December 4, 2021

Market momentum, Bitcoin and reflexivity

“That's what's the matter with network effects. There is really no way to get network effects without the formation of reflective bubbles.And the more we believe that the age of software and the Internet will be defined and governed by networks, and even more so if we think that the participants in the networks will also own parts of them, we better be prepared for a lot of reflexivity. Perhaps in the long term, the entire global economy will turn into Herbalife. I would definitely listen to a startup pitch like this ”, - Alex Danko.

(Post from Deribit Insights analysts)

The reality of Bitcoin is that its valueis in the hands of the owner. Many have tried to measure the value of bitcoin through on-chain metrics, the ratio of available supply to its growth (Stock-to-Flow) and other methods, but at its core it depends on the level of acceptance by many people around the world. The value of BTC is determined by the subjective assessment that people give to it. The more people value Bitcoin, the more it has (hence the frequent application of Metcalfe's Law).

This uncertain nature of Bitcoin's valuationis both an asset and a liability, as it can both create a FOMO atmosphere and convince people that the price is about to go to zero. People tend to value Bitcoin based on the opinions of other people. When they see their friends, family, or hairdressers talk about the value of Bitcoin, they are more likely to agree that it has value. When they hear the bubble talk on TV, they tend to avoid BTC.

This investor behavior creates a strongreflexivity. Once an uptrend has begun, it will most likely continue until the number of participants open to Bitcoin reaches a climax. A downtrend is the same, but in the opposite direction. This dynamic is causing the market to fluctuate dramatically from month to month, year to year, as minicycles occur within the broader adoption of Bitcoin. The prevalence of technical analysis exacerbates this trend, as many in the market use the same simple indicators to assess trends.

A simple glance at the chart gives an intuitive idea of ​​the trending nature of an asset, since there are fairly clear visual trends in the market that are visible to the naked eye.

Chart executed in TradingView

Trader who has spent some timecrypto markets, for sure understands that following the trend is a key element of success in these markets due to the behavior of crypto investors. But looking at a price chart will only give you as much insight as anything else in life; data is needed to test our empirical conclusions.


There are a number of statistical tests thatcan be performed to determine the autocorrelation (in short, the nature of the trend) of the time interval. One of the main tests for autocorrelation is the Hurst exponent. Originally developed by Harold Edwin Hirst to determine the optimal size of a dam on the Nile, it was later used for the more useful problem of autocorrelation of stocks.


In the table above, the Hirst exponent is calculated for twoyears for various assets, and it should come as no surprise that bitcoin came out on top as the most autocorrelated asset. Stocks are relatively average, while oil and bitcoin show a higher likelihood of a sustained trend during these time periods.

The Hirst exponent, of course, does not reflect the completepicture, because it does not quantify the magnitude of the trend - therefore, if the asset rose by only 1 bp. per day, but every day, then you would find that it would have an extremely high Hurst score, useful for asset trading, but useless for understanding the size of the trend.

This helps to understand the distribution of the dailyreturn and calculate the kurtosis ratio, which describes the “fat-tailed” return on the asset, in order to understand the magnitude of the return. The easiest way to visualize this is to use a bar chart that displays daily earnings. The flatter the histogram, the greater the average movement of the asset.


From the charts above, you can see that Bitcoin andEthereum have a relative spread in the rate of return indicating significant average movement. Combined with the Hirst exponent, one can reasonably conclude that cryptocurrency trends are both more general and wider than traditional assets, which should be good for trend following strategies (and not so good for mean reversion strategies) ...

To reaffirm the aboveresearch, we ran several retrospective tests to see if a simple trend following strategy really works better in crypto markets than in traditional markets. We have defined a basic trend following strategy as follows:

  1. Long only;
  2. Z-score of the price up to the 20-day moving average;
    - Z-grade Is the normalized measure of distance from the mean. 1.8 here means the price is 1.8 standard deviation from the average price distances from the 20-day moving average;
  3. Go long when the Z-score exceeds 1.8 (indicating that the asset has already shown strength);
  4. We close the long when the Z-score is below 1.8, which indicates a loss of momentum.


Above, you can see that crypto assets like ETH and BTC have better returns and higher performance than TradFi, oil and SPX. Our initial thesis seems to have some merit.


To add even more credibility to him, onon an older timeframe, we tested the forward yield of the regular week for bitcoin and the week after the big positive move. From the table above, you can see that after bitcoin shows a strong week, it is more likely that next week's profit will be higher than the weekly average.

These tests take us back to the originalthe conclusions we made about reflexivity. Unlike company cash flows from SPX, or a physical barrel of oil that can be consumed as a source of energy, crypto networks such as BTC and ETH derive their value from sentiment-driven use and confidence. In short, as Soros put it, a higher price leads to an improvement in fundamentals, which leads to a higher price. This concept of reflexivity is actually a very useful active trading tool. This market is much more favorable for playing with the established trend. Determining the moment of trend decay is much more difficult.

The very nature of cryptocurrency markets mustencourage the average trader to be very careful in assessing the decay of the current trend. In future posts, we look forward to continuing this line of reasoning by examining the indicators to look out for when assessing the strength of a trend and its potential reversal.

BitNews disclaim responsibility for anyinvestment recommendations that may be contained in this article. All the opinions expressed express exclusively the personal opinions of the author and the respondents. Any actions related to investments and trading on crypto markets involve the risk of losing the invested funds. Based on the data provided, you make investment decisions in a balanced, responsible manner and at your own risk.

Facebook Notice for EU! You need to login to view and post FB Comments!