Losing 99% of funds while short-selling Luna which crashed 99.98% within 1 week

MarketObserv
2022-07-10

This crypto startup firm, Uprise, lost 99% of its client funds while shorting Luna during its price crash. Luna crashed 99.98% within 1 week from Thursday, 5 May 2022 to the following Thursday, 12 May 2022. Just 1 week ago, Luna's market capitalization was USD28 billion. 1 week later, this USD28b crashed by 99.98%. This is the fastest and biggest price crash in crypto history. A short-seller who shorted Luna should be making huge sums of money in quick time. Yet, Uprise lost 99% shorting Luna.

How can this happen?

The answer can be found in the hourly price chart of Luna. Luna surged violently by 708% within an hour from 13 May 2022 10 am-11 am. I guess this is the hour that pretty much killed Uprise. Even if Uprise survived this hour, there were several other hourly intervals when Luna surged above 100% and even 300%.

Short-selling in a strong bear market is dangerous even if you get the direction right because bear markets have very violent rallies. The most a short-seller can earn is 100% but he can lose several 100%, sometimes in a short time. It is quite easy to be killed quickly if what you shorted surged more than 700% within an hour.

There is little time to react if the security you shorted surged multiples of 100% within an hour. This can even happen when you are sleeping if it trades round the clock like crypto or futures. This is less of an issue for machine algorithms than for humans. Since Upsize runs on machine algorithms, it should be fast enough to exit but I guess the position size was too large and there wasn't enough exit liquidity.

Uprise employs Artificial Intelligence trading strategies. These strategies are developed based on historical data. Machine learning strategies will fail for rare events because there is not enough data to train the algorithms to cope with these rare events.

Here are some lessons that I extract from this fiasco.

Lesson 1) Watch position size relative to liquidity. If position size is too large and liquidity is not deep enough to exit fast, the violent price reversal will kill you, especially in a bear mark rally.

Lesson 2) Algorithms based on Artificial Intelligence need to recognize when the market enters into rare situations where the algorithms are not trained to operate in. In these situations, either stop trading or reduce position size drastically. Leverage should be cut to 0.

Lesson 3) Try to avoid trades with the risk of losing more than 100%. Putting 10% of capital into a single position is reasonable and not reckless position sizing but even that moderate size with zero leverage can cause serious injuries if the position goes against you by 700% within an hour. Short-selling stocks and trading in securities with built-in leverage such as options and futures carry the risk of losing more than 100%. So, be acutely aware of these risks when you attend courses that teach these strategies to achieve financial freedom as some trainers may downplay the risks to sell the courses.

Global financial markets have slipped into bear markets today. It is tempting to try out short-selling in this bearish environment. However, unless you have good reason to believe you have an edge in short-selling, it is best to avoid it.



Macro Trend
Monetary policy, various types of price indices... Here is everything about the macro economy!
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Comments

  • SLHengHuat
    2022-07-11
    SLHengHuat
    great info, thanks
  • JessieTheresa
    2022-07-10
    JessieTheresa
    Helpful lessons, great thoughts
  • Juliaaa11
    2022-07-10
    Juliaaa11
    hmm Never know this would happen. Thanks
  • monkeyking
    2022-07-11
    monkeyking
    市场有风险,慎之。。。
  • Blessedme
    2022-07-11
    Blessedme
    Dont let greed control us by trying to short the deal
  • Eng22
    2022-07-11
    Eng22
    Algo can submit and resubmit orders and set to work restlessly .. If the model is not back tested with realistic assumptions it is worrisome. The human factor is still impt for a learning algo.
Leave a comment
1