Hi, everyone. I finally found a good way to add an intraday trading to the advisor "Intraday Rush". So now it is mostly daily trading (new trades every new day). I am going to add to the new version another timeframe - 2-hours.

It is going to be the same logic, but you will be able to choose between daily and hourly trading. 2h-timeframe has 12 times more trades than daily. It also bring more flexibility of course, because the advisor can react to market fluctuations much faster.

The indicator "IRush" will also be updated! So stay tuned! The new update will be published no later than next week!

​A good pic is better than a thousand words. I have made a pic and I will add words to it to explain it.

Neural networks open possibilities. However, too many users these days do not understand the possibilities.

Neural networks learn prices, fluctuations, patterns... In fact, whatever you feed into a neural network (really - whatever!!), it will learn it. Can I feed random prices? Yes! Can I feed random prices of random symbols? Yes! Can I feed bullshit? Yes! It will learn it.

It will learn it. It is not a problem. The problem arises from learning too much.

Why this post? Because too many users buy bad strategies that show an incredible growth in the historical chart. Too many users do not buy good strategies because they think that a mediocre historical chart will not allow them to make a profit on Forex. Good strategies fall into oblivion, that is to the bottom of the ranking system, which is based on greed and not on analysis or credibility.

So what? Make a real live signal, run it for half a year and show a profit with your good strategy. Prove that it works. Yes, this is what I tried. And they still don't buy, because they found hundreds of strategies that show an incredible growth (from 1k to 1 trillion in several years, literally) in the nice historical chart.

Why this post? I do not want to be a liar. I do not want to sell nice historical charts.

This new strategy "Intraday Rush" shows several levels of accuracy in the training. These levels are defined by the property "Trading pattern". The "reliable" level has a much higher chance to win in the unknown period. While the "ultimate" level, despite the mind-blowing profitability in history, has a much worse chance of winning in the unknown period. This is it. I have written it. I have explained it.

Does the market never repeat itself? Is it true that, from this standpoint, an absolutely unprofitable strategy in history should have the best chance to win in the unknown period? Maybe yes!!!

This is what I think. History has some valuable patterns. A good strategy should try to learn the very basic pattern-like moves from history. Therefore, it should demonstrate profitability in any long period running for at least a year. A good strategy should have drawdowns and wrong trades leading to frustration in some periods. And it should win in other periods. This is what makes a good strategy.

Sometimes users do not want a profit on Forex. They want to buy a hope at this particular moment of their lives and be happy for a few days with that false hope. Yes, I understand it. Switch to the "ultimate" level,  this is your false hope. Some users can't stop believing that a nice historical pic will not lead to a profit in the unknown period. Yes, I understand this, too.

Why this post? After months or even years of playing around with so many "nice" strategies out there, come back and read this post again. And switch to the "reliable" trading pattern of "Intraday Rush".

Very mixed feelings about the current updates )))

First, there's an article stating that everyone chasing high profitability runs the risk of encountering an advisor optimized for a past period, which won't meet expectations when trading on real (new) data... And at the end of the article, a call to everyone willing to reset their deposits - to use an aggressive setting...

And in this context, another update comes out in the evening with the ultimate setting. Is this to make it easier for the adventurous seekers? ))))) I would like more information.

And did I understand correctly that during the training of RELIABLE, random data (to prevent overfitting) was around 40%, and for Aggressive - 20%?

If other sellers do not mention overfitting for their strategies, it does not mean that their nice historical chart is a great strategy.

It may be that for a month or 2 months a nice historical chart will reproduce itself in the unknown period with a satisfactory result. And after that, the abrupt end.

Overfitting is measured by a loss in the unknown period. In my tests, any loss over 0.18 is bad. Reliable has a loss of 0.15-0.17, which corresponds to the loss in the training and validation. Aggressive has a loss of 0.2-0.25, while Ultimate has a loss of 0.25-0.45. Both Aggressive and Ultimate show a loss significantly higher than the loss in the training. They still show some profit, so they are usable. I use randomness for inputs, I use dropouts and I use sorting for hundreds of passes in the final xlsx and so much more. I bet no other seller does half of that.

However, a higher loss means the price of one single error is high.

The nicest historical charts with other sellers out there have a loss of over 1 or 2 !!!

