You can't be a consistently profitable trader if you're looking for absolutes, if you need to know for sure how many traders trading how much volume are going to come into the market to support your price excursion desire at a given point in time. Mark Douglas tells it like it is below. If you're uncomfortable with this environment and you're unable to become comfortable with it, then just move on, because there is no Holy Grail of Certainty. You have to learn how to think in probabilities. In other words, you have to get your expectations aligned with the way the market actually exists. And when you do, when you learn these kinds of mental skills and you’re able to execute your trades without fear, without hesitation, without analyzing, or even without thinking for that matter because you don’t need to think… I’ll give you an example: What a professional trader thinks about when there’s an edge present is…does he think about whether the edge is going to work? Absolutely not, because he has learned there’s no point in analyzing or judging or building a case for or against whether that trade is going to work, because he understands the human component. But, what he does think about is he thinks about the risk. How much do I have to risk, how far am I going to let the market go against this position to tell me that other traders are either going to come into the market and make me a winner or not, and he also has a plan for how he’s going to take profits. What does the typical trader do? The exact opposite of the professional. Once they make up their mind it’s a winning trade, they don’t pre-define their risk and they also don’t have a plan to take profits because they think it’s going to go on forever. We don’t want to get into trading with the possibility of being disappointed, with the possibility of being dissatisfied, or being even betrayed, because a lot of traders feel that way; they really feel betrayed, and the problem is that when that potential exists it has the effect of affecting the way that we see market information in detrimental ways. All of us have these mental pain avoidance mechanisms that affect our perception of information. So, for an example, if I’m in a losing trade and I got into this trade thinking I was going to be right (in other words, I did all my evaluation, I did all my analysis, I did my work, I built a case), as the market’s moving against me, I’m going to have the tendency to focus on information that tells me that I’m right and ignore the information that tells me that the market is actually trending against me. I can identify a trend, but I won’t be able to identify that trend if I’m putting an inordinate amount of significance on the information that’s telling me that I’m right…[and] ignoring the information that’s telling me I’m wrong. The problem is, if I’m susceptible to being disappointed or betrayed, meaning I get into a trade expecting it to do what I think it’s going to do, I’m going to have this tendency to distort market information, that causes me to hang on to my losers, and in a winning trade what’ll happen is that instead of letting a winner run…it’s the retracements we focus on instead of the fact that the market’s still trending in our favor. Most traders, because they evaluate, because they judge and because they analyze, and build a case for the pattern being right, they actually talk themselves out of believing that the risk even exists. There’s no way to know the sequence of wins and losses. If we want to be able to trade our methodology in an effective fashion, to be able to utilize this methodology in a way where we can extract the maximum amount of profit that it makes available to us based on the pattern that it identifies, we have to do it in certain ways. Our mind has to be free to be able to execute these trades without making trading errors and the trading errors come from believing that because the pattern is present, that it’s going to give me a winning trade on THIS one; THIS trade is going to be a winner. You can’t think that way. That’s the way the typical trader thinks. The typical trader thinks ‘I’m not gonna put on this trade unless I think it’s gonna be a winner or why would I do it?’” “Trading a technical methodology or a technical pattern does not have anything to do with being right or wrong. It’s just an odds game. You’ve got to be able to take every single trade because you don’t know the sequence of wins and losses. You’ve got to be able to identify what your risk is and that’s simply ‘How much am I willing to spend to find out if other traders are going to come into this market and bid it higher than my price or offer lower than my price if I sold?’” The solution is to change your mind, to change the way you think. “[You’ve] got to eliminate the potential to think that the market’s going to disappoint you. And the way [you] eliminate the potential is by understanding that trading is not about being right or wrong. It’s a probability game.” There are stages of development such as learning how to think in probabilities so the market doesn’t have the potential to cause us to feel emotional pain. “When you put on a trade and it doesn’t work, all it really means is that some of the traders didn’t come into the market that had the same belief that you had, or the same conviction about this market doing whatever it is you thought it was going to do. You have to learn to walk away.” We can’t predict collective human behavior. “The methodologies that we have access to, these mathematical formulas, do that for us. But you have to understand that there’s no possible way that these mathematical formulas can predict the outcome of these patterns on a trade by trade basis, only on a series of trades. So when I get a signal from my methodology, at the most fundamental level what this is telling me is that the odds are in my favor that somebody is going to come into the market (this is what the pattern means) and bid it higher than here if I bought or offer it lower than here if I sold. That’s all that it’s saying. Now they’re either gonna come or they’re not, and so as a result I don’t look at this as being a ‘right’ or a ‘wrong’; I look at this as ‘How much distance am I going to give the market to move away from my entry point to tell me that they’re either going to come or they’re not, and any further is not worth the cost of finding out.’”
