In post #75 of this thread you attached a screen shot of your chart. In the stochastic pane, one bar is apparently highlighted by your software and the value for the fast line, i.e. %k is given as 46.4 Do you know what that 46.4 says about price in your chosen instrument at the moment that the stochastic indicator that you have chosen to base your trading decisions and potentially your future financial well-being upon? Do you know what price would have to do in order to make that value 100 instead of 46.4? In order to make it 0? Do you know, in short, what the stochastic indicator is telling you about price at any given moment? When the stochastic prints a divergence, do you know what exactly is diverging (and I do not mean price vs the indicator - I mean why is there a divergence, what has caused the divergence)?
Before you do that, I suggest that you make certain that you understand "this model." I am not a "know it all" and I sure hope I'm not coming across as one to you because I see so many of them here ET and I absolutely abhor those highly "vocal" posters who seem to know everything about everything even the things about which they are so obviously clueless. For example, so many are so quick to pronounce that all indicators, not just this or that particular indicator with which they have had experience, but all indicators are incapable of generating an edge. Now, I don't mind that so many think that, and the more who think it the better for me. But at the same time when I see someone such as yourself giving this technical indicator thing the old college try I feel I might as well be of some assistance in pointing out what I feel are some basic errors. I recognize these errors as errors because I myself made them as well. Most who have tried TA have made them. You are making them right now. If you keep making them, you'll turn out like one of those know it all know-nothings who start "TA is Bullshit" threads instead becoming a BSD like @Handle123. Look, you have to understand what the indicator is calculating and what that value shows you. A divergence for example, is what you say you are using to signal that a trade opportunity is either present or potentially on its way. So you are looking to enter a long trade, for example, when price makes a lower low, but your stochastic is making a higher low. But why? Do you know why your stochastic will sometimes be at a higher value as price makes a lower low but at other times it will be making new lows along with price? And why is that enough for you to base a trade on? If you learn this, you will learn a lot about price action that you do not now know. You will also find that you will very quickly be able to recognize situations where price is "diverging" from itself. And once you can recognize that, then other tools, such as the DOM or tick stream can become potent tools for you as well. But to learn this, you have to look at that formula you posted above and take it a part. What, for example, does the "highest high - lowest low" represent? Also, make sure as you ponder your stochastic, or any indicator for that matter, that you keep the look back period at the forefront of your awareness - this will help you as you attempt to unravel what the indicator is telling your about current price activity (at least it helped me when I was where you are. You need to take that into account when you start to break that formula down. Do you know why it is called %k (percent k and not simply, k)? Do you know why the quotient is multiplied by 100? I could give you these answers, and no doubt so could @Handle123. But you know who can't? Not one of the close-minded egotistical loudmouths who pronounce TA, PA, or TI or all three to be useless, because each of them has failed at precisely the point at which you are failing: The Beginning. And guess what? I did too. For a long time I was lazy and I thought I was smart. Overbought? Of course! Oversold? Obviously! Divergence? Please ... I wrote the book on divergence lol. I was such a cocky dumbass, let me tell you. I would backtest the crap out of something, but it was years before l I finally realized that I didn't even know WTF it was I was backtesting, and because of that I couldn't begin to design a proper system to back test, let alone a proper back test lol. As I said, I could give these answers to you, but you really should figure it out for yourself. And focus on the %k until you figure out what that is telling you. The %D was just a filter added on and filters tend to be more harm than good in my experience. The only filters I ever found worth a damn was the concept of trend and range. Learn your stochastic, find a way to own the concepts of trend and range, and I think your efforts will be far more fruitful than they have been to this point. And the reason that will be is because you will get yourself focused on price. You might think you already are, but you are not. Please believe me, I know. I was there. You are not anywhere near price itself. And indicators can only be useful tools in your trading tool box if you know how they are derived from price. You see, while most who criticize technical indicators are quick to do so in part because they are "derivative," that is a red herring. It is because these tools are derived from price that they can be, in the right hands, very, very useful. I never intended to write that much, so I will shut up now. Best to your trading, iwilldoit. I hope you do.
