20 EMA and Patterns

[JMC 10]

Also, a 20 ema on the 5 min chart is equivalent to a 34 ema (or 33, close enough) on a 3 min chart.

 

I don't have any problem believing there's a lot of good entries. Enough pics have been posted on that. The obvious thing left to do is to calculate the tipping point for the MAE and the average target which gives you the best stats.

 

Otherwise, I think we're in Woodies CCI land here where everyone who believes in it believes it's a winner IF just the right trade management were uncovered...but there are no winners in the long run on the brokerage statements.

 

[Disclaimer: I wasted the first year of my futures trading education (8 yrs ago) looking into the Woodies CCI snake oil. I saw 1000's upon 1000's of good historical setups. But it is and always has been a losing system]

 

That's a good observation. There are many winning entries that look obvious in hindsight. However, in real time, there are many entries at or near the EMA that would result in getting stopped out more frequently than you'd expect. Obviously, you could increase your winning percentage by waiting for some type of confirmation (ie. candlestick pattern). But, of course, you get the tradeoff of taking more intial risk, i.e. larger intial stop loss.

 

In terms of MAE/MFE, I think it would be very helpful to keep track of these stats. My experience in CL has been that an intial stop loss of in the range of -.15 to -.20 and a target of +.60 to +.80 seems to capture most typical moves. However, we all know that typical doesn't always happen as every day is somewhat unique.

[phoenix01 10]

From peoples perspectives and experience do you think a fixed target or trailing stop is better? I can't see how a fixed target can work when volitilty and conditions are always changing. Iam trying to trail behind candles that go in my favour but having varying success.

[joshdance 10]

From peoples perspectives and experience do you think a fixed target or trailing stop is better? I can't see how a fixed target can work when volitilty and conditions are always changing. Iam trying to trail behind candles that go in my favour but having varying success.

 

I have found that in order to set up an initial workable trade idea, I need to have at least some idea of a target. I fought this for some time as I wanted to capture the huge moves. However, an initial target determines feasibility of a good R:R for starters, and gives you at least a plan. Then, if you discover that the movement gains momentum, you can move your target further away, and if price gets to your initial target location, place your stop pretty close behind and give it a chance to push through. If it pushes, you have at least guaranteed close to your initial target while giving yourself the opportunity to make more. If it doesn't, then with a close stop you can still capture the bulk of what you would have captured anyway.

[phoenix01 10]

I have found that in order to set up an initial workable trade idea, I need to have at least some idea of a target. I fought this for some time as I wanted to capture the huge moves. However, an initial target determines feasibility of a good R:R for starters, and gives you at least a plan. Then, if you discover that the movement gains momentum, you can move your target further away, and if price gets to your initial target location, place your stop pretty close behind and give it a chance to push through. If it pushes, you have at least guaranteed close to your initial target while giving yourself the opportunity to make more. If it doesn't, then with a close stop you can still capture the bulk of what you would have captured anyway.

 

How would you suggest you determine that target when markets voltility is always changing. When some days bounces lead to large extensions because possibly some news came out and fuel it further and sometimes small bounces which you have to take before it comes back for your stop.

[joshdance 10]

How would you suggest you determine that target when markets voltility is always changing. When some days bounces lead to large extensions because possibly some news came out and fuel it further and sometimes small bounces which you have to take before it comes back for your stop.

 

Base your target on the current volatility -- like, what are you seeing today so far, in the last 10 minutes even?

 

I think there's no easy answer to your question. How many markets do you trade, and what hours do you trade them?

[brownsfan019 17]

Exits are tough, no doubt about it. Sometimes you will buy the lod or sell the hod; and that same trade the next day better be a 'hit and run' or you will get stopped.

 

I've eventually come down to a few things that I look for...

 

1) I'd like to have my target greater than my risk (not always per trade but most).

2) I don't like big stops, so I try to enter at good spots and identify where my stop could reasonably be placed.

3) Win more than you lose.

4) You have to find what your risk tolerance is that will allow you to trade w/o having emotions interfere (the less the better).

 

If you can keep the numbers in your favor, you can come out ahead most days. Some days are easier than others. For example, last Fri was a difficult day for me as the chop/slow ate up trades and then a day like Tue or Wed was just gorgeous (on the ES) for me. Same setups - entry and exit - just some days are easier than others.

 

Contrary to what some on this forum or others will tell you, no system is perfect every day. Any reasonably decent system should make money during certain market conditions and your goal is to make the rough days as least painful as possible.

