My Highest Return

[spamula]

My current strategy has achieved 54-57% profitable trades over the last 30 days for about 47 trades on an intraday basis. I have included 15$ commisions and 0.05 slippage. The stock I'm trading is the SMH 1000 shares its giving me a profit of $1600+

Do people think this performance is good enough to run for real with IB? Also is the amount of slippage high enough?

What is the minimum % profitable that should be achieved before running for real in back testing?

Could this percentage be improved with profit targets ?
Also how do I take a picture .jpeg of my back tested results?

[FibbGann]

I would say 75%, but thats just me. I have a bunch that can do 55 to 60%.

I would say keep plugging away at it.

Fibbgann

[spamula]

To be honest I'm getting stuck now I have been for a while
I'm not sure if I can improve upon the stop ratio code that works ok with IB.

I tried an EMA strategy that works great over long periods i.e.
months + years but poor intraday.

I spent a fair bit on getting the stop ratio code to work with IB it works good on an intraday basis trading for a few hrs a day. It has stop loss, trade entry time stamp etc.

Any ideas how I can get the stop ratio code - winning trade percentage higher than 54-57%. What works best profit targets or letting the winners run?? Any idea on a filter that may help this code?

[DuaneG]

Hi Spamula,

I can help with part of your question. In regards to posting an image of the Back Tested Results you'll want to take an image of the window by using the eSignal Snapshot Utility.

Once the image has been captured, you'll want to either attach the file directly to the post or upload into FileShare and then include the link in the body of the post by using the Insert Image tool at the top of the posting body.







Once the image is inserted into the post, everyone will be able to see the image that you have included. Hope this helps.

[spamula]

How can I improve this stop ratio strategy, ideas please - check my other post below !

[Doji3333]

Here is what I would suggest..

Analyze your stop system..

1. Find out how many trades moved against you right away and took losses. Then find out how many of these turned back into profits (had you held them).
2. Find out how many trades "drifted" before getting stopped out and what the average # of bars of the drift was (you might fine 2~5 or more). Now, find out the max/min and average price retracement that caused your stop to be hit.
3. find out how many trades were actually correct, but eventually got stopped out for a loss (or minor profit). These are trades that you will want to attempt to adopt a containment system for losses and a system to protect profits. You'll want to find out the average and minimum profit factor that these trades achieved and possibly the number of bars (max/min/avg) that it took to achieve the minimum profit factor.

From this information, you will be able to determine some specific characteristics of your stop system. You should then be able to formulate some improvements to your stop system - like maybe..

a wide stop for the first 2 or 3 bars after a trade, then tighten it up to your normal level.

a "breakeven+" stop adjustment within x # of bars if your minimum profit factor is reached.

a stop adjustment if a multiple of your minimum profit factor is reached thru a profitable trade. For example, if your minimum profit factor is 0.10, then multiple would be 0.20, 0.30, 0.40...

You also might consider adding a "sell for profits" position to you system that sells 60~70% of your holdings for profits at certain points in time. This helps your system maintain profitablilty and contain losses.

There are many things you can do to improve your system. Although, if you can maintain a 55+% accuracy and improve the money management - you should be fine. You just want to be cautious of the extended drawdowns that may occur. In otherwords - find the faults of your system and try to improve them.

[spike500]

Spamula,

the number of profitable trades is irrelevant.
What is important is the amount of money you win or loose.
If you have 25% of profitable trades that generate each time big profits and you have 75% of losing trades , but you loose small amounts, than you have a good system.
When you try to have a big number of winning trades you will automatically protect your ratio winners/losers by getting out to soon when you win (to be shure that your trackrecord of winning trades will be excellent and to prevent that your winning trade would become a losing one). So you will miss a part of the profit.
And if you have a losing one you will try to turn it around in a profitable one, with the risc of loosing more than initially.
Watch you profits, your losses and your drawdown. Don't mind about the percentage of winning trades.

[gspiker]

spike500 is correct. Percent of trades that are winners in and of itself doesn't make or break a system. The ratio of winners to loosers and the ratio of how big the wins are to the losses is more important. However there is a bit more to it than just raw profit and loss after back testing as well.

For example..how big was the drawdown? This will tell you the largest potential loss, and how much you might have to "sweat through" the trade going against you. What good is a system to you that generated good profit and loss on a backtest, if it also showed that you might have to start swilling the pepto during half your trades?

Also, when it comes to a low winner/looser ratio there are two important things to consider:

1) The number of potential bad trades IN A ROW goes up pretty fast...what would happen if the first days of trading you hit a bad streak and lost 10 in a row? My guess is it would hurt your account and your trust in the system...

