Building Your First Strategy

Building Your First Strategy

To start creating your trading strategies, just navigate to the Strategy Builder tab on the website, and the canvas with the Strategy Details node will pop up. You can use any amount of space on the canvas to easily bring all of your ideas to life.

Firstly, you'll need to give your strategy a unique name (you don't want your awesome strategy to be overwritten with another one you prototype quickly, right?), and select exchange (ie Binance-Futures), trading symbol (ie BTC/USDT), strategy direction (Long or Short), and time period you want to run the strategy on (2023-01-01 - 2024-01-01).

Now we can get to the fun part and start prototyping our strategy! We can start with something simple, such as an EMA crossover strategy. For this strategy, we need 2 exponential moving average (EMA) indicators with different averaging values (slow and fast) and then we will be able to get a signal when fast moving average crosses slow moving average upwards.

Define Entry Signal

Exponential Moving Averages we can find in the TA Data Point, and since we need 2 of those, we just need to drag and drop 2 TA Data Points on the left side of the Strategy Details. In TA nodes we now need to select "Exponential Moving Average" in the top dropdown section. Then we need to select the period for the Fast EMA (let's try 10) and for the Slow EMA (let's try 20). The TA nodes have an input field indicated by the white dot with "Calculate On" name. This allows you to select on which data you want your TA indicator to be calculated. For EMA crossover candle close price is the most common choice, so let's stick with it. Now we will need 2 Data Point Nodes which will allow us to select candle "close" price.

Now that we have our Moving Averages, all we need to do is get a signal out of them, and we can do this with the Create Signal node. Drag and drop Create Signal node on the right side of the TA Data Points, so we can pass TA indicators into the Create Signal like so:

In the Create Signal node, we can give a name to our signal, then we need to select "Cross Over" as a comparison operator which will look for cases when Datapoint A crosses Datapoint B from the below upwards. Next, we need to choose the time frames we want to use for both TA nodes (ie 15 minutes), and attach the Create Signal node to the "OR | Entry Logic" of Strategy Details.

Define Exits

Now that our entries are sorted, we can create Take Profits. We can start with 3 Take Profits (2 standard take profits and one Trailing Take Profit), and see how frequently they get hit. For example, we can do take profits at 1%, 2% and 3% for the trailing take profit with 0.5% trailing scaling out of trade by 40%, 30% and 30%.

Now we need to define our stop loss to make sure we don't stay in a losing trade for too long, which we can do in the Strategy Details Node - let's set it to 1% for now.

Additional Settings

Now we just need to finalise our settings in the Strategy Details node: * set our Starting Balance to $1,000 * set Leverage to 5 (this would allow us to use 5x of our balance to enter into the trades, so $5,000) * set Periods Between Trades to 20 - this means that after we entered into a trade on our signal, with 15m Time the next entry can be done only after 20 periods of the minimum Time Frame we use pass. In our case we use 2 TA indicators Frames, so if we entered into a trade at 13:00, the next time we can enter into a trade would be 13:00 + 15m * 20 = 18:00 * use 99% of our % Portfolio For Entry because we want to have some spare funds to cover commissions and funding fees * % Max Usage of Portfolio could be set to 100%. This means that the system would allow us to use our balance * leverage which is $5,000 when we start the backtest * No Activity Stop Loss Periods we can leave at 100. This setting uses the same logic regarding how many periods on the minimum Time Frame should pass as Periods between Trades. The settings means that the position will be closed after 100 periods pass.

Run Backtest

That's it! Now we just need to click on the Save Strategy button, after that you can click on Load Strategy and you will see your saved strategy in the popped up screen:

The fields are not populated because we didn't run a backtest yet, and once we do, the fields will be populated with the result of the latest backtest we ran for this strategy name. Now click on the Run Backtest button to start the backtest execution.

In the top right part of the Strategy Builder where your notifications pop up, once your backtest finished, you should see simplified backtest results:

While, we can see that the strategy achieved 144% PnL over the backtested period, we can also see that our Excess Returns are - 11,731.79% meaning we would've made significantly more if we just bought SOL and held. This tells us that this strategy is very bad but can YOU come up with a better strategy? Or even improve the one we build together here?

Important

When creating strategies, always be mindful of overfitting (adapting your parameters to historical price movements), long losing streaks (can your strategy and/or you survive a long losing streak when you go live?), outliers (these 1-2 trades that made most of returns in your strategy might not happen when you go live).