 

As for how to exit, there's a few threads around TL as I went on some very long and exhaustive hunts for the 'best' exit. My end result - I use fixed targets. Sometimes those exits are perfect and sometimes they are terrible in hindsight. At this point I've realized that I will probably never develop a perfect, consistent exit method so I'm better off dealing with it and making sure my entries are the best they can be.

[phoenix01 10]

Base your target on the current volatility -- like, what are you seeing today so far, in the last 10 minutes even?

 

I think there's no easy answer to your question. How many markets do you trade, and what hours do you trade them?

 

I trade European indices and the ES so usually two markets. Trade from the European open till the close and therefore catch the first two hours of the US session.

[phoenix01 10]

As for how to exit, there's a few threads around TL as I went on some very long and exhaustive hunts for the 'best' exit. My end result - I use fixed targets. Sometimes those exits are perfect and sometimes they are terrible in hindsight. At this point I've realized that I will probably never develop a perfect, consistent exit method so I'm better off dealing with it and making sure my entries are the best they can be.

 

How have you created that fixed target? Is it just from experience from watching the markets or using an indicator. How soon did you adapt to say the recent vast increase in volatility. It completely threw me, I was so set in my ways of expecting this much profit per trade on average, having tight stops and moving to breakeven quiet soon.

[joshdance 10]

I trade European indices and the ES so usually two markets. Trade from the European open till the close and therefore catch the first two hours of the US session.

 

Cool--I ask because I have found it much easier to get into the rhythm of a single market. A while back I was watching CL and ES at the same time, and it proved too tiresome for me to try to focus on more than one at a time. My reason behind this is that for you gauge the current market's mood, you really need to be focused, and you really need to know the character of the market you're trading. I think all markets are essentially equally hard to trade; however, each one has its own personality, and the better you know it, then the better position you are in to be able to determine the probability of the move going in your favor. Basically, what do you see happen over and over? If you've watched a single market day in and day out, 6+ hours a day, 5 days a week, for a year, then you should have some good idea of what it likes to do, particularly how it likes to behave on average around certain times of the day.

 

For example, in a CL market which has had some good activity in the morning and has slowed down around 11:30 or noon, I will take a trade and hope to get 20 or 30 out of it, not much more; if I decide to hold longer, I can be pretty sure that I will be holding for 2 hours to get a larger target, and will be guaranteed to sit through some back-and-forth, and may even find a 50 tick gain completely erased. That's just typical behavior for that market. I know that, and so it allows me to put the current low volume activity in context, and base the target off of that.

 

Conversely, if it's pre-market, say 8:30am, and price has based, and I get good signals from price and volume that we have consolidated and are ready for a move up, then I can expect that if we do break north out of the basing range, that we will revisit the VWAP, or perhaps the daily pivot, or perhaps the overnight high, or some other measure of value, so if I take the risk to get in near the bottom, then when we break north, and IF (this is a big if) the mood of the market gains some momentum in my favor, then I may be more likely to hold the trade longer for the target. Or more likely, I will take the trade off at +40 or so, and put a limit near the top of the consolidation to buy back, and so on.

 

Learn how the market you're in likes to behave, and develop some basic plan of action. Base your target on some area where you have observed that other traders are likely to counter your direction, and trade accordingly.

[brownsfan019 17]

How have you created that fixed target? Is it just from experience from watching the markets or using an indicator. How soon did you adapt to say the recent vast increase in volatility. It completely threw me, I was so set in my ways of expecting this much profit per trade on average, having tight stops and moving to breakeven quiet soon.

 

It's just based on each market. On the ES for example, my risk is typically 1.5 or lower. As long as I make 2.0 on winners and have enough of them, the #s work. On the CL the risk is usually 10-15 ticks, so I like to keep my winners at 20 or more.

 

Intraday trading, esp in this volatility, can lead to very quick retraces or some extended runs. No matter what your exit method is, you just have to capitalize on that - either get your profit and out or make sure your winners can run.

[ake 10]

From peoples perspectives and experience do you think a fixed target or trailing stop is better? I can't see how a fixed target can work when volitilty and conditions are always changing. Iam trying to trail behind candles that go in my favour but having varying success.

 

If I catch an intraday trend reversal (failed retest of a high or low after the break of a trend line) I'm more likely to trail the stop. If I entered on a pullback to the EMA I use a fixed target.

 

For fixed targets I use horizontal lines of support or resistance, which I draw in a one hour chart by loading 3 months worth of one hour data, zooming out until I can barely tell a bar from the next and identifying trading ranges that turned into reversal points. Then I draw a horizontal line, sometimes on the higher boundaries, sometimes on the lower ones, depending on how price behaved around that price on different days.

