Simple S/R based intraday ES strategy -- comments?

[austinp]

<i>"A 1.50 stop on the ES using a 15 min or longer term chart is imo pretty tight.

--------------------------------------------------------------------------------

Understatement of the century. I use ATR.</i>

**

A -1.5pt initial stop in ES is actually 1.25pt factoring bid/ask spread wouldn't work even if trading a 100 ~ 200 tick chart with focused precision. That type of distance is well inside normal ES random noise.

Using technical stops via ATR, fib levels, prior support - resistance levels, etc have zero edge over simple dollar stops long-term. Aynone who has spent time writing mechanical systems has seen this tested out statistically.

ATR is always a lagging measure. Stops place outside of prior swing lows, trendlines, etc are routinely gunned down in deliberate stop hunts in ES. That's why wycoff spring = 2B patterns are so effective as failure reversals... they are fundamentally gunned stops ahead of planned reversal or pullback continuation movement.

Good news is the ES offers a lot more profit potential than 2pt scalps. Intraday swings of 8pts, 10pts, 12pts, 20pts are firing off multiple times daily. Widen your stops, and widen the profit objectives. Then you'll be in harmony with present ES market conditions.

 

[austinp]

<i>"I'll backtrack 2 weeks and see how the system performs with 2-3 different stop/target levels. Phew. I've got my work cut out for the long weekend "</i>

That won't show you anything but holiday-period behavior. What the ES has done in December versus what lies ahead from now thru April are probably two very different things.

Backtrack at least three months, preferably six months and you'll be off to a good start. Testing first, demo trading to walk results forward, real money only after all that.

Or, learn while you burn... no one ever earns while they learn inside R&D mode with this profession of ours.

 

[rolextrader]

Quote from austinp:

<i>"A 1.50 stop on the ES using a 15 min or longer term chart is imo pretty tight.

--------------------------------------------------------------------------------

Understatement of the century. I use ATR.</i>

**

A -1.5pt initial stop in ES is actually 1.25pt factoring bid/ask spread wouldn't work even if trading a 100 ~ 200 tick chart with focused precision. That type of distance is well inside normal ES random noise.

Using technical stops via ATR, fib levels, prior support - resistance levels, etc have zero edge over simple dollar stops long-term. Aynone who has spent time writing mechanical systems has seen this tested out statistically.

ATR is always a lagging measure. Stops place outside of prior swing lows, trendlines, etc are routinely gunned down in deliberate stop hunts in ES. That's why wycoff spring = 2B patterns are so effective as failure reversals... they are fundamentally gunned stops ahead of planned reversal or pullback continuation movement.

Good news is the ES offers a lot more profit potential than 2pt scalps. Intraday swings of 8pts, 10pts, 12pts, 20pts are firing off multiple times daily. Widen your stops, and widen the profit objectives. Then you'll be in harmony with present ES market conditions.

If your palms don't drench your mouse with sweat, use an initial 5 point stop, but I'd still use ATR settings of 3,3,2. to trail the price action. And even using a 699 tick chart (my preference) you will see many, many decent winners (5+ points) get stopped into small loses, scratch or slight profit as ES backs and fills. The typical trader doesn't have the mental conviction to see good profit evaporate. At least I don't.

Go for a high batting average of winners, not the daily grand slam. One doesn't need to catch huge moves to make good money in ES.

 

[trackstar]

Good luck with that stop. I dont care how spot on you are. 50% of the time you will probably be stopped out of good trades you would have had otherwise if the stop was reasonable.

 

[DeepThought]

quote from austinp:

I'll backtrack 2 weeks and see how the system performs with 2-3 different stop/target levels. Phew. I've got my work cut out for the long weekend "

That won't show you anything but holiday-period behavior. What the ES has done in December versus what lies ahead from now thru April are probably two very different things.

Backtrack at least three months, preferably six months and you'll be off to a good start. Testing first, demo trading to walk results forward, real money only after all that.

I've spent few days backtesting the strategy, and I came pretty much to the same conclusion. The daily volumes drop off dramatically in mid-December. Also going back few months will get me interesting volatility environment that might still continue, but not forever.

So I need more data. So far I've been using NinjaTrader with the data IB provides. That's only few weeks. Also manually testing the system, especially with multiple parameter values, is far too time consuming.

