CTX Delegation Rewards Generator

This repository contains code to calculate rewards discussed in this proposal.

Setup

1. Create a database in postgres create a database named ctx_delegation_stats in postgres

CREATE DATABASE ctx_delegation_stats;

2. clone the repo

 git clone https://github.com/cryptexfinance/ctx-delegation-stats.git

3. Change directory to get access to the python scripts

 cd ctx-delegation-stats/stats

4. Setup env variables create a .env from the existing env.sample file

cp env.sample .env

set the DATABASE_URL. Please refer to the sqlalchemy docs for more details.

5. Install python packages In a python virtual env install the python libraries by running:

pip install -e .

6. Create the database models

python run.py create-models

7. To fetch all delegation data from the subgraph, run

python run.py fetch-and-store-data

Generate Distribution

Please make sure that you complete all the steps in setup before generating the excel sheet for the distribution.

To generate the excel sheet for the distribution run:

python run.py generate-distribution

This will generate a file named distribution_data.xlsx in the current working directory.

Here's a link to the excel sheet generated using this code: link

Distribution procedure

The code for the distribution process can be found here

Distribution Math

Formula Let’s say there are 2 keepers K1 and K2 and 5 users U1, U2, U3, U4, U5.

Let’s say the total reward pool is R.

We will allocate 20% of the rewards i.e. 0.2R to the keepers as rewards and 80% to the users i.e. 0.8R.

Note: The numbers above can be changed based on feedback. Those are tentative numbers.

The formula for the distribution is explained using an example.

Let’s say that K1 voted on n1 proposals and k2 voted on n2 proposals.

The reward for K1 = (2 ** n1 * 0.2R)/(2 ** n1 + 2 ** n2)

An exponential function is chosen here to reward active keepers more than non-active ones.

Similarly reward for K2 = (2 ** n2 * 0.2R)/(2 ** n1 + 2 ** n2)

To understand the user distribution, let’s an example.

Let's the user balance was as shown in the table at the time when each proposal was voted on.

User	Delegated to	No of days	Amount staked	ProposalId
U1	K1	d1	a1	1
U1	K1	d2	a2	2
U2	K1	d3	a3	1
U2	K2	d4	a4	1
U3	K3	d5	a5	2

Also, let's say the table for the keeper votes looks as follows:

Keeper	proposalId
k1	1
k1	2
K2	1
k3	2

We can see that K1 voted on 2 proposals, K2 voted n 1, K3 voted on 1 and K4 didn't vote on any proposals.

We first need to calculate rewards for each of keepers delegators. The reward pool for each keeper will be calculated as User reward pool for K1 UrK1 = (2 **2 * 0.8R)/(2**2 + 2**1 + 2**1) User reward pool for K2 UrK2 = (2 **1 * 0.8R)/(2**2 + 2**1 + 2**1) User reward pool for K3 UrK3 = (2 **1 * 0.8R)/(2**2 + 2**1 + 2**1) User reward pool for K4 UrK4 = 0

We further divide the keeper rewards for each proposal based on the number of proposals the keeper voted on K1 User reward pool for proposalId 1 UrK1P1 = UrK1 / 2 K1 User reward pool for proposalId 2 UrK1P2 = UrK1 / 2 K2 User reward pool for proposalId 1 UrK2P1 = UrK2 K2 User reward pool for proposalId 2 UrK2P2 = 0 K3 User reward pool for proposalId 1 UrK3P1 = 0 K3 User reward pool for proposalId 2 UrK3P2 = UrK3 K4 User reward pool for proposalId 1 UrK4P1 = 0 K4 User reward pool for proposalId 2 UrK4P2 = 0

Reward for U1 for proposal 1 for delegating to keeper 1:U1K1P1 = (d1 * a1 * UrK1P1) / (d1 * a1 + d3 * a3) Reward for U2 for proposal 1 for delegating to keeper 1: U2K1P1 = (d3 * a3 * UrK1P1) / (d1 * a1 + d3 * a3) Reward for U2 for proposal 1 for delegating to keeper 2: U2K2P1 = (d3 * a3 * UrK2P1) / (d1 * a1 + d3 * a3) Reward for U3 for proposal 1 for delegating to keeper 3: U3K3P1 = 0

Please note that U2 has been rewarded twice becasue they had delegated to 2 keepers during proposal 1

Reward for U1 for proposal 2 for delegating to keeper 1 U1K1P2 = (d1 * a1 * UrK1P2) / (d1 * a1) Reward for U3 for proposal 2 for delegating to keeper 3 U3K3P2 = (d5 * a5 * UrK3P2) / (d1 * a1)

Total reward for U1 TU1 = U1K1P1 + U1K1P2 Total reward for U2 TU2 = U2K1P1 + U2K2P1 Total reward for U3 TU2 = U3K3P2