Trade bitcoin with 120% profit using AI and the Stock to Flow Model in a bearish market

Published in
4 min readFeb 19, 2022


We would like to share with you the results of one year of hard work by our cryptocurrency traders and programmers.

Looking at our backtest results, we are buying at the beginning of the uptrend and selling just before the price drops. you may wonder how that is possible!

It’s a combination of indicators and on-chain data and machine learning.

Machine learning signal

Based on the simple moving average of bitcoin price for a year, we used TensorFlow to predict the next candle close using our model. We can use that prediction to decide whether to buy bitcoin or to wait.

Bitcoin stock to flow model

This model treats Bitcoin as being comparable to commodities such as gold, silver or platinum. These are known as ‘store of value’ commodities because they retain value over long time frames due to their relative scarcity. It is difficult to significantly increase their supply i.e. the process of searching for gold and then mining it is expensive and takes time. Bitcoin is similar because it is also scarce. In fact, it is the first-ever scarce digital object to exist. There are a limited number of coins in existence and it will take a lot of electricity and computing effort to mine the 3 million outstanding coins still to be mined, therefore the supply rate is consistently low.

Stock-to-flow ratios are used to evaluate the current stock of a commodity (total amount currently available) against the flow of new production (amount mined that specific year).

For store of value (SoV) commodities like gold, platinum, or silver, a high ratio indicates that they are mostly not consumed in industrial applications. Instead, the majority is stored as a monetary hedge, thus driving up the stock-to-flow ratio.

A higher ratio indicates that the commodity is increasingly scarce — and therefore more valuable as a store of value.


According to that, when we decide to buy or sell bitcoin, the first thing we do is check the S2F model price to see how far we are and if we need to sell the bitcoin or wait for the price to increase.

Average true range

Average true range (ATR) is a volatility indicator that shows how much an asset moves, on average, during a given time frame. The indicator can help day traders confirm when they might want to initiate a trade, and it can be used to determine the placement of a stop-loss order.

As you can see from the backtests, ATR gives us a very accurate signal when it’s time to sell.

Moving average

A moving average (MA) is a stock indicator that is commonly used in technical analysis. The reason for calculating the moving average of a stock is to help smooth out the price data over a specified period of time by creating a constantly updated average price.

With moving averages we can determine if we are in a bullish or bearish market. If we are in a bearish market, we won’t enter and will wait until the market is bullish.

Over the course of a year of research and backtesting of twenty indicators, eight on-chain data, and several machine learning models, we found the perfect combination to make 120% profit in a market that was 20% down.

Stay tuned as we share this great strategy with the public.

This strategy and any strategy the users want to create can be used with the SmartStopSet application.

Get £100 if you sign up today, only for the first 100 customers

Send feedback to info[AT]


Any information found on this page is not to be considered as financial advice. You should do your own research before making any decisions.

Join Coinmonks Telegram Channel and Youtube Channel learn about crypto trading and investing

Also Read