is considered to be lower/higher than the anticipated target region. Problem is, no one knows where that end balance point will actually be! 
So surely the objective of our TA model is to a) identify a move, b) suggest an entry point, c) in some way suggest a suitable target point to exit, and d) identify the point at which our scanario is nullified and the trade closed out.
If we can design our model to function reasonably well within our chosen timeframe then our profit consistency becomes more a question of exposure management and patience and discipline.
All this is not easy, but it is not outside the reach of diligent and persistent traders.
These are, of course, only my own modest thoughts on these issues, but I hope this helps in some way, or at least prompts some thoughts of your own on your approach that might improve your situation!
. I believe past predicts future but I also think at the point you enter a trade (even during a very strong trend) the chances of it going your way are 50:50 at best! (In my case usually worse than 50:50).
