Beware of Retail Support and ResistanceBitcoin / TetherUS PERPETUAL CONTRACTBINANCE:BTCUSDT.PPrinciplesofmathprobI don't see any reason to short here. People love seeing "resistance". If price bounced there before, how could it not bounce again, right? Then they try to trade there and blow up their account in no time. This kind of thinking is very dangerous because it's static and markets evolve dynamically (see my last post). If someone shorts at what later gets identified as "an important resistance level", how do we know that participant didn't exit his position once price went down, and why would we assume he'll sell more or that more sellers will show up in the area? A more appropriate way of trading and looking at markets starts with identifying the trend instead of levels where price bounced before, and then determining what makes sense to do in that market. In this case, it's clear that it's going up, and the more appropriate play is to buy at support. Not because it already bounced there before, and not because support is high probability, since there's also an enormous number of people who go around saying "this is high probability," and it's all made up; they actually can't give you a number even by accident. The truth is that, if they have any risk-to-reward, what they are seeing is, at best, 50/50. Furthermore, a line or trading context having prior validation, under certain circumstances, actually makes the trade worse instead of better, because it's the direct observation that other people already traded before us, and why would they have incentives to keep trading each time at worse prices? They might, or they might not. Let's look at where they already traded: Does this make it more reliable, safer, a better trade? Not necessarily. That's generally where I want to trade. Now that I know there are already people well positioned there, why would I assume that those people would re-accumulate Bitcoin at worse prices? It can happen, or it may not. People who talk about "high probability" because of this or that variable generally don't have the slightest idea what they're talking about. At best, they might scrape one unit of risk (the same as what they're risking) with a probability slightly above 50%, but that's about it, and if they have any risk-to-reward, believe me, if they have a 60% win rate, it's basically a miracle. Markets don't increase probabilities by having more of something: more bounces, more stop runs, more patterns of whatever it may be. Probabilities change at the precise moment when a side that was imbalanced in one direction loses that "strength" at a critical point and the opposite side becomes imbalanced in its direction and regains control, typically at extremes. That's the real economic mechanism. That's what we constantly trade in this system and call "Test and Retest". When prices reach an extreme, we don't know if they can continue to the stop or reverse; that's the "test". If that test then produces the reaction that we have measured across thousands of cases, then we know that, in the general case, the phenomenon that occurred there is that basic economic phenomenon of supply and demand, where one side of the equation is in a situation where it's no longer economically favorable to keep doing more of the same. Meaning, if, for example, prices have been going down, by the law of supply and demand we know buyers will tend to appear instead of sellers, and the curve tends to shift and subsequently balance out. That's essentially what we measure, which is exactly the same thing high-frequency algorithms measure when they're scanning the order book looking for imbalances and volatility, except instead of doing it on such a small scale looking to grab a few ticks, like those algorithms do (including Market Makers), we want to grab a multiple of what we're risking. Given the conditions we're looking for and that we have measured for the general case (Law of Large Numbers), only then do we enter on a retest at the calculated point. Without worsening our price by chasing it. Placing limit orders so that price comes to us where we already know a new imbalance was generated in the direction of our trade, the "test with good separation" that happens at the extreme. What does this have to do with how many times price bounced before? As far as I know, absolutely nothing. Furthermore, as I said, sometimes it can even worsen the scenario, because if there are people who already traded before, maybe they have no incentive to accumulate more inventory of whatever it is; maybe they got out, or maybe the numbers no longer work for them. Buy at support, the lower extreme; sell at resistance, the upper extreme, if there's confirmation from the market that it's re-accumulating, and most important of all, if it fits our Money Management and Risk Management numbers, meaning the size of the stop, where we place it, and how much we get paid for the trade.