So what I'll need to do next is to buy or steal few years of ES history data and import it to NT. Then I'll write a little plugin:

- Allows me to enter trade parameter (stop, target levels etc.)
- Allows me to point trade locations on the screen
- Calculates P/L for all different parameters
- Spits out the result into Excel

Luckily I was a programmer before management track swallowed me up. If I get this done ok, it might be useful for others too.

Or maybe there is something like this available on the market already? The problem is that the entry is somewhat discretionary, so the system can not be completely automated. I'll need to point the entry spot and then have the system take over.

== DT ==

 

[DeepThought]

Quote from austinp:
Good news is the ES offers a lot more profit potential than 2pt scalps. Intraday swings of 8pts, 10pts, 12pts, 20pts are firing off multiple times daily. Widen your stops, and widen the profit objectives. Then you'll be in harmony with present ES market conditions.

It's really a choice of preference ... do I want to try to make small profits from the short-term movements, or catch the longer-term moves? For now I'm looking at the 5-30 minute timescale, that's why I have the 2-4 point profit targets.

So far the test results have not been too encouraring. Slavishly fading (almost) every move against a pre-determined S/R level produces or success/fail ratio of 1 to 1.5. Luckily the profit/loss ratio has been over 1.5 to 1, so the system has been breaking even

Simply widening the stop without increasing the profit targets decreased the performance. More winners, but the losses are much much worse. Oops.

I'll still keep testing and optimizing the short-term trade strategy. One thing I've found in my years of swing trading is that for learning it's better to stick with one plan and try it out -- if it doesn't work in the end, then change. But at least I've learned more than if I changed approaches every week. I've done that too, many times. It doesn't work very well

But your advice is not forgotten. It might be that the end result of my testing and simulation and actual trading is that I'm not able to make the scalping work.

==DT==

 

[DeepThought]

Quote from richardm:
Interesting support / resistance concept - - what criteria are you using to determine levels (N # of ticks, candles, etc.)? What I am curious about is how this might work with the market profile. I have database and programming experience and would like to apply this to futures markets. Thanks in advance...

I identify the S/R levels with the old Mk I Eyeball instrument I look at candlestick charts of the previous 1-2 weeks, using 1 day, 60 minute and 15 minute candles. S/R levels are the clearly visible congestion areas over longer time, or tight areas where the price had rebounded in the past 1-2 days.

Additionally during the trading day I make a note of short-term S/R zones where the price has rebounded during the day.

Very unscientific, I know. In my opinion price action is the visible product of the intentions and actions of the horders of market participants. It's a chaotic system, driven by mass psychology and certain rules and conventions (either machines or institutions). Human brain is an amazing machine for finding patterns and I believe that S/R zones can be seen more reliably than found by any machine.

I don't know, maybe I'm wrong and there is a realiable system. I'd be happy to be proven wrong. I'm a newcomer to daytrading and futures and I don't try to pretend to know anything. I pick a hyphothesis and try it out. If it doesn't work, I'll try something else. So far I've seen this forum to be a great source for opinions

==DT==

 

[athlonmank8]

Part of the problem is fading. When fading you need to open profit targets more.....hence the reason why you're fading

 

[Hombre]

Quote from athlonmank8:

Part of the problem is fading. When fading you need to open profit targets more.....hence the reason why you're fading

True. Even bigger problem is when you try to fade against the current trend. This is the missing component of your system.
Fix that and you are ready to go.

 

[DeepThought]

Quote from Hombre:
Even bigger problem is when you try to fade against the current trend. This is the missing component of your system.
Fix that and you are ready to go.

Makes sense. What I was thinking is that in an uptrend I can go short at a resistance level for a brief pullback, but that would be a short-term move and the profit target smaller by necessity. Going in the direction of the trend would yield bigger profits.

My question is how to determine the trend? MA crossovers, oscillators etc ... Let's say for the sake of simplicity that the trend is based on MA crossover on the next higher timeframe. So if I trade with 1 minute bars, I could look for the MA order in 5 and 15 minute bars and trade accordingly.

Better ideas?

In any case, I've got my NinjaTrader backtest function almost working, so I can easily try different variations.

==DT==