How does SMI manage risk?

How does SMI manage risk?

SMI, has both active and passive risk elements.

Active risk management elements

We understand active safety as all those elements that tend to protect SMI in the event of a negative event, such as our airbags in the car, which protect us once the accident has already occurred. They minimize the consequences of the accident but do not prevent it.

• Stop Loss "fixed"

Each and every one of the robots that operate SMI has a fixed stop loss, said stop loss, is placed at the same time that the buy-sell order is placed. That is, our buy and sell order, placed by our robots, have our stop order incorporated.

This offers us an additional guarantee of security, on the one hand, at the opening of the trade we know what the maximum possible loss will be, and on the other, in the unlikely event that the data center where our server is located has a problem of any kind if the price reaches the stop level, said stop will be executed on our broker's platform.

In the case of our systems, for example, it will operate with a trailing stop (without a fixed stop) and the data center or the MT4 platform has some kind of problem, we could have unexpected results, hence we have chosen to incorporate it in all systems and in all trades a fixed stop.

Fixed Stop Loss have been calculated and tested in the system generation process and subsequently validated in the back tests carried out for each system and pair, prior to their real activation.

• Stop loss "dynamic"

1) Closing by variable criteria (parameterized in programming)

Most of our robots has incorporated into their programming that, at the moment that the conditions to carry out the buy or sale order (that is, the circumstances that motivated the opening of the trade) are no longer fulfilled, the robot will close the trade.

Closing by "dynamic" stop loss will be executed regardless of whether the trade is in profit (and has not reached its take profit) as if it is in losses (and has not reached the fixed stop loss level).

Explanation behind it is that if we have opened a trade for some reasons, and these reasons cease to be fulfilled, regardless of the development of the trade up to that moment, it seems logical that we close without waiting any longer, we will therefore have multiple operations that will close without have reached our take profit or stop loss.

2) Trailing stop

In order to prevent trades that initially go positive and then turn around and generate losses, in those systems where the back-test allows it, our systems also have a trailing stop loss incorporated (it is a fixed stop that evolves with the trade).

Not all systems allow to have a trailing stop, since many need to be able to develop against until taking the correct direction, the decision or not depends on the back tests and the results of the analyzes that we carry out on each system in each pair.

Passive risk management elements

SMI, is the result of many hours of work and each of the decisions in its design have been prioritized based on safety and risk criteria. Many of these decisions have to do with the passive safety that SMI offers from its conception and design.

We understand as passive safety all those elements that tend to prevent the negative event from occurring, for example our rain sensor in the car that turns on the windshield wipers to prevent the accident from occurring due to lack of visibility. That is, all the elements that work to prevent the active management elements from having to be activated, or to avoid that, if they occur, an unforeseen event with a great impact on a specific currency or pair could put the entire SMI at risk.

• Number of robots

From our point of view, leaving SMI in the hands of a few automated systems would be taking an unnecessary risk that we have eliminated. If we were only interested in profitability, the ideal would be to operate with the best robot and in a single pair, in SMI on the contrary we think that operating with a large battery of robots, although it will reduce the profitability curve, it will also stabilize it, hence the manager's commitment is to try to have a minimum of 10 systems operating, and the generation of new systems is first of all intended to meet this objective.

Just as the law of large numbers reflects that the behavior of the average of a sequence of random variables as the number of trials increases tends to be fulfilled, we believe that, with adequate money management, we must work with a large number of systems that tend to meet the averages of the joint backtests of all systems. Thus avoiding that the level of risk that a system fails damages the joint profitability of all the systems. This point is developed in How many robots operate in XIC and why?

• Forex market itself

Forex market has been selected, because within the existing markets, Forex, although it may not be the most profitable market, offers us additional security due to its intrinsic characteristics.

Each pair in which we invest is made up of two currencies, so whenever the main currency rises, the secondary currency goes down and vice versa, so it always offers us opportunities and by having a balanced risk and being all the currencies correlated, it allows us to balance the currencies well. positions to balance risk.

It is not a market with a trend, unlike other markets that are trend, Forex has a mean reversion so we are not marked by any type of bullish or bearish bias.

The size of the market is gigantic and its liquidity is very very high.

Working with a high number of pairs, looking for a minimum deviation between the risk of each currency, forces us to make an effort to generate systems in pairs that are not "adequate" and to work them all at the same level, avoiding dependence on the mother currencies; US dollar and euro. You can see more in In which assets does SMI operate??

• Balanced portfolio

Unlike standard portfolios that work with risk and return variables, SMI portfolio has been created solely with risk criteria.

It is sought that level of exposure to each of the 8 currencies is the same and to determine it, not only the different lots (which come from the distances to the stop points) but the number of operations that are placed in each pair are calculated, so that the risk deviation in each currency is under control.

In the same way that we do not take into account the profitability of a system, we do not take into account the investment, since such investment as we see in the money-management part is fixed (in € of risk per operation within a closed level) and always looking for the optimal lot so that the Darwinex risk manager does not harm us. This means that for example we can be investing much less lot in each operation in one pair versus another, since the criterion is not the nominal investment but the distances to the stops in euros in each of the pairs and the frequency of the operations.

• Long and short

As the Forex is a market with reversion to the mean (pseudo seasonal), it allows us to design active systems that work both the long and the short parts, avoiding the risk that appears in markets with a high trend (such as indices) where systems are typically only long and if there were to be a medium-term correction at some point, we believe it would temporarily prevent our systems from working optimally during the correction.

• Volatility analysis

All our systems have volatility as the central axis, either in their design or in the variables themselves that determine the rules for entering or exiting operations. We can always have problems at specific moments of very high or very low volatility, but since this variable is so internalized in SMI, we believe it will be less affected than other strategies.

• Correlation analysis

Forex pairs are correlated, hence, for example, arbitrage opportunities that both market-makers and high-frequency funds exploit in the market.

At SMI, we cannot avoid such market correlation, but we do validate that systems operating in SMI are not correlated. The generation of systems is totally focused on working with robots that, in addition to not using input-output variables, do not have input-output levels that are not even similar. Our robots and their results confirm this in the back tests carried out during their generation, both in the out of sample part and in the in sample part, but also a final correlation test is carried out with the final data extracted from the platform itself. trading.

• Money-management

One of the differentiating points of SMI is that money-management of the underlying strategy has been specially designed so that the Darwinex risk manager does not interfere with SMI's results.

A differential aspect of SMI is that it calculates the risk positions of each operation based on a percentage term in € of the total equity of the underlying strategy account, that is, it invests a fixed % of its equity in each system. Money management is completely confidential.

• Trend, anti-trend, or mean reversion systems

We operate all types of systems, seeking diversification and de-correlation, we can find systems with reward-risk ratios higher than 2 or lower than 0.5, win factor from 75% to 30%, we seek that SMI is balanced in all senses.

• Time frame

We operate only in H4 and D1, this timing limits the number of operations per system, but drastically reduces noise compared to lower timing. We'd rather put more effort into the generation than risk the noise of lower timelines.