Quick notes on stops, more later.

[Victor]

Hello, I've just finished my first phase of research on using stops and money management using the momentum systems as posted on this site.  In summary, here are some interim findings:

All stops are on a closing basis, so you'd have to monitor & execute near the close each day, or take the slippage into the following day's open.

Price Momemtum Weekly
Initial Stop @ 18%
If it closes above 10% move stop to breakeven
If it closes above 18% move stop to +3% (recovers costs and a bit)
Trailing stop 30% below highest close

Version 1 (one week delayed entry)
Initial Stop @ 15%
If it closes above 18%, move stop to breakeven.
Trailing stop 30% below highest close

Version 2
Initial stop @ 20%

Version 3
Initial Stop @ 15%

Version 4
Initial Stop @ 20%

Version 5
Initial Stop @ 14

Getting fancier with stops on versions 2-5 didn't accomplish anything.  Either not enough trades or the system itself precludes the use of other kinds of stops.

For versions 4 & 5 the stops don't enhance the overall performance, but they do cut trades short, putting the capital back in your account sooner.

There are other kinds of stops to consider, but I think this a decent kick at the can.

I'd had trouble getting into the forum system earlier.  So I sent my findings via email.  They include a lengthier description, which I hope to post later, if there's interest.

Ciao for now,

Victor

[Victor]

Ok, here's a subset of test data I have put together based on the stop techniques mentioned in the previous post. The nake system is the Price Momentum Weekly without any stops.  The number of trades is not the same as the website as I wasn't able to get data for 32 of the trades (stock no longer listed).  Had they been included they would have had a positive impact on the outcome as there was an upward bias in returns, despite there being an even split between winners/losers.

So here's the return info

....................

See Excel file posted below...

[Victor]

The numbers in the previous post refer to total days of exposure.  So each day that each trade is on adds to this figure.  The return / exposure is an interesting metric.  How much bang for the time spent watching the markets!

Having all the stop mechanisms in place reduces time exposed to the market by 38%. And less than a third of the time are you watching losing positions. (27% of the total exposure vs. 33% for the system, in days, 941 vs. 1857).

However, performance is only improved marginally overall, by 5%.  To get more bang, go with the initial stop, trailing stop and retracement stops.  With this setup you improve performance by 19%.  Total exposure is reduced a bit, by 10%, and return / exposure is top quartile material.

Other good combinations are Stop + Trailing + Skipping 'til breakeven, and adding retracement stops to this.

Well, I guess I'll wait for feedback before forging further ahead.

Ciao for now,

Victor

[Super Stock Picker]

Any tips on getting excel info posted?

To Victor,

To get Excel info to the forum, I advise you to attach your file.

To do so, click on "Additional Options" below the posting box and you can see this option.

[tuzo]

Thanks for doing this.  Do you happen to have the maximum drawdown?  I would even be willing to sacrifice some performance to lower the Max drawdown from the ~ 40% of the momentum portfolios.

[Victor]

Here's the excel file with the summary results.  Easier to read than the posts above.

As far as drawdown, do you mean as a whole portfolio?  'Cause the stops effectively limit your drawdown per position.  As a portfolio it's more difficult to figure out in large part because you have to make allocation assumptions.

I'm currently working on a money management technique to determine position sizes.  This should be a little easier given the stop info.

Cheers,

Victor

[Victor]

I removed my previous posts with the data as the Excel file posted contains it all in one place. 

Please note that stops limit your downside, but can't do much in the case of gaps.

Cheers,

Victor

[Victor]

I sorted the trades by end date.  The max drawdown is 25% or less for the top performing "stopped" porfolios.  Using this method the max drawdown on the system peaks at 40%, with two other distinct peaks at 39% and 32%.  The 25% or less figure is seen as a single peak in the data where there are a number of consecutive losers.

Note this does not capture the entire portfolio effect as a number of these trades are on at the same time.  A little more complex to model.  Next project, maybe.

Cheers,

Victor

[bryanmcn]

Hi Victor
I hope your stops kicked in recently. This last month we have seen the markets give back all the gains they have made since March!
Has anyone been actually using stops lately?

[Victor]

Hi Bryanmcn,

So far I've been researching the topic.  In the meantime, I've been putting a few beaver bucks on the line, and yes, it has been painful.  I have been using a money management strategy that has limited my losses, so the pain is bearable for now.  I'm definitely going to be applying my stops for future trades.  Hopefully by then I'll have the data to understand how best to apply the strategies across the different portfolios.  At the end of the day today I've got ALS below its stop for a couple portfolios.

So for now, I'm intending to apply stop strategies to future trades and will use the system as is for existing trades.

Cheers,

Victor