Money On Chain — March 13th crypto downfall analysis

Alejandro Bokser, Co Founder and CTO

Introduction

This document presents different evaluations that have been performed to analyze the behavior of the model behind MoC (Money On Chain) during the March 13th Bitcoin (BTC) crash.

We executed simulations of the model against actual data with different parameters, which showed the robustness of the model under extremely unfavorable conditions.

In addition to the simulations on BTC values, we present the performance of the model with real data from the main RSK1 blockchain.

First simulation

In this simulation a series of real values of the BTC price were used, covering from 2014–02–01 to 2020–04–02. The BTC value graph can be seen in the following figure along with its moving average.

In some periods the value of the moving average exceeds that of BTC, while in other periods it is the other way round. We always use the minimum of both prices to calculate the coverage needed to sustain the model.

In a bull run, with real market conditions, it is expected that new collateral will enter the system, since people will seek the free leverage of the BitPRO, therefore increasing the coverage.

However in our simulations we assume that no new collateral is added, with the purpose of increasing the stress of the simulation and the chance of liquidation.

The simulation is done with the following parameters:

• 120 days exponential moving average
• BitPro sale threshold with discount 2
• Settlement interval 30 days
• It starts with 1000 BTC and 1000 BitPro
• No more BitPro are minted
• All possible DOCs are issued every day
• The BTCX have a leverage of 2
• All possible BTCX are issued every day, leaving a minimum of 60% of free DOCs
• 0.5% of DOCs are redeemed daily
• BitPros are not redeemed
• All BTCX are redeemed in the settlement event
• If the emission of BitPros with discount is enabled, only 2% of the maximum possible number of tokens are issued per day.

We observe that even under very negative conditions, the coverage did not drop below two in the March 13th Bitcoin crash, not requiring the sell of BitPro with discount (a mechanism designed to provide incentive for adding collateral to the system, which reduces the risk of BitPro liquidation).

Second simulation

This simulation is focused on the interval from 2020/03/02 to 2020/04/02, in order to have closer data visualization.

This simulation was made with the same parameters as the previous one.

Real data performance evaluation

In the following analysis, actual data series taken from the main blockchain were used.

The date range used is from 2020/02/13 6:40:50 (RSK block 2111334) to 2020/03/17 10:47:02 (RSK block 2197734) when the BTC price dropped about 50%.

Graphs are shown with the BTC block value and the moving average, with the overall coverage and objective of the system.

Conclusions

Under a worst case scenario, the model showed robustness and sustained the Bitcoin crash by maintaining coverage out of risk at all times.

Mechanisms designed to handle extreme conditions were not triggered in the BTC crash.