Watch here the first part: The ultimate trading system

Transcription

Hi guys, I'm here to tell you about the system.

You described some possible solutions to why the system doesn't work in real time.

My plan was to disclose the solution in my next system trading magazine, but on this occasion, I want to disclose it here now.

This is the equity line one-minute chart and one of the objections was that we are actually looking into the future so that we don't know what low and high are for the next bar.

This is not the solution, of course, because in reality, we are not looking into the future at all.

Multicharts does not allow this kind of tricks, the meaning of the instruction is that once one bar is completed and we are starting the next bar, we take high and low of the completed bar and in those levels, we place the orders.

In fact, you see here, we are not selling at the top, but we're selling at the same level of the high of this bar or in this case buying at the low of this bar.

This is what we are doing!

So, sometimes we'll get the top, but in some other case not.

So we don't look into the future because the high and the low of the code are the high and the low of the last completed bar.

So this is not the case to understand why the system cannot work in real time, the problem is somewhere else and one other objection was: "Yes, but you are not including commissions or slippage".

So to give an answer, this is the one-minute bar, let's go to the one-minute system.

This is the 5-minute system and here we have again around $2,000,000 with this system without commission and slippage.

This is the 15-minute system, so let's take the 15-minute first and let's include commissions into the system and here we go commission, properties.

We include a $5 round turn.

Is that enough?

I would say it's more than enough because I'm paying around $2.05 I believe on Interactive Brokers on one side, so that will be a little bit more than $4 and $5 is just to cover other cases where you are paying something more.

That said, we are selling and buying at a limit.

We are not using stop or market orders, this means that we are not jumping levels, so with limit orders, we have not slippage in a certain sense, so the 5 dollars are covering decently what we intend to do.

Here it is the 15-minute result, which is obviously much worse, this is clear because we are leaving on the table about $750,000, you see here below, but we're still making a lot of money.

The equity line is still very nice, there are some points here which are not that clean, but it happened on several occasions and it always has recovered fine.

You can see here, year after year the system is still making decent money, it takes a while because you know there are many trades, so you see it's not the numbers we have seen before, but we're still collecting a lot of money.

So this is the 15-minute and if we look into the trade analysis, we can see that we have still an average of over 6 dollars and theoretically, even by paying more commission, we will stay inside this.

The situation changes a little bit on a five-minute chart because here the original average trade was a bit smaller, but we're still making money, you see, we are 5 something, so we're still good, the equity line is still beautiful.

We had about 2 million now we have 1 million, we are living 950.000 dollars on the table in the year starting 2015.

So, also, in this case, we're still making a lot of money, obviously, the first report was tricky considering the commissions I did not include, but including the commissions, it does not change the nature of the system.

Also if we have a look at the one-minute results, which were those in only 1 year, you see 592.000 instead of 2 million and here we have 1.300.000 left on the table, because of the commissions, but the equity line is no longer regular as it was before, but is still an equity line that everybody would obviously like to have in his system.

So the point is not the commissions, that is not the problem of the system.

The problem of the system is completely different and I said we don't have, because of limit orders slippage, but in reality, we don't have the classic slippage.

The slippage as we intend, where you want to enter at 10 and usually you enter at 11, this is the concept of slippage that we normally have and we think about.

In this case, there is a different kind of slippage and the different kind of slippage derives from the limit order where we expect to be executed.

With no slippage as said, but look at this.

So let's have a glance here inside.

As said this is the last completed bar, we are short, so we place a buy order at the low of this bar, it's not reached, then we place a limit order at the low of a new bar and in this case we are executed, this place a sell order at the higher of this bar executed and then we place a limit buy at the bottom of the new bar here and we are executed.

But look at this and also at this one, there are 2 cases here and also this one, where we are executed exactly at the top or the bottom of the bar.

It means we were so good to identify exactly where the market was stopping during that minute, this is a one-minute chart.

We are great, yes, but if you look at the miniS&P500, you see there are, at each level of the book of the order book, hundreds of contracts.

So I am, let's imagine here, 27.60 for example.

I have to place a buy order at 27.60 for the next minute.

I place it, limit buy 27.60, but obviously, for the next minute, for no reason in the world, I will be the first on the list of buyers.

There will be other contracts being placed before, sitting there, waiting to be executed.

