Skip to main content
Featured

Pulling Uniswap V3 On-Chain Swap Data

Dr. C

Have you ever wondered how to obtain historical Uni V3 data? 🧵

We created a short tutorial on how to do it using Python and GBQ (LINK HERE)

NB: ☝️ can also be extended to other protocols.

In short, we created a query that:

  1. Selects the relevant columns from GBQ’s public Ethereum logs database
  2. Filters info by pool address (e.g. USDC-ETH pool at 0.3%)
  3. Extracts relevant transaction data (e.g. amounts, sqrtPrice...)
  4. Cleans & sorts the data.

NB: You will need a Google Big Query credential. See here.

You will also need the following (standard) python packages:

  • pandas
  • matplotlib
  • http://google.cloud
  • google.oauth2
  • numpy

Why do we want this data? It contains a wealth of information! As an example, by investigating the data for the ETH-USDC pool at a 0.3% fee, one can plot the asset price in the pool.

im1.png

We can also look at the histogram of the time between any two transactions:

im2.jpg

Notice that this distribution resembles an exponential distribution with parameter λ = 0.007, meaning that, on average, there is a transaction every ~136 seconds in this pool.

This tutorial is intended to be the "first step" in our series of #ResearchBites investigating Uni V3 pool data (e.g., prices, transactions, etc.), as well as trading strategies in these pools. Stay tuned for more upcoming #ResearchBites from the Panoptic_xyz team!