 

I place a limit order to take profits at the first horizontal support or resistance line that is more than three times the size of the risk away. In the EUR/USD 1m with a 34EMA this is generally between 30 and 40 points, with a risk of 7 points. In the DAX 1m is about 20 points with a risk of 5 or 6 points.

 

I protect profits as price moves towards my target. I move the stop to break even once price has moved in my favour the size of the initial risk, and I move the stop to protect profits about the size of the initial risk once paper profits are twice that.

 

I have found that my returns are better when I stick religiously to the target and don't get out early because I thought I saw a loss of momentum.

[phoenix01 10]

I have found that my returns are better when I stick religiously to the target and don't get out early because I thought I saw a loss of momentum.

 

I find this extreme difficult to do however from backtesting it seems to produce the best results.

[ake 10]

BB - what do you think of the 2nd arrow? It got through the EMA but gave an inv hammer. Would you consider a short there or no b/c it busted the EMA?

 

 

For whatever that's worth, I've recently back tested this in the EUR/USD and the DAX.

 

The outcome is that the probability of success appears to be marginally lower but the potential profits wouldn't be, on average, different from trades where the entire body of the signal bar is below the EMA.

 

This wouldn't deter me from taking these trades. However, I don't take them because they add a degree of ambiguity: how far above the EMA is OK?

 

I find it easier and less stressful to look for the simpler, "clean" version of the setup --the signal bar opens below the EMA, pierces it and closes below it.

[ake 10]

I find this extreme difficult to do however from backtesting it seems to produce the best results.

 

More frequently than not, I get stopped out on my way to the target, with a profit about twice the risk.

 

Because of this, the rationale behind how I set the target as well as its accuracy may not be as relevant as I'd like to think. Any arbitrary target will probably do, provided that it is far enough, we protect profits on the way there and the win/loss ratio adds up.

 

Sometimes I find it easier to stick to the target when I'm not sure if that particular support or resistance level is accurately drawn.

 

This may be because I stop thinking that the rationale for the target is correct and price should be attracted to it. I just accept that I may get stopped out for a profit and I'm less likely to exit at market. Sometimes these are my most profitable trades.

[JMC 10]

Oil getting sold today along with stocks. Two nice entries at the EMA into a couple of large moves around 8:50am and 11:34am EST.

 

[ake 10]

Obviously, you could increase your winning percentage by waiting for some type of confirmation (ie. candlestick pattern). But, of course, you get the tradeoff of taking more intial risk, i.e. larger intial stop loss..

 

Thanks for this, JMC. I've been trying out recently entering with a limit order, with a stop order a tick above the high or a reversal candlestick and at market at the close of a reversal candlestick.

 

I can't comment on the last one because I'm still "forward testing" it. However, in regards to the first two, I've found that the optimal way to manage the trade immediately after entering is around these lines:

 

1. Upon entering long with a limit order, tighten the stop to a tick below the newly formed low as the market starts to move in favour of the trade

 

2. Upon entering with a buy stop order, tighten the stop loss to a tick below the entry bar at its close

 

I've found that although, as you mentioned, the risk initially is higher entering with a stop, the average loss is smaller if the stop loss is tightened as above.

[ake 10]

I've come across two books that discuss this pattern from different angles.

 

The first is "Winning the trading game", by Noble Drakoln, 2008. Drakoln says he uses the 9EMA, the 20SMA and the 50SMA to trade off a daily chart. When price pulls back to any of these levels, he waits for two to four bars for an appropriate candlestick to setup, then enters the market "immediately". He discusses other ways of trading the moving averages as well.

 

The other one is "The encyclopedia of trading strategies", by Katz and McCormick, 2000, where the authors focus on automated testing of entries when price bounces off moving averages on daily charts. Take this with a grain of salt, but the outcome of their research is that, in sample, the optimal entry is with a stop. Out-of-sample, the limit order performed best.

 

There is a considerable amount of detail in this book, which lends credence to the author's research. As an explanation of why a stop entry performs better in their tests, they speculate that optimal entries appear to have both a counter-trend element (entering against the pullback) and a with-trend element (entering with a stop).

[pat15 10]

Of course I think context is the most important. To use an EMA as confirmation is the best use, IMO. In this case, we had support just above prior double bottom, lower highs at the top, and a false break out of the triangle-shaped formation, which trapped enough longs (see the volume for the bar and the delta) to get enough shorts interested in taking it down. The EMA provided nice confirmation and additional incentive to shorts.

 

 

Josh you had mentioned delta ..is it available in ToS and more imp how does one uses it. Very nice thread and following it now.

