Beyond conventional indicators, a study of Twitter sentiment versus crypto prices.

Social media are ever more present in everyday life, reflecting and influencing behaviour. This is more and more being recognized with the traditional financial sector, as evidence is increasing that Twitter sentiment can forecast asset prices. See for example Twitter as a tool for forecasting stock market movements: A short-window event study or Predictive Sentiment Analysis of Tweets: A Stock Market Application.

Looking to the future or past?

Within the crypto space indicators are mainly technical, meaning that they are relying on past prices to predict future returns. Though sometimes with some effectiveness, past prices do not capture changing market conditions. Prediction asset prices means gathering information about the future, not about the past. A technical indicator cannot incorporate if there is an impending change in regulation, boosting the price, or a Pyramide scheme reducing dropping the price. This makes technical indicators (if they do seem to work for some short time) only good for picking up pennies in front of a steamroller.

Exploring the future

Knowing this, I set out to see what sentiment analysis has to offer in relation to cryptocurrency prices. More specifically, I started scraping and analysing Twitter data in order to gauge if this would prove to be an helpful indicator. Without getting too technical, I gathered all Tweets about the 100+ largest cryptocurrencies over time (using so-called cashtags in order to minimize noise), using fancy Python based code. Following on this all these Tweets are scored using an algorithm that determines textual sentiment to see if they are positive or negative on a range from +1 to -1 (ie, a score of 1 means the Tweet is reallllyyyy positive). But before diving in to the technicals, first have a look at the results. The plot below shows sentiment (polarity), a bull/bear indicator (which looks at short term sentiment compared to a longer term average) and sum of market cap of all cryptocurrencies analysed. Sentiment Indicator

Interestingly it shows a decrease in sentiment prior to the market declining at the start of July. When markets started to pick up again sentiment was already recovering for some period. Though the sample is obviously too short to make any bold statements, there seems to be some value in using sentiment as an indicator. That being said it cannot be stressed enough that all indicators are imperfect, and the future is always fussy. Moreover, this indicator should not be seen in isolation and adds so much more value by placing it in to broader context.

What about sentiment on an individual level?

Having seen the first results on aggregate level, it’s interesting to zoom in on a more individual level. As an example Stellar (XLM)is taken, for which below the market cap is shown (top chart) versus several sentiment indicators (bottom 5 charts) individual analysis of XLM

Stellar is a nice example because it has shown some serious decline and rise in market cap over the past months (graph 1). When doing technical analysis on price the sudden increase might have come as a surprise. Sentiment (graph 2), however, was already increasing several weeks before. Other Twitter based sentiment indicators shown are: objectivity corrected sentiment (graph 3), Sentiment of Twitter accounts with a significant follower base (4), recipients (Tweet * followers) of Tweets (graph 5 and 6). Those indicators also spike somewhat before prices, though not as fast as sentiment.

More to explore!

The short analysis above hopefully shown the value of exploring sentiment as an indicator, also for Cryptocurrencies. Having a tool that monitors current sentiment is a valuable asset to have, even though it is never perfect. That said, there is still much more to explore. Due to limitations from the Twitter API there is not much history available. For doing analysis historical data I am limited to the time the tooling is up and running, so more data comes available as time progresses. Moreover there are much more ideas to be explored.

How to stay informed?

These indicators are shared frequently on via a newsletter. Moreover next to sentiment analysis of Tweets, also news messages on global crypto websites are monitored. So if these analyses interest you feel free to subscribe!