Which is exactly what I said here: http://www.elitetrader.com/et/index...-via-price-action.281995/page-97#post-4040933 And, yes, it does take a while to find n instances of a given setup (or hypothesis) in a given bar interval and determine the probability of success. That's why so few do it. But it's also why so few are successful.
That is an excellent quote ND. I absolutely don't want to be dealing with absolutes. Each moment is different at Douglas would say. This is partly why I didn't get around to testing enough appearances of the BOPB setup. The little rule of the bar having to close above the trendline just seemed too much of an absolute. If one bar is off by a tick, this may mean it doesn't close above or below the line. If a swing point that is used for the 5 min trendline is off by a tick, this could be the difference of a couple of points once that line is extended over a few hours. Being aware of this, I wondered how on earth am I going to collect decent stats when I'd in fact have to be a little "generous" with what I'm looking for since I don't want to be so firm, so absolute with precisely what I'm looking for down to the tick. Its just like with levels.. having read some old stuff from Db recently and also from Wyckoff. The levels are almost more like zones. Its nice when price stops at the exact tick at a potential support level that you are looking at, but even if it pushes through by a point and quickly rallies back, this doesn't invalidate the support level. So being aware of the need to be a bit loose, to not require tick precision, means that the testing was now a bit more difficult. This always brought me back to behavior because along with seeing where price goes, where it cannot go, how long it takes to get there, the pace, the hesitation, all of these things in a way help to see what is going on, so although a level might be breached by a few ticks, what happens right after the breach is quite telling. This is of course all lost on the static chart and there is no way to backtest this. If I'm going to use this in trading live, then I can be called out for placing a trade because of a "thought", since its very difficult to backtest this, unless you have years of experience to draw on and your brain has been collecting the stats. So I still have no way to rationalize the idea about trading behavior vs. just trading bars. Don't get me wrong... I think just trading bars is easier. There is less to think about, less to worry about, less concentration is almost needed. I am absolutely all for the statistical nature of this endevour. All my fear comes from not having the statistics of this trade worked out that might be setting up before my eyes. Geez, in his journal from about 6 years ago I think, was a wonderful inspiration because he knew his 50% win ratio and his 1:2 risk to reward ratio. If I had this worked out, I know I could trade it, but since I don't have these stats to fall back onto, I'm never quite sure of where my stop should be that completely invalidates the trade, nor do I have a profit target worked out. So each trade is therefore met with apprehension. But when you get into Wyckoff, and also so much of what Db says, it focuses on the market telling you what to do. I know that you have your profit targets and stop losses worked out to the tick ND, but by following what the market is telling you, a trade that you thought might only get 10 points based on where the next level is might actually give you 20 points if it blows right past the level where you thought you'd find resistance. Likewise, an exit for a loss can generally be made even before your automatic stop loss is hit, and I think you've even said this before as well. So it seems that these profits targets and stop losses are quite fluid, and then in this case, the collection of stats seems fluid as well. The last thing I want to do is argue with all you lovely people that are trying to help me, but I'm just trying to figure out what absolutes are necessary and what aren't. There is the absolute of having a trading plan... the absolute of testing a particular setup at least 100 times, the absolute of not thinking about a trade and having to just put it on because your parameters to put it on are met, but clearly the absolutes also stop and things do get fluid very quickly. The years of experience of course allows you to be fluid, and each rule can be broken as long as you know how I suspect. But its just difficult for this new guy to put it all together because not everything lines up. Of course if it did, more people than just a select few would be successful at this I take it.
