Algo Trading, and the importance of proper PNL Reporting
We have all seen the promise of infinite wealth in regards to making passive income from trading bots. Percentage returns in the thousands, perfect entries, and perfect exits. If only it were this simple. Many times when results of a particular algo are listed you are seeing the “signal” entries and exits. This is the price at which the algorithm called for trade execution, and while this method of reporting works from a testing and development standpoint, real-world performance can be significantly different.
Backtesting on platforms such as Tradingview gives you results based on “paper trading”. The optimal entries and exits and the exact point at which an algorithm called for placing a trade. This does not take into account real-world factors such as slippage, market movement in the time between the signal being sent and the trade being executed, funding while in a perp market trade, and trade fees. Unfortunately, many platforms chose to use this data in their PNL reporting, as this often times shows the best possible “perfect world” results. In addition to this, it is the easiest way to track and report data. We believe this is not the proper way to report PNL.
At CompendiumFi we looked at various ways to record and display PNL of the various algorithms on the platform to come up with the best possible way to ensure accurate, “real life” forward testing data, while also protecting the intellectual property of our third-party authors. Every bot that is marked as a “Verified Strategy” undergoes a lengthy forward testing process where signals are sent to our system from the authors' code and executed on a dedicated live trading account. This is real money, being traded on a real account, no different than a customer signing up to a bot. THESE are the results that are shown. They include all slippage, market fluctuations, trade fees, funding fees, etc. Only after the strategy has proven itself in a live trading account for a set period of time, does it make it to the marketplace.
We believe in total transparency for our users, as well as protecting the intellectual property of the algo authors.
If you are a bot author and are interested in listing your strategy on CompendiumFi you can contact us at firstname.lastname@example.org or join our discord https://discord.gg/JmxFxrmQ5M. Complete details on how to list a bot can be found here.