원저자: TechFlow
For the leeks, paying attention to the calls of various KOL bloggers is an important source of obtaining the code to wealth.
So, is KOL’s shouting a sure win, or just a series of accidental coincidences?
Different bloggers have completely different answers to this question. A 100-fold correct call or a wrong recommendation that returns to zero may become a very subjective survivor bias.
From the perspective of the entire industry, what is the final performance of KOLs in bringing in orders?
In February, several researchers from Harvard Business School, Indiana University Business School, and Texas AM University jointly published a paper titled Cryptocurrency Influencers.
The article studies the crypto-asset-related returns mentioned in approximately 36,000 tweets posted by 180 of the most prominent cryptocurrency social media influencers (KOLs) over a two-year period ending December 2022, covering more than 1,600 tokens.
Key Results
After using machine learning to categorize tweets and tracking the tokens mentioned in tweets and subsequent price performance through various statistical descriptions and tests, the key results are as follows:
1. Crypto influencers’ tweets initially correlated with positive returns. But these tweets were followed by significant long-term negative returns, suggesting they generate little long-term investment value.
2. The above effects of these tweets are most obvious when the elements of small coins, large Twitter followers and self-proclaimed experts are met.
3. The paper uses machine learning methods to classify tweets and finds that the above result pattern is stronger when the tweets have more positive sentiment or are related to purchase recommendations.
Data Examples
Crypto influencers’ tweets show positive short-term returns:
The average one-day (two-day) return rate for tweeting a certain coin is 1.83% (1.57%).
For cryptocurrency projects outside the top 100 by market capitalization, the return rate after one day of shouting is 3.86%
The earliest that gains began to decline significantly was five days after the tweet. The average return for days two through five was -1.02%, indicating that more than half of the initial gains were wiped out within five trading days.
Taking a longer-term perspective, the average cumulative returns ending 10 and 30 days after the tweet are -2.24% and -6.53%, respectively. We further document that these negative ex-post returns are more negative for lower-cap cryptocurrencies (where information and liquidity issues are most severe).
A rough estimate suggests that an individual investing $1,000 in a non-top 100 crypto token on the date of the tweet and holding the investment for thirty days would incur a loss of $79 (7.9%), or an annualized loss of 62.8%.
So-called experts: When influencers claim to be experts, post-event returns are more negative; and when these experts have more followers, the returns are even worse.
Overall, the results show that the average long-term investment advice provided by cryptocurrency influencers is unprofitable. Profits can only be made by exiting the position immediately after the tweet is published, but this strategy may not always be feasible due to insufficient market liquidity. In addition, this immediate selling behavior goes against the never sell culture in the cryptocurrency community.
think
The collective evidence in the paper suggests that investors should be cautious about following investment advice from cryptocurrency influencers, as most gains disappear shortly after the tweet is published.
But the article’s authors acknowledge that the evidence is not yet conclusive. Crypto influencers may simply be chasing trends or promoting tokens that will earn them the most visibility and followers, thereby benefiting them financially.
Additionally, a more innocuous alternative explanation is that the cryptocurrency influencer truly believes that crypto assets will eventually experience high levels of growth. It is also possible that the influencer is simply concerned with recommending short-term purchases and assuming that investors know to sell immediately.
Nonetheless, the paper’s results are still informative, as they provide clear evidence that investment advice is unlikely to be useful if one holds coins for more than a few months or even years.
The paper also suggests that regulators and the business media might prompt more scrutiny of such practices to determine whether these activities are associated with more relevant conflicts of interest.
Appendix: TOP 25 Twitter accounts mentioned in the paper (Affected by the time of the paper research, the table is the ranking 2 years ago)
This article is sourced from the internet: Data interpretation of KOL order effect: How are the returns related to crypto assets performing?
Related: Dismantle the top KOL investment list and find new Alpha
Original|Odaily Planet Daily Author: Wenser In April 2024, Web3 asset data platform RootData released statistics on KOLs participation in project financing in the past six months . Dingaling ranked first with 21 project financings, GmoneyNFT and DCF GOD ranked second and third. Today, Odaily Planet Daily will work with you to disassemble KOLs investment strategies from this table and find new Alpha. https://x.com/RootDataLabs/status/1779848819364245812 Popular tracks: Games, BTC L2, Social According to the above table information and Rootdata official website information , many KOLs main investment targets are concentrated in popular tracks such as games, social networking, and BTC L2 network . Among them, the following projects have received more recognition and investment from KOLs: EESee has 7 KOLs investing in it. It is a decentralized aggregation trading and lottery platform…