Reader Questions Answered

Trading With Pivot Points, MAs and Oscillators

The other day a trader named Sam put the following question to me:

I day trade via PIVOTS POINTS, MA’s and an Oscillator or so ( John Person was a day trading influence for me , not John Carter). My question is do you feature any trades similar to this mentioned by me?

I do not trade or do any analysis based on pivot points. I don’t even know how pivot points are calculated. My issue with pivot points is that they are just calculated levels with potentially no inherent meaning otherwise. Granted, if market participants think they are important then to a certain extent they become so – a self-fulfillment sort of thing. Regardless, pivot points tend to be mostly used in day-trading time frames, which generally are not ones I operate in.

Moving average and Oscillators are also trading tools I do not employ. Over the years I have done loads of system testing with them. Some of the systems were interesting. Others were not. In the end I just came to stick with simple things – what price is telling me.

Do not take this to be me implying that other methods are useless. Different things work for different people. The key to successful long-term trading performance is to find what works best for you.

Reader Questions Answered

Can I optimize position sizing during development?

Here’s a second trader question on the subject of evaluating trading system performance test results.

Is there a way to optimize for position sizing during systems development, using software or otherwise? Or am I supposed to work it out on a trial-and-error basis?

Optimization in this context refers to the impact of position size on system expectancy and standard deviation.

Optimization is one of those things in trading that can be very, very tricky. There are definitely ways to do it, but before we even get to that point we need a set of constraints. Specifically, we need to know the absolute risk parameters. By that I mean we need to know things like how large a drawdown we are willing to accept, how much of a single trade loss we can mentally/emotionally tolerate, and things of that sort.

Now, if you’ve not done much real money trading it can be difficult to sort out those risk parameters. We can all come up with objective ideas like, 10% per trade and 50% for a drawdown, but we may find out when it’s real money at stake that taking losses that large give you sleepless nights and ulcers.

When you have your risk/loss parameters sorted out there are indeed system trading applications which can help you with optimization so that you don’t have to do it via trial and error by hand.

One of those applications which just about everyone has access to is Microsoft Excel. You can use the Goal Seek functionality. You’ll have to do some reading up on it for yourself because this is not the place for a lengthy discussion on Excel. If, though, you have the ability to set your system up in the spreadsheet, you likely won’t have too much trouble with Goal Seek. I’ve done a great deal of system testing with Excel because it’s extremely flexible.

As for other programs, they vary in capability. I know that MetaStock definitely has the functionality, and I’m sure others do as well. I’m not an expert on all the different packages, however, so if you want to know about any specific one I suggest you look to the manufacturer and/or check with other traders on a forum site like Trade2Win.

Reader Questions Answered

Trading system equity performance

A member of my mailing list asked me this question about trading system performance:

For the systems I have tested so far, the equity line shows a significant drawdown at the beginning, but the system later recovers from that initial loss, in a kind of J-curve. This could be due to the specifics of the system, but I was wondering if this is a normal pattern in most systems, in which case my question is what is the rationale for this behavior in the equity line.

The formation and pattern of trading system equity lines will vary based on two primary things. One is obviously the way the system is trading. The other is what the market in question is doing at the time.

I would definitely not call it “normal” to see a J type of pattern to the equity line. If that sort of thing is being seen repeatedly in one’s testing then it suggests to me that the systems in question are of a certain style, but that the initial part of the price data being used has a pattern which is not beneficial to that style. For example, the systems being tested are of a trend trading nature, but the price data in the early part of the set is actually mostly range-bound. That would create a series of whipsaw trades, and thus a drawdown at the start of the equity curve, with things then improving later on when the data became more trendy.

If you are experiencing something like what the questioner has described, try testing your system with a different start point in the data. That should change the pattern of the equity line.

Reader Questions Answered

How do you set position sizing in Metastock system testing?

Developing and testing trading systems is about more than just entry and exit points. It also has to include a risk management element. The other day I received a question along those lines from a reader of my introductory trading book wanting to learn more about using Metastock.

Hi John,

Chapter 8 makes the following statement:

“An entry rule, in all its glory looks like this:

Go Long/Short n units when x occurs,

where n is either a fixed number or a formula based on risk”

Well, I thought it would be bold to just ask a MetaStock question to you, so I did check out the product’s help, couldn’t find an answer. Then went on to trade2win, and found some comments, but since I still have some doubts, I kindly ask your help.

My problem is I can go to the system tester and enter a condition for entry long or short, but it seems I can’t find a way to enter a value or a formula for “n”, in other words, how do you take care of money management when building your test?

Thank you in advance for any help, it would be greatly appreciated.

My response is to provide a picture. The following is a screen grab of the System Editor dialog box from Version 10.