I'm cueing, so maybe I have 200 contracts in front of my single contract.

Who is granting me that I'm executed?

If also, only one single contract of those 200 is executed, the chart is plotting that level as the low of the bar, but nobody is telling me that my contract will be executed.

So these executions at the limit of the bars at the bottom or at the top, are all executions which I cannot know if they really took place in the real-time trading.

I cannot know it, there is no way to know unless I test it.

But if you consider, especially on a very short timeframe, that in one minute you must be sure that all the contracts in front of you and your contract as well will be executed, obviously you have not many odds at your favor.

In any case, this is not the biggest problem on Multicharts and also in other software, of course.

We can go into back-testing and change the assumptions, saying that we consider the limit orders filled only if the level where we placed them is exceeded by at least 1 point or one tick.

By doing so we are sure that we would have been executed in that bar because there have been trades also below that level, in case of long entries or above that level in case of short entries.

With this condition, I want to show you how, in this case, the 1-minute chart changes.

If I were able to do it, this is dramatic believe me.

You see, this includes the commissions, as we said before, but also the condition that there are only fills in case the level is exceeded.

Remember that, if you did not put that, of course, you are sure of the executions only on losing trades, because in that case your level is traded through and you certainly in the market.

This is the equity line.

This is self-explanatory, this thing is dramatic!

On a monthly basis, this is the last year, because this was a 1-minute chart, you see we are losing everything, I would say.

Now you would say: "Okay Andrea, you said that this is on one minute, what happens on a 5-minute bar?", because in that case, the solution might be different because there is more time and so on.

Believe me, a 5-minutes is still small time, but in any case also if we include this commission on a 5-minute chart, we have exactly the same result.

This is the equity line and the total trade analysis changes to -16 from about 5 something that was before, year after year we lose all our money.

It takes a while because we have still many trades, even though less than before, but you see.

And last but not least, we can go on a 15-minute chart and see what happens in this case.

Again, nothing different from what we know, this is my equity line.

This is not the real performance of the system, because also this assumption is extreme, because also this assumption is taking into consideration only those orders where the level had been certainly exceeded, so we can imagine that we could stay in the middle, 50% chance to be executed and 50% not, which is an assumption.

I have made it here, I don't know how true it might be, but that is something that we could imagine.

Also in that case, if you make some math you see that you are still not on the right side of the corner to make some money.

So this is the sad real life and I can tell you, you will probably go back and play with some tricks.

One solution will be to insert "market if touched" orders, which means as soon as the level is touched, you buy.

In this case, looking it on a 1-minute chart with commissions, obviously what happens?

In this case, you are losing one tick entering the market and you are losing one tick exiting the market because when the level is touched you send the market order and you are buying one tick worse.

In this case, you are sure to enter the market, but losing one tick in the entrance and losing one tick exiting, you are losing 12.5 dollars entering and 12.5 exiting, which is $25 round turn.

This is the performance for the one minute, you have an average trade of 2.26.

Now if you lose 25 out of 2.26, you can easily calculate what the result of your system will be.

You should also remember that we had 6 something on a 15-minute bar, so we are still far from covering the 25 we could put on the table using this method, so this is obviously not something that works.

I imagine you will go back testing what happens on a 30-minute bar, I've tested all of these, believe me, there is no way to make this thing work, no way, believe me, no way!

Many people are using this trick to sell you something which does not work.

This is an extreme case, but there are also other systems using limit orders on miniS&P or on other very liquid markets such as Bonds, Tnotes, etc and the limit orders execution is always a big question mark.

If the average trade of the system could not cover a "Market if touched" entry, that system is scratch, you can throw it into the waste paper, it does not work.

Believe me, there are more gurus out there trying to sell you something that does not really work when you put it live on your charts.

That's it, I hope you got it and I saw a lot of interest in this point so I will probably put some more questions in the future about other systems, tricks or general technical issues that might happen when you trade live.

Good luck guys, subscribe, comment and see you next time.

NEXT UP ON THE BLOG

World Cup Award Ceremony

Interviews

Read More...

---

Robbins-PFGBEST World Cup Trading Championship Awards Presentation at CME Group

Interviews

Read More...

---

On forex markets - Dukascopy

Interviews

Read More...

---