Why I did this. Because some users think that if I do not show a nice historical chart, it is because my strategy is not so good. It is good, I can make a nice historical chart like all those other sellers. And I am telling you that any nice historical chart is not going to win in the unknown period.

Reliable mode is preferred for both daily and h2 charts. It has the highest chance of winning in the unknown period.

I think most trades will be closed by the newer version. You cannot run the same advisor in different charts, it will be a mess

Hi, everyone. I am currently testing improvements for Intraday Rush:

1) Adding more features like day of week and day of year as inputs for training the network. All along, I was mistaken that only inputs from indicators or prices can be used for a recurrent network. But recently the gentle voice of the assistant from ChatGPT crushed my delusion, by saying: "Yes, it is possible". It should be added as a bar feature not as an additional bar.

2) Bitcoin or Oil or both.

So let me know if you have other suggestions for the next update coming out next week.

Hi, everyone. I added the following passage to the description of this advisor.

Let us focus on the "Trading Pattern" property of this advisor. The Ultimate Trading Pattern is a neural model which learned the history very well. This is, beyond any doubt, very good that it can draw a quickly growing historical chart. However, if price fluctuations differ from the history, the Ultimate Trading Pattern won't make it through. In technical language, this trading pattern has a high neural loss, that is, one error may be fatal.

On the other hand, the neural model of the Reliable Trading Pattern explored the basic and major price fluctuations in the history. Despite a hardly satisfactory historical chart, the Reliable Trading Patterns has the highest chance to make correct trading decisions in the future unknown period.

So how is it possible? What is the practical difference of the Ultimate Trading Pattern from the Reliable one? For example, the Ultimate mode opens a buy and keeps this trade despite a quick rollback of the market. A floating negative profit will be growing, and it will be impossible to cover it up with other trades after one month. However, the neural model of the Ultimate Trading Pattern will still insist that it is a buy. The Reliable Trading Pattern should be faster to recognize a reverse trend and open a reverse trade. So here it is the high "cost" of one error, which is in the nature of the Ultimate Trading Pattern, and not in the nature of the Reliable one.

The Aggressive Trading Pattern is somewhere in between the Ultimate and Reliable. However, I recommend using the Reliable Trading Pattern for live trading.

Hello!
After some short travels, I've also come to the conclusion that Reliable is the most dependable option, allowing us to move forward in line with the expectations set during testing. However, now I'm sorely missing at the expert advisor a magic-param...
I want to leave a series of unprofitable deals open with Ultimate, let them recover... But simultaneously activate Reliable...
Without magic, i'll have to close the losses and manually open a deal to implement this.

I published a new update. I added a magic for this, so you can change and apply new settings.

I changed to the new defaults. To apply the new defaults, click "Reset" when started the advisor. It is the settings I use in my account on myfxbook.com.

I tried to explain the best I could the difference among different settings of risk. I guess it was still counter-intuitive.

Below are the pictures that I now use everywhere. Because 2 pictures are better than 2 thousand words.

>>>>>  >>>  >>>>>

What's up, guys! I am starting a new search for the best configuration for Intraday Rush.

These are some of the changes I am currently planning:

1) Trading Pattern only Reliable and Ultimate. Ultimate to show that this advisor can draw a nice historical chart. And Reliable ready to trade live.

2) New symbols, I am currently considering and testing NZDUSD, USDNOK and USDNOK.

3) I will add the property "Stoploss", because many users think it is required. It is required, of course, but it is not absolutely required with this strategy. The neural model of this advisor makes forecasts quite often and can close any open trade long before it can reach its target value.

4) If you have time, you are invited to post your suggestions here or via the messages.

All profit will be ours!

Hi Evgenly,

Any chance of adding Magic number to the EA.

I have this request so that this EA can work separately from other EA on the same trading account.

When you'll add this option?

I can add it in the next update. But as I said it does not matter at all with this strategy.

Hello. I found such a moment in the advisor. The EA opened on a signal at 0 o'clock on a large spread. Is it possible to ask the author to enter an additional filter by time or spread? Thanks

Здравствуйте. Обнаружил в советнике такой момент. Советник открыл по сигналу в 0 часов на большом спреде. Можно ли попросить автора ввести дополнительный фильтр по времени или по спреду. Спасибо