Yup.. this is what backtesting is all about I guess. But I gotta tell you.... I'm fighting this because then its just basically creating an automated system... perhaps even an automated scalping system so that enough of these setups happen every day or every few days to make it a worthwhile enough of a trade worth pursuing. My other problem of course is having the pre-existing conditions well formulated. With everything I know thus far, its important to know where you are in the trend in reference to trend channel limits and the mean, how close you are to daily support or resistance, where the hourly swing points are, etc. So this makes backtesting extremely difficult. Now of course I can just throw all this out and use a 1 minute chart and just look at what's happening an hour before the open and go from there, but this now means that the all important context isn't being backtested, and Db has already cautioned me about this. Don't get me wrong, if I developed a scalping system that worked well enough for 2 or 3 points I'd be delighted, but I'm fairly sure the algos have this worked out much better than I do, and my brain just cannot compete with the computers that never get tired and have a shit load of stats pre-loaded and ready to apply as needed.
If it were easy, then everybody would be doing it. You're rationalizing your reluctance to do the work in order to justify the reluctance. Perhaps you ought to ask yourself how many years you're willing to continue down this path in hopes that there might eventually be a revelation. "Since I don't have these stats to fall back onto": why not?
Think of it more as collecting stats on your favorite setup irrespective of context. You won't ever have confidence in taking said setup if you don't know how it performs. Ok, I can now see why your head is spinning w all these things you think you need to test - right away. Why not 1) test your favorite setup as recommended above. Once you have collected a fair amount of stats as to how the setup performs on its own, 2) THEN try looking at the context and look at context as a means to filter (prevent) taking the setup in certain conditions
Yes, I can certainly see how its feeding onto itself. Ok.. I guess no point in continuing to go back and forth. I will just look for 100 instances of something and go from there.
Ok.. one last time before I go silent for a while. O - Just before the open, we had a great rally overnight, and on my chart is this high from Nov 3 that we just broke. The selling starts right away though. A - We hit a low of 65, go up, stopping exactly at the 50% level. A short right here would be our first RET. I also mark in a 5 min SL that I had on my chart. Price does close inside/above it, but the fill would happen below the line. Given that we couldn't go above 50% of this initial down move, and given that the 5 min SL broke, this should be a high probability trade. B - As we come down some more, I note that at 4148, we have the 50% of the move up from the low yesterday where we magically turned at 4120 (the Sept high) to today's high at 4176. It looks like we might be holding here since we come up strongly. C - We form even a HL. D - Drats, but now a LH. The nice thing about this is we will have a hinge, and since this hinge happens here in the middle of nowhere, its gonna be a solid hinge, not like a hinge in a range which is not that useful. E - We drop out the bottom, and it sure does drop hard. So we wait for the retrace back up, which goes just as high as where it broke out, and we put a sell stop below this bar here. These are pretty big bars here... 5 points high, which I consider to be quite big. F - Our short works, but we only go down as far as this. Yes, this level has been sitting on my chart for a while now, since the 3rd, and all I've done is made this line thicker now since it clearly is more important now. G - I mark in this long. If the trade triggers, we can certainly say that for the time being we had support at 42 again, and if price should drop back down, because this bar is fairly small, this might be only about a 4 point risk. This of course means that I've given up on my short from E, although that could have been closed for a couple of points of profit. H - We successfully go through the apex of the hinge, but stop below the 50% of this drop from the high at 76 to this low at 42. I'm not saying that I would be looking for a short just because we stop here for now. It certainly looks like we are heading up, this has been the trend now for 15 minute since we bounced off 42, but since we could not go past the 50% level of this drop, if we are still holding some shorts from the top, then this gives us reason to keep holding. Yesterday we had