2) Related to #1...it takes a certain kind of trader to be able to trade systems with low winners/looser ratings. I can't do it, I have to know going in that I have a better than 50% chance of winning. That's me, others will not have this issue. Perhaps a bigger issue however is how many people will continue to use a system that generates 10 bad trades in a row...and that may be just when the system is about to deliver the big winner that will bring you back to a profit position.


Also :

big number of winning trades you will automatically protect your ratio winners/losers by getting out to soon

Or you have found a way to filter out low probability trades and are only taking a very small number of high probability trades. This works also.

Garth

[FibbGann]

I found this to dramatically help my returns.

EFS is set to trade only during these time periods.

9:45 AM EST to 11 AM EST.

AND

2:20 PM EST. to 3:45 EST

Fibbgann

[spamula]

Hi Fibb,

I found 2.05 to 3.45 EST to be the best and only time to trade,
the odds are stacked far more in your favour.

I used to like the morning from 9.50 to 12.10 also, but it doesen't back test very well.

In my back test for the morning I get 41% winners with a loss of
$600. I f I trade the afternoon only I get profits of $2000 on 1000 shares and a 55% winning trade average on a 30 day test.

Thanks to everyones posts I have soaked them all up.

Regards
Spam

[spike500]

Originally posted by spamula
Hi Fibb,

I found 2.05 to 3.45 EST to be the best and only time to trade,
the odds are stacked far more in your favour.

I used to like the morning from 9.50 to 12.10 also, but it doesen't back test very well.

In my back test for the morning I get 41% winners with a loss of
$600. I f I trade the afternoon only I get profits of $2000 on 1000 shares and a 55% winning trade average on a 30 day test.

Thanks to everyones posts I have soaked them all up.

Regards
Spam

Spamula,
This is the way i analyzed the results of my trading system:

Make an excel sheet with the following information:
1. Entry price
2. Exit price
3. Maximum open profit
4. Maximum open loss

With this information you can calculate for each trade:
1. your maximum potential profit
2. your maximum potential loss
3. your drawdown ( which is very important to mesure your risc)
4. your optimal stoploss.

I have an example sheet with random prices to show how I did it.

BUT IT IS IMPOSSIBLE TO POST IT APPARENTLY UNLESS ALEX CAN HELP US.

By varying the stoploss you can see the corresponding returns and choose for the best value for your stop.
Don’t look too closely to the sheet because it was made in a hurry. The purpose was to give you an idea of a kind of analysis you can do on your system.

With a starting capital of 5000 $ and a stop loss on 1 point in the S&P minis and trading only 1 mini it gives the following results:
1. Result without a stop: -2065.78 $
2. Result with a stop: 607.15 $
3. Return: 30357.50 $ or 607.2%
4. Draw down: 5 % (this indicates that your maximum risc will be 5% no matter when you started trading during the given period)

I did not include commissions or slippage. But they don’t interfere in the logic of the analysis.


Spike

[spamula]

If you take your highest possible profit (best run up) for each trade and then divide them will this be your optimal profit target?

Any ideas how you would work out the optimal stop loss?

[ACM]

Spike
Excel files cannot be attached. You could either zip it and attach the zip file or upload the file to a FileSharing group and then post the link here
Alex

[spike500]

Originally posted by spamula
If you take your highest possible profit (best run up) for each trade and then divide them will this be your optimal profit target?

Any ideas how you would work out the optimal stop loss?

This is the excel sheet. You can improve it at your convenience.

You can see what the highest profit was that you could achieve. The next step is to see if you are better off by trading the signals till the exit or if it is better to place take profits. The statistics can help you to decide where the take profit should be placed.

Use as much trades as possible (i did my calculations on approx. 1000 trades) because the results will be much more reliable.
Analyzing the trades will give you a better idea of the striong and the weak points of your system.

The most important thing is to keep the drawdown at a reasonable level. This is necessary to protect your capital.

[spamula]

I have noticed many traders prefer to trade the e-mini or futures to stocks.

Please explain why? I'm having difficulty relating your stats
to ones I can create for the SMH tracking stock.

MAX OPEN PROFIT - is this the max. run up or biggest profit the emini went to?

RESULT IF STOPPED IN LONG? -1, What does -1 REPRESENT? $50?

845.39 if long with a stop of -1 so you'll be stopped out at 844.39
with a loss of -$50 correct?

RESULT FROM TRADE -11.69 -- but would have only lost -1 if stop in place?

Just making sure I'm on the same lines as you.

Thanks for your time!
Spam