A pullback strategy that works

Hi all,

I recently got interested to try out a daily pullback strategy using a variety of public sims and ranking systems available here on p123. While some of them looked promising in the beginning (usually ‘previous close’ as transaction price…). I soon found out that none of them actually worked over a longer time frame.

I tried the following setup:

GENERAL
Starting Capital $10,000
Commission $10 (Flat Fee)
Slippage Variable
Transaction Type Long
Rebalance Frequency Daily
Price for Transactions Next Average of Hi and Low
Allow sold holdings to be re-bought at current rebalance Yes

POSITION SIZING
Nr of positions 2

UNIVERSE AND RANKING
Universe All Fundamentals - USA
3 different ranking systems that I tried:
https://www.portfolio123.com/app/ranking-system/54062?st=1
https://www.portfolio123.com/app/ranking-system/54315?st=1
https://www.portfolio123.com/app/ranking-system/44149?st=1

BUY RULES
MktCap > 250
Rank > 90
close(1) < close(2) AND close(2) < close(3)
close(1) < open(1)
low(1) < low(2) AND low(2) < low(3)

SELL RULES
I tried the following three as single sell rule:
noBars > 1
noBars > 2
noBars > 3

Futhermore:
StopLoss None
Hedge/Market timing Disabled

I was not able to generate a profit for any one year period or above. Does anybody have an advice of what I have done wrong?

Best regards,
Florian

Florian,

The above is a backtested strategy from Larry Connors’ book: “Short Term Trading Strategies That Work.”

It is been a while since I traded pullbacks and I’m not claiming I ever got it to work for me but then I didn’t have P123 at the time either. You may need a little stronger pullback. Maybe the stock being in an uptrend might help too: Connors always uses this.

I think I was trying the 5 day stochastics below 20 followed by a strong 2 day decline in an up trending stock as the entry.

Also, you might look at pullbacks in up-trending ETFs. Leveraged if you want. I think I started looking at technical strategies for ETFs after being inspired by Gil Blake’s success with mutual fund technical strategies (see The New Market Wizards). I forget all the reasons I fell into this: perhaps the multiple stocks in the ETF made the strategy more consistent but it certainly could help slippage. Logically, one would think that regression to the mean is more expected, statisically, for a large basket of stocks. Gil Blake makes a consistent argument based on the binomial theorem. As I recall a lot of my trades were in GDX which was pretty volatile: again I’m not saying I got rich doing it.

Don’t know if that helps.

sister -

  • Set your price for transactions to “next Open”. “Next Average of Hi and Low” is not point in time and won’t give you good results on a short time frame.
  • To start with, set your commissions (variable and fixed) to zero as they are impacting your investigation. Add commissions only after you have a healthy average profit.
  • You should be using close(0) i.e. close(0) < close(1)
  • As Jim said, you need a significant pullback, at least 10% over two or three days
  • Use RSI(45) > 55 to get only stocks rising on a longer term basis
  • choose a comprehensive fundamental ranking system, not with technical indicators

Steve

Aside from what Jim and Steve pointed out, I noticed your model rebalances daily.

One of the most unfortunate things about the modern high-automation era where the average investor has so much access to technology is a belief that faster is better. Sometimes it is and sometimes it’s not. If you’re going to use any sort of daily strategy, you should make a thoughtful assessment of whether your ideas, whether they be fundamental, technical or both, are the sort that are likely to be incorporated into future stock prices that quickly. Many powerful ideas take time to get reflected in stocks. I’m not sure how effective the kind of modelling we can do on p123 (lacking intra-day prices, depth in bid-ask data, etc.) is at creating the sort of rapid-pace strategies that would seem to be needed for day trading. And if someone does come up with something, I’d be amazed if they set it for public visibility.

Pullback strategies can be fine on p123; it’s the daily part that leaves me skeptical. Even a week is not a slam dunk.

Hello Florian,

Take a look at these:

https://www.portfolio123.com/port_summary.jsp?portid=1289498
https://www.portfolio123.com/port_summary.jsp?portid=1289509

The universe I use is the same as the Buy Rule: mktcap > 35 & close(0) > 1 & AvgDailyTot(20) > 100000

Problems:
The main point is Low(-1)<Close(-1) kills the profits and slippage is high.
The pull back works well for micro caps so liquidity is a problem. AS you increase liquidity say good bye to profits.
Can you handle a 60% drawdown?

If you can overcome these problems then you can run a pullback system. As Marc said you need to be a good day trader. There are some P123 members who run daily sims hopefully they can offer you some advice.

Regards,
Mark V.

Gentlemen,

many thanks for all your valuable input! Plenty of ideas to significantly improve my current sim…