How Stochastic value is used?
- Thread starter hkaddah
- Start date
Stochastics were one of the first indicators I fell in love with when I first started trading. It's a beautiful indicator in hindsight. You look at it and say, "wow, every time it crosses 20band from 0, the price goes up. Look there, the price went down, and wow, stochastics just crossed back under 80 band!"
Then you try using it in real time, and the results are somewhat different. Yeah, for sure, you get to catch some amazing reversals every now and then, and it's pretty useful in oscillating markets. Of course, as anyone who's tried it will tell you, you get absolutely slaughtered relying upon it when the market trends.
From my own experience, I think a big danger in attaching yourself to stochastics (especially if you've had initial success with them) is that you'll soon begin to see every dip as "oversold" and every rally as "overbought". Why is this dangerous? Well, I'll paraphrase I quote I read on these boards and say, "the market can remain overbought (oversold) for a lot longer than you can remain solvent."
I personally found, after trying indicator after indicator, (and the heart ache and anguish that comes with picking tops and bottoms) that the only really consistent way to make money in the market is by going with the trend and having a "feel" for the market you trade. (Successful bottom pickers, and I know you're out there, feel free to flame me.)
Also, after trading and looking at stochastics for long enough, I think you can have a pretty accurate idea of where the stochastic would be at without even having it there.
I know this doesn't answer your question regarding which length stochastics are best for particular styles, but they're such a "pet hate" of mine I had to chip in with my .02.
Given that there is some "contorversy" about what exactly what constitutes a "scalper" as opposed to and "intra day swinger", I think it would be better if you defined the styles for us as you understand them. The simple answer though would probably be, for scalping use a very short length stochastic (or depending on how you scalp, don't use one at all!) and lengthen it as your time horizon/profit objectives broaden.
As for what indicator to use along with stochastics, I would recommend using any trend following indicator, to give some "counter balance".
Daniel
Then you try using it in real time, and the results are somewhat different. Yeah, for sure, you get to catch some amazing reversals every now and then, and it's pretty useful in oscillating markets. Of course, as anyone who's tried it will tell you, you get absolutely slaughtered relying upon it when the market trends.
From my own experience, I think a big danger in attaching yourself to stochastics (especially if you've had initial success with them) is that you'll soon begin to see every dip as "oversold" and every rally as "overbought". Why is this dangerous? Well, I'll paraphrase I quote I read on these boards and say, "the market can remain overbought (oversold) for a lot longer than you can remain solvent."
I personally found, after trying indicator after indicator, (and the heart ache and anguish that comes with picking tops and bottoms) that the only really consistent way to make money in the market is by going with the trend and having a "feel" for the market you trade. (Successful bottom pickers, and I know you're out there, feel free to flame me.)
Also, after trading and looking at stochastics for long enough, I think you can have a pretty accurate idea of where the stochastic would be at without even having it there.
I know this doesn't answer your question regarding which length stochastics are best for particular styles, but they're such a "pet hate" of mine I had to chip in with my .02.
Given that there is some "contorversy" about what exactly what constitutes a "scalper" as opposed to and "intra day swinger", I think it would be better if you defined the styles for us as you understand them. The simple answer though would probably be, for scalping use a very short length stochastic (or depending on how you scalp, don't use one at all!) and lengthen it as your time horizon/profit objectives broaden.
As for what indicator to use along with stochastics, I would recommend using any trend following indicator, to give some "counter balance".
Daniel
Magna
Administrator
Well said, a downright remarkable indicator in hindsight, my experience exactly.
I think the quote you were referring to was by John Maynard Keynes when he said, "The market can remain irrational longer than you can remain solvent." But I like your paraphrasing, pretty damn accurate.
Another appropriate quote is a recent one by our own BrandonF when he said, "Trend days have a habit of becoming very overbought/oversold, and then becoming more so."
34,3,21 for all time but not daily, for daily I use 15,3,3
34,3,21 generate signal when line cross, 15,3,3 generate signal when cross 20/80 line. MACD is often use to confirm the stoc. or another way around, it is all depend on how fast/slow when you set MACD, I use 5,13,6 on all time frame except 1m, for that multiply it by 5. Overtime, you will change the setting because market change and people change.
daniel,
well said. stochastics is a sexy indicator that will kill you. Even George Lane, who pretty much invented it and teaches classes on it, says not to use it the way most people do, ie buy when it crosses 20 and sell when it drops below 80. He advocates using it to spot divergences, ie whne price makes a lower low and stochastics makes a higher low for buying, reverse for selling, then use a confirming indicator such as a trendline to get you in the trade. As for what period is best, it depends on your market and your trading style. Obviously, a longer lookback period gives a smoother response and less noise, a shrter one gives you faster signals. To derive an optimal period you need to know what the dominant cycle is in your market, then I think the idea is to use half the cycle length for the lookback. Personally, I never found cycles very useful.
well said. stochastics is a sexy indicator that will kill you. Even George Lane, who pretty much invented it and teaches classes on it, says not to use it the way most people do, ie buy when it crosses 20 and sell when it drops below 80. He advocates using it to spot divergences, ie whne price makes a lower low and stochastics makes a higher low for buying, reverse for selling, then use a confirming indicator such as a trendline to get you in the trade. As for what period is best, it depends on your market and your trading style. Obviously, a longer lookback period gives a smoother response and less noise, a shrter one gives you faster signals. To derive an optimal period you need to know what the dominant cycle is in your market, then I think the idea is to use half the cycle length for the lookback. Personally, I never found cycles very useful.
As with most technicals, you'll see great patterns After the fact. We try to get our people away from "spacing out at chart patterns" looking for some signal that will help them make a decision. Trading is so much easier if you look at why things are happening, rather than just what did happen.
A simple example: When you see a trough on stock, support and resistance at say, $20 and $30. Many times something as simple as dividend yield may be the cause of the pricing action. When the stock is $20, then their dividend may "beat the street" (as far as interest or ROI goes, but at $30 it no longer makes sense, so people sell the stock back). This applies to investors more than traders, but you get a lot of the same intraday reasoning for traders as well.
Don't get me wrong (gotta keep the peace), of course we understand all the basic TA, and the basics of Fundementals as well (that is just a given), but we don't concern ourselves much with the intraday charting.
Don
A simple example: When you see a trough on stock, support and resistance at say, $20 and $30. Many times something as simple as dividend yield may be the cause of the pricing action. When the stock is $20, then their dividend may "beat the street" (as far as interest or ROI goes, but at $30 it no longer makes sense, so people sell the stock back). This applies to investors more than traders, but you get a lot of the same intraday reasoning for traders as well.
Don't get me wrong (gotta keep the peace), of course we understand all the basic TA, and the basics of Fundementals as well (that is just a given), but we don't concern ourselves much with the intraday charting.
Don