Sudoswap trade analysis

Dashboard

Tools used

Sudoswap Explorer is a dashboard for visualizing trading activity on sudoswap. It's complementary to other dashboards, such as on Dune (link).

The project has 3 main components:

  • Data fetching:
    • Sudoswap trades are being fetched from Flipside using their SDK
    • Floor- and estimate prices for each NFT collection present in the trading data is fetched from NFTBank.ai using their Python SDK. Both data sources are stored in an AWS Postgres instance.
  • Data processing: we investigate the concept of wash trading (as explained in this tweet from Nansen). The basic idea is that one trader is trading the same NFT using two distinct wallets, thus making it seems like there is more trading activity for a given NFT than there actually is. Those trades also occur within a short time difference of each other. For our example, we consider a trade of type "wash trade" when two addresses trade the same NFT address back-and-forth (i.e. 2 trades) within a 1h period.
  • Data visualization: we build a dashboard (link) for visualizing trading activity (incl. price estimates of each NFT collection) as well as wash trading activities.

Getting started

Environment variables

Set the appropriate values in the sample.env and rename it to .env. An API key for ShroomSDK (from Flipside crypto) is also needed.

Data fetching

To fetch all data, install Python (we suggest Miniconda) and run

conda create -n sudoswap python=3.9
conda activate sudoswap
cd <repo-root>
pip install -r requirements.txt
cd data_fetching
python main.py

Starting dashboard locally

Assuming you installed the dependencies as described in the previous step, now run:

cd dashboard
streamlit run app.py