header-langage
简体中文
English
Tiếng Việt
Scan to Download the APP
Partnered With:

Will the price of a project with a higher frequency of GitHub code update be increased?

Read this article in 4 Minutes
Not only to see if the team is doing something, but also to see what it is doing.
Original source: The TIE
Original compilation: 0x26


The TIE, a cryptocurrency research institution, released a A report on the correlation between GitHub Commits and cryptocurrency prices to study whether Commits can be counted as a unique leading indicator of currency prices and become a guide for investors.

 

Before understanding whether Commits can be used as a research on the leading indicator of currency price, first understand what is GitHub Commits. Commit: A commit or revision is an individual change to a file. When a developer commits to save work, Git creates a unique ID (aka "SHA" or "hash") to record the specific changes committed, as well as who changed them and when. Commits usually contain a string of messages that briefly describe the changes made.

 

So Commits can represent the continuity and frequency of project updates. Simply put, it can be understood as whether the team is "doing things".

 

The following is the full text of The Tie's research.

 

I don’t know if you have this curiosity, does the activity of blockchain project developers drive the price of cryptocurrency? Let's study together to see if GitHub commits (Commits) can become a leading indicator of currency prices and become a new "wealth code".

 

At first glance, if we compare GitHub commits (blue) with prices (black line), we don't see any apparent relationship.

 

 

To test for meaningful signals, we tried a simple trading strategy of buying and holding for 14 days if commits for the day were high (top 75%). Otherwise, backtest by holding all assets equally.

 

Red line=Trade based on number of submissions

Black line=Equal holding Yes

Grey line = holding a certain token alone

 

< img width="undefined" height="undefined" src="https://image.theblockbeats.com/upload/2021-02-01/cea1b65e3f62a831a3f0038086e56f89cbba1f73.png?x-oss-process=image/quality,q_90/watermark, image_d2F0ZXJtYXJrbmV3LnBuZz94LW9zcy1wcm9jZXNzPWltYWdlL3Jlc2l6ZSxQXzMw,g_center">

 

At first glance, this strategy does appear to be better than simply buying and holding, but It might just be due to luck. In fact, strategies that trade these assets randomly outperform commit-based strategies 20% of the time.

 

Perhaps the number of GitHub commits by itself is not an adequate measure of the added value of these assets, as many kinds of activity can be counted Incoming commits, including simple bug fixes or documentation updates.

 

The chart below shows the last four years of Ethereum-related activity on GitHub, broken down by code. There are some interesting situations, such as the shift from Parity dominance to Solidity dominance; the short burst of Sharding and so on.

 

Perhaps taking this change in commit style as a metric is a better indicator of project development than directly observing the number of commits obvious changes in .

 

 

We compared the percentage of submission types at each point in time to the previous two weeks to measure how the submission types differed from day to day.

 

Following the simple trading strategy described above, the chart below shows how we can use this commit variation as a potential buy Signal; buy and hold for 14 days when the change is greater than 75% (horizontal line). Otherwise, all assets are held equally.

 

 

This strategy seems to be picking up some excellent buying opportunities (and of course the recent XMR buying highs), such as the first signal on EOS and the penultimate on XLM two signals. In fact, this strategy does outperform strategies that just look at the number of commits.

 

This means that if submissions do have an impact on price, it is likely not just the number of submissions but Mostly the content of these submissions.

 

There is only a 1% chance that the random strategy will outperform the commit change strategy above. This already seems to prove that GitHub activity should be a major consideration for traders. However, it is still possible that this result can be attributed to luck.

 

 

To verify that these strategies make sense, we backtested them on more historical data. We found that the 2020 results do seem to be attributable to luck. As the table shows, no single strategy is consistently optimal in all years.

 

 

And looking at the historical data, the results of these strategies are more likely due to luck, especially the 2016-2017 backtest. This is due to a lucky position for Monero (XMR) in 2017 and a nearly 40x increase due to it.

 

If a strategy can randomly capture a fraction of these lucrative events in the market in 2016-2017 , it would define excellent performance throughout the year.

 

The final (probably unsatisfactory) answer is likely to be that it's unclear whether GitHub's activity will affect the price. The few strategies we tested in 2020 seemed to work well, but when they were compared to previous years, their performance was inconsistent.

 

As we have seen with Dogecoin’s recent “explosion” , the crypto market remains driven by strong emotions. It's unclear whether the project's code activity matters.


Original link


Welcome to join the official BlockBeats community:

Telegram Subscription Group: https://t.me/theblockbeats

Telegram Discussion Group: https://t.me/BlockBeats_App

Official Twitter Account: https://twitter.com/BlockBeatsAsia

举报 Correction/Report
Choose Library
Add Library
Cancel
Finish
Add Library
Visible to myself only
Public
Save
Correction/Report
Submit