Following your system . . . . with a delay

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Following your system . . . . with a delay

Post by rhc » Mon Jan 23, 2012 8:43 pm

You have made it your goal to take every single signal that your Long term trend following system offers relative to your risk parameters.
Your system issues a sell short signal in a particular instrument.
Upon review of the risk-to-stop (relative to your account size) you find, unfortunately, that the risk is too great and you must pass on this trade. You really wanted to follow your system & take the trade but it’s simply too rich for you.
A couple of weeks (or so) go by and you notice that prices have rallied and are now above your initial entry point ( that you could not afford to take)
Because of this rally you can now afford to take this trade.

Question: Do you take the trade?

Note the key words in the above passage . . . “A couple of weeks (or so) go byâ€

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Mon Jan 23, 2012 9:56 pm

This can be tested, but I doubt you will be impressed with the performance for the reason you state:

"Every losing [and winning] trade will always retrace back to its stop eventually"

When the risk to stop is only a couple ticks, any size account can trade.

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Post by rhc » Mon Jan 23, 2012 10:03 pm

Yes, this is true which is why I was careful to include the proviso that you only get to enter within the next "couple of weeks (or so)" after the initial signal that you could not afford.

This of course only applies to long term trend following systems that have an average hold time of around about 6 months to 1 year
For a system with a hold time of a few weeks my above comment doesn't make sense

stopsareforwimps
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Sun Oct 10, 2010 1:47 am
Location: Melbourne Australia

Re: Following your system . . . . with a delay

Post by stopsareforwimps » Tue Jan 24, 2012 12:36 am

rhc wrote:Do you take the trade?
I would suggest it depends on what you tested. Your ability to take trades based on size/risk is part of your system. If you tested taking the trades later on and it worked out well, all good. Otherwise it is not part of your system.

I am currently coding up this very test. (See my earlier post in here viewtopic.php?t=6097&postdays=0&postorder=asc&start=20).

No idea at all how it will turn out.

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Tue Jan 24, 2012 2:43 am

surely the odds of this beinga winning trade have now changed - the market has moved against the initial position and hence is moving in an un-favourable direction - it may well be you get in and then in tanks in your direction - hey presto you risked less money than teh system and made out good - but this type of trade has the odds stacked against you

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Post by rhc » Tue Jan 24, 2012 5:57 am

Chris67 wrote:surely the odds of this beinga winning trade have now changed
That is an interesting comment.
However I don’t entirely see it this way.
If you had enough money in the first place you would have taken the original trade. No problems.
Now if you are in the original trade and it moves against you then how is this different to entering the trade a bit later on at the risk you can afford.
Your system is saying you should be Short (or Long) and you are Short (or Long).
Where is the conflict?

If we reason that the “odds of this trade being a winner have now changedâ€
Attachments
PlatCC.jpg
PlatCC.jpg (132.15 KiB) Viewed 5918 times

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Re: Following your system . . . . with a delay

Post by rhc » Tue Jan 24, 2012 6:01 am

stopsareforwimps wrote:I am currently coding up this very test. . . .
No idea at all how it will turn out.
Do make sure to post the results of this test on this forum.
Could be interesting.

Demon

Post by Demon » Tue Jan 24, 2012 8:05 am

Personally I wouldn't take it. I agree with Chris, the odds are worse than at the time of the original signal.

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Tue Jan 24, 2012 8:19 am

If you have optimized your entry point then getting in at a different price will likely degrade your system. If you get in at a more expensive price you missed some profit, if you get in at a cheaper price you are more likely to get stopped out.

sluggo
Roundtable Knight
Roundtable Knight
Posts: 2986
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo » Tue Jan 24, 2012 12:25 pm

A philosophically similar conundrum is: A fund manager receives new investment dollars, two weeks after a trade has been entered. What should she do?
  • (A) Immediately add to the position, using the same (#contracts_or_#shares per dollar of AUM) that she used two weeks ago at trade entry

    (B) Do nothing; do not invest the new money, do not add to the position; wait for a new trade entry

    (C) Wait for a "dip", an adverse price move, and only then add to the position

    (D) Wait for a "pop", a favorable price move, and only then add to the position

    (E) Something else

Eventhorizon
Roundtable Knight
Roundtable Knight
Posts: 229
Joined: Thu Jul 08, 2010 2:36 pm
Location: Boulder, CO
Contact:

Post by Eventhorizon » Tue Jan 24, 2012 1:22 pm

I would say it depends on what information you have about your system and how it trades in terms of probabilities.

Beware of the possible implications of what you find. For example, if the expected return of a trade entered with a delay is higher than that of a trade entered immediately, should you delay entry on all trades? Perhaps the expected return will be lower than entering immediately but greater than zero. Perhaps you can find some delay at which expected return drops to zero - this might give you a "safe" window in which to enter. Analysis is only limited by your imagination.

The problem will always be having a large enough sample size to draw meaningful conclusions.

Bottom line would be: test it and see.