something similar in that we had a big drop, but on the 5 min chart you could see a nice retrace back up that initially didn't break this 50% level, but then we formed a higher low and eventually broke this 50% level, and price just kept going higher. So these 50% levels can be quite telling. If we break them, often we keep going up, and if we cannot break them, price resumes the move in the initial direction. I - Well back down we come, going right through the apex of the hinge which seems to be providing nothing juicy today. Here we have a congestion area. I'm hunting for a long, which would end up being the higher low from F. I can see a nice RET in here to the right, which would fill, but look at the huge drop after. Why I'm hunting for a long I'm not sure given that price is coming down and we rejected the 50% level, but I guess I'm just wanting my friend at 42 to hold as support again. Of course even if a long was taken here, the ultimate test is at 42, so we are quite a bit above this which makes a long here risky anyway, not a trade that should be considered until we have more confirmation. J - We drop below and this drop looks solid. K - Slowly we climb back up after our drop to test this level as resistance now, and yes, price bounces off. Lets short below this bar. L - So that one fills, but we form a higher low from the previous low at 37 that caused the penetration. But here we have another REJ of 42... so this short still looks solid. M - Drats, now we break through... so much for resistance. N - Now we almost even have a test of this level as support again and price just shoots up. Well... these levels can be a bitch. The worst scenario is where price goes above and below, testing both moves several times. Who knows, maybe this level isn't even all that prominent and I'm making it out to be more than it is. A long above "N" would have obviously worked beautifully though. I might have gotten chopped up a bit by trading long and short around this level, but I do think that the small loses would easily be covered by getting into the move that finally took off. CONCLUSION Ok... so moving forward, lets start the back testing. 5min/1min BOPB was a great trade today. Not sure if its legit based on price closing on the wrong side of the 5 min trendline, but I can't have this much precision, so the fact that the 5 min trendline was broken and we had a solid retracement makes this is a winning trade. So the score is 1:0, with 99 to go. The hinge setup worked really well today, its just that support was not far below. Being aware of this, a few points could still be made. So score 1:0 for a nice hinge setup by making sure the hinge is in the middle of nowhere and waiting for the RET on the 1 minute chart and then taking the trade 1 point above or below the crest/trough... 99 more to go here. The other juicy trade today was that level at 42. This one is tricky because I have to have my levels right, which takes lots of work just to setup and backtest. But if price is at a level that you think is going to be important based on previous action around this level, if the bar is sufficiently small enough that bounces off this level, taking the reversal can be a solid trade, and if price happens to break through and retrace back up, taking the continuation trade after the retrecement is another solid trade (this is a case of the level first acting as S and then R). But like I say, this one is tricky to test because its dependent on getting the levels right. The 5min/1min Multiple Time Frame Approach (MTFA) is also tricky because as I outlined yesterday, getting the lines right also makes a difference, but I guess if I wasn't lazy, there would be no reason why I couldn't track several and see which one works best. The hinge is really nice because if its not in a congestion area, it provides a beautiful move. I also like how the hinge, based on what I read, acts as a reset, so whatever happened before gets reset after the hinge exit. So with a hinge, you only really need to focus on the hinge I think. If you want to take the exit right away with a market order, go for it. If you wanna be safe and wait for the RET, this is fine too, but your price might be terrible, like 5 points away from where the hinge actually exited. On top of this, you gotta be aware that price may very well take off in the other direction eventually. But if you are prepared for any outcome and know what to do, then this should be a piece of cake. As a final note, the last price bar on my chart is at 53, and since then we are having trouble getting past 57. Well it just so happens that the 50% level from the top at 76 down to the new low of 37 is a hair under 57, give or take a tick. Just saying.....