"Does the Team's Activity Really Relate to Cryptocurrency Prices?"

The Four Bull and Bear Cycles in the Crypto Market

The genesis block of Bitcoin was created in 2009, and in the subsequent 14 years, its price has undergone several cycles of bull and bear markets. Notable industry narratives such as the “ICO Era,” “Public Chain Explosion,” “Defi Summer,” and “NFT Wave” have emerged.

For analysis convenience, this article defines Jul 2015 - Jan 2018 as the first bull market, Jan 2018 - Mar 2020 as the first bear market, Mar 2020 - May 2021 as the second bull market, and May 2021 - present as the second bear market.

The first “ICO” bull market from Jul 2015 - Jan 2018 is too distant, with too little data available for rigorous analysis. Therefore, this article focuses on the latter three cycles.

The Four Bull and Bear Cycles of the Crypto Market

What factors indicate that “the team is actively working”? We’ve identified six factors!

In the industry, the vast majority of projects are based on blockchain technology, and their code is open-sourced on GitHub (a platform for code hosting and sharing).

Therefore, Falcon uses six factors from GitHub as quantitative standards to measure if “the team is actively working”. These include: Stars, Forks, Commits, Issues, Pull Requests, and Watchers. The following are the specific meanings and types of these six factors.

Detailed Introduction to the Six Factors of GitHub Data for Projects

All the GitHub data of the projects mentioned in this article can also be viewed on Falcon’s product. Visit the link: https://falcon.lucida.fund/ch/asset_tracker/73/github?uid=

Product Page Screenshot

Effective Sample Size and Glossary

The team analyzed coin price trends and corresponding GitHub six-factor data for three market cycles. After outlier processing, the effective token samples retained were 81, 330, and 596 for each market cycle respectively.

Here are explanations for the terms appearing in the following charts:

Specific Explanation of Terms

First Bear Market (Jan. 2018 - Mar. 2020): GitHub data had some resistance to price declines, but the effect was limited, possibly related to the small sample size

Starting with the first bear market:

Descriptive Statistics for GitHub Data’s Six Factors and Coin Price Fluctuations:

In the first bear market, token data was more dispersed, characteristic of the early stages of the crypto market’s rise. During this period, the standard deviation of seven statistical measures was far from the average, indicating significant differences between different coin types in terms of price and GitHub data. In this phase, more mature tokens like Bitcoin and ETH had extremely high attention on GitHub, but many emerging tokens had low GitHub visibility and developer contributions.

The statistical situation of coin prices that fell less than the average (highlighted in bold black) and their corresponding GitHub data’s six factors:

Tokens marked with grey cells represent those contrary to market trends. We consider these tokens to have unique characteristics that require comprehensive analysis with market conditions. In this period, only Binance-exchange was an exception. Observing its six GitHub factors, star and fork values were in the top 10, but commit, issues, pull requests, and watchers were extremely low. This was mainly because BNB, before 2019, was only considered a “platform coin” without “public chain” attributes, hence the code was not open-sourced. During the second half of 2018, market focus was on platform coin segments, and BNB’s rise was significant, resisting the downturn in that cycle. For this coin, only star and fork factors on GitHub had some correlation with its price.

Among tokens that fell less than average, 40% had GitHub factors in the top 10 of the statistics. The remaining tokens generally had lower GitHub profiles, suggesting that GitHub factors had a positive effect on reducing price declines, though not significantly.

Second Bull Market (Mar. 2020 - May 2021): More active GitHub projects saw greater increases during the bull market

Descriptive Statistics for GitHub Data’s Six Factors and Coin Price Fluctuations:

In the second bull market, token data was more concentrated, indicating increased maturity and prosperity in the crypto market. The standard deviation of seven measures was closer to the average compared to 2018-2020, indicating a more concentrated sample distribution. Market analysis reveals that tokens developed more maturity by 2020, with tokens that emerged in 2018 experiencing growth and an increase in corresponding fundamental GitHub data. Additionally, the number of tokens issued during this period significantly increased, further centralizing data distribution.

The statistical situation for coins whose price increases exceeded the average (highlighted in bold black) and their corresponding GitHub data’s six factors:

Out of 330 data points, 11 had price increases above the average, with 5 of them having GitHub factors above average, accounting for about 45%. Preliminary analysis suggests a correlation between increased GitHub data and price rises, with specific correlations detailed in the third part of the article.

Projects that didn’t rise but fell during the bull market were typically very inactive on GitHub

Coin Price Anomalies (Price Drops in Bull Market):

Out of 330 effective samples in this period, 28 tokens went against the trend and fell in price, highlighting their weakness. Correspondingly, 90% of these tokens had GitHub data below average and tended toward the minimum.

Second Bear Market (May 2021 - Present): More active GitHub projects contributed to resistance in the bear market, but the effect was not substantial

Descriptive Statistics for GitHub Data’s Six Factors and Coin Price Fluctuations:

Sorting by the star factor, the top 20 tokens and their other six statistical measures (tokens exceeding the average highlighted in bold black):

With further development in the crypto market, data in the second bear market became more dispersed, likely due to increasing industry divergence. The standard deviation of seven measures varied greatly from the average, indicating that token data was more diverse during this bear market phase. The token market in 2021 was still in a robust development phase, attracting more people to the token market, with initial focus on well-developed and mature token projects. These tokens had GitHub attention with tens of thousands of statistics, but emerging tokens of this period still needed time to gain public recognition and naturally had lower visibility and development.