TRADE ONE: SHORT/PROFIT/ BRD/60/15H30 Finally a profitable trade I realized there was BRD on the 60 at 15h30 → nearly missed it hadn't I checked (was in a bleak mood today). 15h33 I set my limit order to 105.50, TARGET= 103.75, INITIAL STOP = 107 EXECUTION Price rise up to fill my limit order at 105.50 at 15h34 (USING the DOM) → IN TRADE PERFORMANCE Then I see price move up to 6.25 to hit the 18SMA→ got a bit scared→ but hey I'm a tough guy I dont mind getting stopped out. EXIT After Price hit the 18SMA, it dropped to quickly hit my PT at 103.75 as planned (textbook) Then price dropped like a rock in an ocean → if I had two positions I would of left it. WHAT I LEARNT Need to be more mentally prepared→ I was thinking about non-career issues prior to noticing the divergence Perhaps my targets are a bit too shallow→ I have noticed that on profitable trades → price does run a bit more past my planned target. I wonder how this would of played out in a real live "money" trade? It is also worthy to note that there was indeed BLD/60/9H30→ but I had work commitments at that time→ also it would of resulted in a losing trade based on the model's ruleset. So this time I missed a losing trade and was available to take a profitable one. Stochastics was also at overbought levels on both the 30 & 60→ perhaps adding extra confirmation to the trade? @Handle123 @Redneck @NoDoji @Buy1Sell2 @aquarian1 @Gueco @pak @slugar @zbestoch @dratsum @wrbtrader
Dear @zbestoch, Thank You so much for this post, I always value sincere and specific useful advice, and am always open to constructive criticism. I wish I was available to reply to this post earlier. I have been backtesting models using Stochastic Divergence for the past 15 months, and according to the backtest I find it quite reliable on larger timeframes. I have to be honest, and admit, that no I do not entirely know the relationship between the candle price action and the stochastic formula, but your post has definitely motivated me to research and understand this vital relationship. Nevertheless, I have heard from various sources, you should have a full understanding of any indicator and how it is derived. Now I am going to take a look at the 30 and 60 charts, and look at the formula and see the highs and lows from the past 14 candles→ and see how K% is calculated Please do not "shut up" and continue to make positive contributions to my thread. Have a wonderful evening, and I am lucky to have you on my thread. @Handle123
One of the things I did in late 80s was learn the indicators I was using then and change values of closes highs lows over several months, I wanted to learn by sight of what happens by making one tick differences and what the indicator does, took long time as you have to go into data itself and make changes, reams of charts I made on printer, so I know what indicators should do 99% of the time. Now what I also learned in this time was price patterns, better idea of what bars form to continuation of trend and when price should make a reversal. I have learned much of breaking down both indicators and price. I know better than ever saying what I learned about MACD is same like stochastics, but I love using indicators when they show visual and making it easier to see what is happening. I see way too many on this forum that generalize that indicators are useless and their way is best, so selfish of a stance, it is usually the same folks that never share charts they use to trade, they never show their losing trades or how they mucked up cause they read chart wrong or how many times they have mucked up in their trading, it is always same crap of saying how someone else makes a life and living is doing it wrong. HAAAAAAAAAAAAA
A great exercise would be to put a short term stochastic on a 5 minute chart and then watch in real time for a few weeks. Replay would also be good, except that replay isn't really replay as it is usually at a much quicker pace and the feel of the trading day is lost. Did you identify what it is this represents?: (Highest High - Lowest Low) where Lowest Low = lowest low for the look-back period and Highest High = highest high for the look-back period What does that represent? What was George Lane using math to define for the purpose of calculating the indicator's current value?
Ok so this is my attempt to understand Stochastics: Here I have a 60 chart which displays the Bear Divergence at 15h30 today. The Three candles I will be performing calculations on are denoted 1,2,3 in red The Lookback period is 14 candles which is shown by the red line to look back until the 5h30 candle This is the formula for the Fast Stochastic: %K = (Current Close - Lowest Low)/(Highest High - Lowest Low) * 100 %D = 3-day SMA of %K Lowest Low = lowest low for the look-back period Highest High = highest high for the look-back period %K is multiplied by 100 to move the decimal point two places Here are the calculations: CANDLE 1: %K = (2106 - 2089)/(2110.25 - 2089) * 100 %K = 80 Look at the pink ellipse it matches... CANDLE 2: %K = (2108 - 2089)/(2109 - 2089) * 100 %K = 95 CANDLE 3: %K = (2103.25 - 2089)/(2103.75 - 2089) * 100 %K = 96.6 So it seems that in order for %K line to cross over %D> at 15h30, the 15h30 candle would have to close below Candle 2's (14h30) close of 108! (I recognize divergence when it is confirmed with a crossover) Then %D is simple the 3 period average of these calculated %K: 95+96.6+80/ 3 = 90.5 (which can be verified by the chart image I posted) I think I am starting to understand it? Did I miss anything? That being said, I am not sure how to calculate the slow stoch using an exponential smoothing factor? (not on google) This is actually the type of stochastic I am using to trade. @Handle123 @zbestoch
From what you showing, I don't see Divergence, you don't have enough chart and your status box is overlapping indicator. Plus once Stochastic has gone very low, it negates everything to the left.