 

Pat

[pat15 10]

I am going thru this thread ..Someone mentioned about the NoDoji on Elite Trader ... I am curious see her posts or charts or what setups she uses. On ET I could find her nick and under Author but cud not find any setups etc posted by her....

 

Sorry new to ET too. If someone cud post a link where her posts are on ET forum, I would appreciate that...

 

Thank you

 

 

Pat

[steveh2009 10]

Pat,

 

Donna has 1000's of posts on elitetrader. She has a journal over there which details her green stage in learning. Just use their search facility.

 

A few forum thugs roughed her up recently and she decided not to post frequently anymore.

 

If I had to guess her winning pct profile, it would probably be:

 

30% winners, 10% losers and 60% scratched trades (b/e + 1 tick), give or take.

 

I'm guessing...but I think I'm a pretty good guesser after reading 100's of 1 person's posts.

[joshdance 10]

Josh you had mentioned delta ..is it available in ToS and more imp how does one uses it. Very nice thread and following it now.

 

Pat

 

There are many resources on delta -- I don't use TOS so I don't know if it has this capability or not. I recommend a search on TL for various resources. Volume delta is simply the number of contracts traded at the bid subtracted from the number of contracts traded at the offer--this measure can be taken per bar, or per price. How you choose to use it is up to you...

[pat15 10]

Thank you Steve for your reply ...sorry to hear that about her. I did try the Journal but cud not find any posts by her. May be what you mentioned isn the reason. I will try it again.

 

Thanks Josh for the explanation. I will check and see what I can find. Meanwhile I will plod ahead with this simple setup discussed here. As mentioned by some one need to track 3-4 futs to find a setup that meets this criteria.

 

Does anyone trade commodities? I dont but just curious if some uses this setup for wheat, beans etc.

 

 

Regards

 

 

Pat

[brownsfan019 17]

Take this with a grain of salt, but the outcome of their research is that, in sample, the optimal entry is with a stop. Out-of-sample, the limit order performed best.

 

There is a considerable amount of detail in this book, which lends credence to the author's research. As an explanation of why a stop entry performs better in their tests, they speculate that optimal entries appear to have both a counter-trend element (entering against the pullback) and a with-trend element (entering with a stop).

 

I've always entered on a stop - reasons being:

 

1) You guarantee to fill all your winners

2) You guarantee to not fill some losers (no fills)

 

In contrast, a limit order works as follows:

 

1) You guarantee to fill all your losers

2) You guarantee to fill some of your winners

 

And when you break it down like JMC does on a 1 minute chart, you aren't 'sacrificing' much for that extra confirmation. On a daily those no-fills can be huge. The catch being you are paying for that extra confirmation and you will get tick-in's that will frustrate you.

[steveh2009 10]

A buy-stop or sell-stop order to get an initial position is no bed of roses either. You can just as easily get triggered into the market and then have price retreat from your position, stopping you out. The momentum was good enough to get you in but then it retreats.

 

You don't have to pick one over the other though. You can create a buy bracket (long) or sell bracket (short) order where one order cancels the other. Basically, you're trapping price in a range with your long in a buy bracket getting triggered on either the buy limit or buy stop, whichever one happens first.

 

The first and second pullbacks are your best opportunities.

 

If you're trading a great futures contract like the CL, spend less of your time thinking about these things and more of your time thinking about how to add to your winners when you catch a strong trend.

Edited September 12, 2011 by steveh2009

[ake 10]

I've always entered on a stop - reasons being:

 

1) You guarantee to fill all your winners

2) You guarantee to not fill some losers (no fills)

 

In contrast, a limit order works as follows:

 

1) You guarantee to fill all your losers

2) You guarantee to fill some of your winners

 

And when you break it down like JMC does on a 1 minute chart, you aren't 'sacrificing' much for that extra confirmation. On a daily those no-fills can be huge. The catch being you are paying for that extra confirmation and you will get tick-in's that will frustrate you.

 

Very good points.

 

Katz and McCormick's research shows that a limit order appears to work best when entering with the trend, immediately after a moving average crossover. Again, this needs to be taken with a grain of salt (by discretionary traders) because their test was, naturally, automated.

 

Of the two entry strategies (with-trend at the MA crossover and countertrend when price bounces off an MA), tested with both limit and stop entries, they concluded that entering with a stop when price bounces off a MA was the best performing.

 

This appears to confirm the outcome of my rather more modest (and discretionary) efforts to test this entry.

 

The book admits that the way the tests were carried out was a bit rudimentary and a more sophisticated entry (candlesticks?) could improve results further.