7432
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Wed Dec 31, 1969 7:00 pm
Location: sandy, ut

Post by 7432 » Tue Jan 24, 2012 2:35 pm

thought of a fun experiment while skiing this morning.
consider all signals from your system as 'live' waiting to be entered.
build a risk manager that trades fixed equity of only 250k, or whatever number is low enough to reject many or all trades.
then have the system wait to enter all trades until the market has moved to a level of risk to allow the trade at 250k.
probably many signals will never be entered, but that would give you a bunch of information also.
if the system works it could be an addition to the how to trade futures with a small account thread.

stopsareforwimps
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Sun Oct 10, 2010 1:47 am
Location: Melbourne Australia

Post by stopsareforwimps » Tue Jan 24, 2012 2:59 pm

rhc wrote:If you had enough money in the first place you would have taken the original trade. No problems.
Now if you are in the original trade and it moves against you then how is this different to entering the trade a bit later on at the risk you can afford.
Your system is saying you should be Short (or Long) and you are Short (or Long).
Where is the conflict?
Your system said to go in and to get out in certain circumstances, and you can expect to win by doing that.

It might be profitable to get out if a trade goes against you initially. But you haven't tested that and that is not part of your system. (Actually this is something I want to test!)

There is a difference between staying in a trade and entering a new trade. Part of the difference is brokerage and slippage. And if you enter the trade with limit prices you may not even make the trade. Or you may not enter the trade if the market goes limit up or limit down.

What I am saying is that just because you would have stayed in a trade does not mean you should enter it now.

Eventhorizon
Roundtable Knight
Roundtable Knight
Posts: 229
Joined: Thu Jul 08, 2010 2:36 pm
Location: Boulder, CO
Contact:

Post by Eventhorizon » Tue Jan 24, 2012 3:33 pm

rhc wrote:Now if you are in the original trade and it moves against you then how is this different to entering the trade a bit later on at the risk you can afford.
It's a variation on the Monty Hall problem - the trade moving against you is like one of the doors being opened: you have posterior information.

P("winning trade") != P("winning trade" conditioned on "missed trade but price has moved against me since signal")

I would not be at all surprised to find that you should only enter later if the price has moved in the original trade's favor!

Anthony Marsland
Roundtable Fellow
Roundtable Fellow
Posts: 83
Joined: Thu Jan 13, 2011 5:33 am
Location: London
Contact:

Probability

Post by Anthony Marsland » Tue Jan 24, 2012 3:33 pm

rhc,
First rule of trading - follow the rules. If you do not have a rule in place for entering retracements of too expensive trades, then it is not part of your system at present. Maybe use this trade as inspiration to test/write the rule before the next opportunity to enter at a more favourable price occurs.

By choosing these retracement trades you may change the character of the sector heat, portfolio heat over the course of the backtest. Maybe the retracement trades maxed out portfolio heat and stopped other trades from being entered...

Chris,
My understanding is that there are no time rules in rhc's system therefore time of entry doesn't change the probability of the trade result.

LeviF
Roundtable Knight
Roundtable Knight
Posts: 1428
Joined: Mon Dec 22, 2003 12:24 pm
Location: Des Moines, IA
Contact:

Post by LeviF » Tue Jan 24, 2012 8:20 pm

sluggo wrote:A philosophically similar conundrum is: A fund manager receives new investment dollars, two weeks after a trade has been entered. What should she do?
  • (A) Immediately add to the position, using the same (#contracts_or_#shares per dollar of AUM) that she used two weeks ago at trade entry

    (B) Do nothing; do not invest the new money, do not add to the position; wait for a new trade entry

    (C) Wait for a "dip", an adverse price move, and only then add to the position

    (D) Wait for a "pop", a favorable price move, and only then add to the position

    (E) Something else
I disagree. In your (Sluggo) example I say put on all trades to match the system because that is the system. Sometimes the poster's system skips trades because they are too risky - taking the trades when they are less risky is not following the system (unless it has been designed as part of the system).

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Wed Jan 25, 2012 12:55 am

Interesting stuff

I have always found - and thats just my experience on my systems - that the best edge available in trading - is that the market is moving in your direction at the time of entry and hence the odds - in most systems- should be weighted towards sell MOO (Stop entry whatver) as opposed to waiting for a bounce
As a poster said above howver - analyse your own system and work out the odds - but difficult to see any system whose odds are better by taing trade against short / medium / long term price at that point

nonpareil
Roundtable Fellow
Roundtable Fellow
Posts: 62
Joined: Fri Nov 06, 2009 11:39 am

Post by nonpareil » Wed Jan 25, 2012 9:44 am

Upon review of the risk-to-stop (relative to your account size) you find, unfortunately, that the risk is too great and you must pass on this trade. You really wanted to follow your system & take the trade but it’s simply too rich for you.
It seems to me that this is no longer a mechanical system if you are making these decisions, it is partially discretionary because you should never have gotten a signal if it didn't meet your risk expectations that should have been programmed in.

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Post by rhc » Wed Jan 25, 2012 7:40 pm

To be clear, I do not trade like this since this is not what I have tested.
If I can’t afford the trade I either pass it up or I might look to a mini-contract or perhaps an ETF/ETN.

I thought it might be interesting to bring this topic up because intuitively it seems a good tactic.
I have noticed that in a large number of cases when a LTTF trade is entered, the market will move in an adverse direction some short time after entry.
This is to be expected because in LTTF you are generally buying the breakouts or selling the breakdowns and this is where the markets are the most stretched in terms of the short term timeframe. In other words, ripe for a short term pullback.
Have a quick look at your last 10 LTTF system entries and you might see this phenomenon at work.
It’s very rare that you will enter Long and the market simply takes off and doesn’t look back.
To be sure, this does happen and these can be the best trades to be in. But, more often than not there’s backing and filling and grinding.

The key here is not that you abandon your original system signals so you can play a retracement system instead it’s more what happens if you can’t afford the original signal.

This issue has some of the hallmarks of the “Monty Hallâ€

Post Reply