# We traded a Pump & Dump scheme with a Trading bot

A pump and dump scheme is a type of financial fraud that involves artificially inflating the price of an owned stock through false and misleading positive statements, so as to sell the cheaply purchased stock at a higher price. In our case, we attempted to automate our partecipation  using a bot.

Please note that engaging in such practices is unethical and illegal in many jurisdictions, and this information is provided purely for educational purposes.

Anyway…

We executed two different Python scripts in our experiment. The first script was a basic trading bot that monitored a specific Telegram channel, waiting for the trading symbol to be announced. The second script was designed to track and store Level 3 data for subsequent data analysis.

# Trading Bot

There's nothing special here. We simply used the **telethon** Python module in order to connect to our personal Telegram account and then we followed the Exchange API documentation to set an authenticated API call for placing orders.

# Tracking Level 3 data

This is what we've been seeking: Data. Specifically, Level 3 data, the highest quality obtainable from the Kucoin Exchange. We initiated real-time streaming of all transactions for 628 different tickers, and stored the results in a time series data database.

This is what Level 3 data looks like:

# Deploy the bot on AWS

In our endeavor to stay as closely connected as possible to the Exchange servers, we made the decision to deploy our bot on an EC2 instance. EC2, an acronym for Elastic Compute Cloud, is a service provided by Amazon Web Services (AWS).

Among the multiple regions available, we decided upon us-west.

This decision was driven by the region's low latency or 'ping', a critical factor in trading where every millisecond counts. This region provided us with reliable, low latency access to the Exchange servers, thereby ensuring that our bot could execute trades with minimal delay.

# The Trade

`1704474023.5695188`

What is this? It's a timestamp. Let's transform it into a more comprehensible format:

Friday, January 5, 2024 5:00:23.569 PM

This is the precise moment the channel announced the ticker name.

But what about the actual trade entry time?

Here it is:

`1704474023.5695188`, which corresponds to:

Friday, January 5, 2024 5:00:24.310 PM

This is less than a second difference. Quite impressive!

**HOWEVER, IT'S NOT SUFFICIENT.**

# The Results

The price of that symbol surged by +800% in a single one-minute candle, but we managed to gain "only" 20% in two seconds.

While a 20% gain may seem substantial, it pales in comparison to the potential +800% increase.

# Conclusions

The ultimate aim of this experiment, excluding the speculative gambling aspect, was to observe what happened just prior to the message being released to these Pump Channels. This was in an attempt to find a way to predict the ticker ahead of the initial pump, or in other words, to be quicker than bots like the one we developed.

However.. as you can see, the prospect of making money by participating in these Pump & Dump schemes is slim.
Despite having a well-written script and a reliable server close to the Exchange servers, we only managed to make a 20% profit, even though we entered at a 500% increase from the start of the pump.

It's not worth it. Algorithmic trading is a vast field; you should consider something more sustainable and supported by a serious research and environment.

You can explore our algorithms on TradeStackX, or hire us to develop a unique strategy and solution to enhance your trading business. Take a look at the box below.

<PricingAlgos header="Deploy our Algos or your best strategy." />

We executed two different Python scripts in our experiment. The first script was..