Systematic Trading 101: Four Step Framework

0:00 over the past three years I've transitioned from struggling to find a single profitable trading strategy to having an overwhelming number of strategies that My Equity cannot handle if you're interested in knowing how I achieve this let's jump into my four-step framework from turning an idea into a profitable strategy this is systematic trading 101 and I'm unbiased Trading the first step is idea classification this is the first step in the framework that involves classifying the strategy ideas into different buckets each bucket

0:29 represents a specific category of trading strategies and it's important to understand the unique pros and cons associated with each category here are the four buckets so we have momentum trading this is strategies that capitalize on the strength and Persistence of price movements we have Trend based these are strategies that aim to identify and ride long-term trends in the market we have mean reversions these are strategies that take advantage of the tendency of prices to revert to the mean or the average

0:55 value and we have pattern based these are strategies that rely on the identification and exploitation of recurring patterns in price charts now we need to actually identify the core issues with each of these and this will help you more than really the pros and cons as one I've been trading all these kind of different types of strategies I've always noticed it's simply picking your battle of what one you think you can overcome the the best so let's jump into the four kind of key

1:22 core issues now identifying the core issues with each trading style is crucial for momentum trading the challenges to recognize whether the momentum is strong enough to continue or will it reverse soon next is Trend beasts and the core issues identifying when a trend is no longer happening and you are now in a chop without being super late on identifying and joining the trend with these new parameters to identify when there is chop we have mini reversion and

1:49 this is the exact opposite issue of trend we actually want to identify when there is a trend and when reversing to the mean will no longer likely happen as we're in a very strong Trend and this will normally be a huge um profit loss for most mean reversion strategies if you can't identify a way to Sif out Trend based price movement lastly with pattern base the main struggle here is to identify when and how that pattern variates over time and

2:17 identifying it quick enough as most of the time these patterns will slightly change over time and they'll cause your current version of that pattern strategy to no longer work or be less effective or potentially maybe be a losing strategy now and it's having the ability to either quickly identify what has changed and also not overfitting it to small changes and that's really the key issue with those now qualification this is step two missed statements like I'll take an

2:46 entry when it shows weakness over the pre-market high or not sufficient for actually successful back testing to to ensure the viability of a strategy is crucial to develop a logical framework behind the idea for example a Quantified version of the previous statement could be I would enter a trade if there is a three minute red candle above the pre-market high after it first touches it so here's a little tip as well I picked this up along the way and when you're

3:11 developing your trading strategies try to think how you would code it and if you're not there with experience with coding maybe just break it down into actionable steps that make logical sense from identifying your entry points to assessing the risk to determining your exit strategy every step should be grounded in data and have a clear understanding of why you are doing this that is really the key to being fully systematic and making the most of your

3:33 trading efforts if you want to be 100 systematic obviously this will allow you to have a variance of that so you could have 80 systematic 20 discretion but still you want the core or how I say it is the macro Edge you want the macro edge of your strategy to be extremely Quantified where you understand why each Single part is happening so if it was an open and close strategy that's very simple you're saying that you have these

3:56 parameters that you are looking for in a pre-market gap for example and then you're taking a position at open with a 20 stop that's a very easy and close that uh walkthrough but for other strategies that's going to become more complex but really this quantification will allow you to do the next step which is back testing the next step is subject to the strategy to rigorous back testing back testing can be divided into three key steps so one we have pull historical

4:22 data we want to gather all the relevant historical Market data including price and volume information or any other data points that would be relevant to your particular strategy two is running historical entries and exits so you want to implement the Quantified strategy rules on historical data to generate a series of simulated trades over that particular time period so let's say five years free we want to analyze the results so you want to evaluate the performance of

4:48 the strategy by examining various metrics such as profitability risk adjusted returns drawdowns and other relevant statistics while the concept of back testing can really seem very simple on the overview level it's advisable to dive a lot deeper into each of these singular topics in back testing as there's a lot to learn though you may even consider joining the how to back test bootcamp which covers these topics extensively with over around like 15 hours of content going into back testing

5:15 and also the next step which we'll talk about in just a second now if you're looking to pull data on reckon and checking out polygon polygon API FMP or alpaca if you know how to code and for running historical entries and exits it's usually best to make this fully custom in Python but there are a couple python libraries that can help you like SSN and backtesting.p1 that can help you build some of this framework without even needing to fully custom

5:40 code it but it is important to be cautious when running your python code especially if you're not too experienced even seasoned coders like myself and other amazing Quant Traders I've had the pleasure to talk to and work with have mistakes in the code it's very normal and these mistakes can sometimes lead to look ahead biases or other Niche errors that produce insane results but that are actually unrealistic or they're not true results over that time period to avoid

6:05 this just make sure to double check your code and include realistic parameters like commissions and slippage and also do a general analysis of your Equity curve most of the time you can tell just by looking at Equity curve the amount of drawdowns if there is potentially an error here or you should potentially be looking into your code just in case there is one next step is robustness testing now many Traders make the mistake of stopping at

6:27 the back testing stage assuming that if a strategy performs well historically it will continue to do so in the future however live trading can expose fools and unprofitable aspects that were not evident during a back test to address this I recommend conducting robustness testing through the following tasks one we have Monte Carlo simulations so these employ and generate multiple variations of the strategy by randomizing different factors so you have reshuffling and resampling and a couple others this

6:54 helps the assess the strategy's performance under different marking conditions and various inputs to allow you to see the robustness of it two is out of sample testing so this reserves a portion of the historical data for the validation or optimization purposes ensuring that the strategy's performance is consistent with unseen Market data which would be your other section of data this type this step helps confirm the strategy's ability to generalize beyond the specific historical data used for the back testing or optimization and

7:23 lastly we have parameter sensitivity so this analyzes the sensitivity of the strategy's performance to changes in its parameter or inputs by systematically varying these parameters you can identify optimal values that maximize the strategies of a business and also where it's crucial that if you it's better with an example so let's say your current parameters you're shorting everything that's above a 50 Gap well how will the result when it's 49 how is it when it's 48 how is it when it's 46

7:52 how is it when it's 36 and when does the performance start to significantly drop off not like five percent or something but you know almost turn negative although break even uh when is that kind of point so you can be aware of how sensitive your strategy is to those minute changes um gaap isn't always the best one normally this is commonly done with like EMA strategies and other ones like that but you can apply it to those kind of

8:15 variables as well Now by following this four-step framework you can streamline the strategy development process you can filter out ineffective strategies and focus on those that demonstrate true profitability and robustness remember systematic trading requires a disciplined methodical approach and continuous refinement is essential to keep adapting to market conditions and your strategies changes 97 of Traders don't know how to effectively complete all these parts to identify and create robustness robust systematic strategies if you want to transform your discretion into black and

8:46 white stats I encourage you to join the July 1st how to back test bootcamp and there'll be a link in the description for a free video where it tells you a bit more information about it and you can go through it and just have some free value even if you decide not to join now thank you for watching and if you enjoyed it please like and share it with anyone else who might be interested