Analyzing the top 20 tokens by Star data ranking, it is observed that tokens exceeding the average in GitHub’s six-factor ranking show certain similarities in statistical patterns, suggesting a high correlation among these six factors. It is also noted that tokens with particularly high rankings in these six factors are more mature, mostly issued between 2015 and 2018, including Bitcoin, ETH, and Dogecoin.

Abnormal token price behavior (price increase during bear markets):

Out of 596 tokens, 28 anomalies were observed. Among these, six tokens, accounting for 28%, had one or more factors exceeding the average in GitHub data. According to the data, it’s inferred that an increase in GitHub data contributes to resilience during bear markets, though its impact is not particularly significant. The strong price advantage of such tokens is primarily determined by factors from other categories.

How to quantify the correlation between GitHub factors and price? Which coefficient will be used for this assessment?

As noted earlier, GitHub data plays varying roles in bull and bear cycles.

So, how do we quantify the correlation between GitHub factors and price?

A Q-Q plot uses the quantiles of the sample as the horizontal axis and the corresponding quantile points calculated according to the normal distribution as the vertical axis, displaying the sample points in a Cartesian coordinate system. If the dataset follows a normal distribution, the sample points form a line around the diagonal of the first quadrant. For datasets following a normal distribution, Pearson’s correlation coefficient is appropriate for analysis, while Spearman’s correlation coefficient is suitable for datasets that do not follow a normal distribution.

The results of the Q-Q plots for the six factors in three intervals are as follows:

As shown, the sample points for the six factors – Star, Fork, Commit, Issues, Pull_requests, Watchers – do not cluster around the diagonal of the first quadrant, indicating that they do not follow a normal distribution. Thus, the correlation analysis of these six factors with token prices will be based on Spearman’s coefficient.

First Bear Market (Jan 2018 - Mar 2020): Limited correlation between GitHub factors and token prices due to sample size

Correlation table of the six factors with token price appreciation:

The five GitHub factors positively influence token price resilience during bear markets. The correlation coefficients of star, fork, issues, pull_requests, watchers with price are around 0.260, showing significance at the 0.05 level, indicating a positive correlation with token prices.

The commit factor showed no significant relationship with price appreciation in this interval. The correlation coefficient of commit with price fluctuation was -0.032, close to 0, and the P-value was 0.776 > 0.05, indicating no correlation between commit and price.

The correlations of star, fork, issues, pull_requests, watchers with price align with our previous assessment, showing a positive impact, although not high. A correlation of 0.260 is meaningful for our subsequent research into token price trends and constructing related factor strategies. The result for commit differs slightly from our previous findings, tentatively attributed to the limited sample data. In the second and third intervals, more token data was collected to further examine the correlation between commit and price.

Second Bull Market (Mar 2020 - May 2021): More active GitHub, higher token price gains

Correlation table of the six factors with token price appreciation:

In the second bull market, with the effective sample size increasing from 81 to 330, the correlation of the six factors – star, fork, commit, issues, pull_requests, watchers – with price significantly strengthened, around 0.322, markedly higher than the average correlation of 0.260 in the first interval, and significant at the 0.01 level. The correlation of star, commit, watchers with price was as high as 0.350. In this interval, all six factors had a positive correlation with price, seemingly confirming our conjecture of a negative correlation between commit and price in the first interval, possibly due to limited data and the influence of outliers.

Second Bear Market (From May 2021 to present) GitHub factors have timeliness! Still significantly correlated with token prices in bear markets, but not necessarily resilient

Correlation table of the six factors with token price appreciation:

For the third interval, with an increase in effective samples to 597, the correlation between the six factors – star, fork, commit, issues, pull_requests, watchers – and price strengthened compared to the first interval, with an average correlation of 0.216 under the significance level of 0.01, slightly higher than the 0.205 in the first bear market but significantly weaker than the 0.322 correlation in the second interval.

It is believed that the six GitHub factors are positively correlated with token price appreciation, but they have a certain timeliness!

The six factors demonstrate stronger predictive and contributory power to the price fluctuations of cryptocurrencies during a bull market. However, their utility is relatively weaker in a bear market. In such scenarios, cryptocurrency prices are more influenced by other broad categories of factors, such as volume-price factors and market sentiment (including alternative factors). Data from GitHub serves only as a part of the fundamental analysis, playing a relatively limited role.

Conclusion

Based on the above content, Falcon summarizes the conclusions of this article as follows:

1. With the development of the Crypto market and the flourishing of the developer ecosystem, the correlation between GitHub data and cryptocurrency prices is increasingly strong.

2. From an investment perspective, it is advisable to invest in projects with active GitHub development and avoid those with inactive GitHub development.

3. In bull markets, projects with more active GitHub activity tend to have higher gains; in bear markets, those projects tend to be more resilient to downturns.

4. The correlation between GitHub activity and cryptocurrency prices is significantly stronger in bull markets than in bear markets.

About LUCIDA & FALCON

Lucida (https://www.lucida.fund/) is a leading quantitative hedge fund that entered the Crypto market in April 2018. It primarily trades strategies like CTA, statistical arbitrage, and options volatility arbitrage, and currently manages assets worth $30 million.

Falcon (https://falcon.lucida.fund/) is a new generation of Web3 investment infrastructure. Based on a multi-factor model, it assists users in “selecting”, “buying”, “managing”, and “selling” crypto assets. Falcon was incubated by Lucida in June 2022.

For more information, visit https://linktr.ee/lucida_and_falcon

Disclaimer：

1. This article is reprinted from [mirror]. All copyrights belong to the original author [LUCIDA & FALCON]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.