As the red highlight indicates, when you do a Buy or Sell order you can define the Entry Size. The default setting will come from your overall settings. If, however, you select one of the other options (units, %) the formula bar will become editable. That means you do a customized determination of position sizing.

Reader Questions Answered

Am I using the right indicators for my trading strategy?

The other day I received the following question from a trader named Paul.

I have inherited an illness (SCA1) and now work from home, as I can be at my computer for the majority of the day I tend to short-term trade.

I only trade one currency (Euro/USD), and look to pick up just 10 pips on each trade. I use two moving averages and the MACD indicator for my trading. My moving averages are set to 10 and 5 and I only enter a trade when the two moving averages have crossed over and are pointing in the direction of the trade and the MACD is confirming both the direction and strength of the trade. This system seems to be working really well and I am picking up between two and three trades per day.

My question is:

In your opinion are they the best indicators for this type of trading?

Thanks very much for your excellent service and continual support. It is experienced traders like you that give me the confidence and desire to become a successful trader!

This falls into the arena of the “What inidicator is best?” category. My answer to that has always been and will always be that there is no such best indicator. Whatever suits you best is the one you should use.

In this case, my comment on Paul’s selection of indicators is that he’s using a set of indicators which are basically all based on the same thing. MACD is a moving average based indicator, so Paul is basically using three MA based signals. As such, he’s not really providing himself with any extra information.

My second point is that generally speaking moving average based indicators are aimed at use in identifying trending markets, which means users of them should probably take a trend trading view. That means one need to be prepared for periods of frequent whipsaws when the markets aren’t trending. It also means that the vast majority of gains are made from relatively few trades, by letting a position run. If you close out trades at a specific target you lose any chance you might have for the large gains which are the ones which produce the returns.

If one wants to trade for quick, small gains it might be better to use an approach which is more range/retacement based than trend based.

Now, having said all this, testing is the important thing. No matter what indicator you wish to use you should thoroughly backtest it to guage performance and to intimately understand how it works.

Deep Posts Reader Questions Answered

Is This a Good Day Trading System?

Here’s another question I received from a trader, in this case about day trading the mini S&P future.

I am currently evaluating a system that is profitable 85% of the time locking in 2 ticks (nett 1.6 ticks) each trade (on the ES). The other 15% of the time it losses on average of 8ticks (nett). The system developer advises to stop trading once the system made its 4 ticks profit(gross) the day. From your experience, does this sound like a viable daytrading system.

A good starting point in assessing any system is to determine it’s expectancy. That means what you would expect to make, on average, each trade. In this case we have the required figures to do that.

( 85% x 1.6 ) – ( 15% x 8 ) = 0.16

What I did there was take the win % and multiply it by the expected winner. That comes out to 1.36. Then I took the loss % and multiplied that by the expected loser, which is 1.20. Subtract the loser from the winner and you get 0.16 as the expected trade performance.

This doesn’t strike me as being a good system. It would average an $8 profit per trade, before commissions. If you pay less than that per trade and the system presents a lot of signals each day, maybe it’s worth it, but just on the face of it I would think you could do much better.

The Basics

Trading Systems Defined

In my previous two posts I outlined my thoughts on some topics related to trading systems, especially in regards to performance assessment. It occurs to me, though, that it is worth defining what exactly a trading system is. Here’s something we can use to that end:

A trading system is a set of rules which define when and how we enter and exit trading positions, and the size of the trades involved.

At the most basic of levels, the rules would be framed in this manner:

  • Go Long if/when . . .
  • Exit Long if/when . . .
  • Go Short if/when . . .
  • Exit Short if/when . . .

The “if/when…” is the actual trigger. It is something like “the market closes above the 10-day moving average”. This is a pretty simplistic set of rules, though. We really should get more specific.

A proper trading system rule includes the full discussion of all elements that go in to a trade. That includes what happens if you get a buy signal when already long, or even when short. It also includes the size of the trade to be executed. A better long entry rule would look something like this:

If not long, go long 1 contract at the market if close is above the 10-day moving average.

I think that is something traders generally have a fairly easy time grasping, so I won’t spend any more time on it, unless you want me to do so. The thing I do want to very quickly bring up, tough, is the difference between Trading System and Trading Plan.

In my view a Trading System is, as described above, the set of rules which dictate the way you enter and exit your positions. The Trading Plan is a broader thing. It is that which defines your trading overall, including the market(s) and instrument(s) you trade. It incorporates your risk tolerance. It incorporates your trading time frame (day trade, swing trade, position trade, etc.). It incorporates your trading return expectation. All of these things come together in the Trading Plan.

What it comes down to is that your Trading System is part of your Trading Plan. You build it (or select it) based on the other elements of your Plan – risk, market, instrument, time frame, etc. You are unlikely to get the results you want from your System if you don’t go through the Plan process first.