How to sense market ‘buzz’ on the virtual trading floor

Throught history, people have traded, but they have done so face-to-face, in physical encounters, until quite recently, with the emergence of technology and the internet.

This meant that psychology was also in play, as dealers analysed the mental state of others, when trying to figure out trading strategy.

But when technology took over, did valuable psychological information about the mind of the market disappear? Does modern trading require new skills in getting a sense of marketplace psychology? Would that help reduce the uncertainty with which we trade today?

Until recently, most markets, such as commodity exchanges and stock markets, boasted physical “trading pits”, where the traders communicated with (or, as they got excited or flustered, shouted at) each other, while dealing.

In 2001, Joshua Coval (professor of business administration at Harvard Business School) and Tyler Shumway (professor of finance at the University of Michigan) analysed the noise level in the Chicago Board of Trade’s 30-year Treasury Bond futures trading pit.

Sound levels of such exchanges were referred to as the “buzz”, but did it contain useful information?

Entitled “Is Sound Just Noise?”, the study found that following a rise in sound levels, prices became more volatile and market depth declined. Market depth refers to a security’s ability to tolerate the execution of large market orders without having a large effect on the security’s price.

Since trades can be conducted entirely with hand signals or body language in an open “outcry” setting, such as these trading pits, the authors of the study argued that periods of high sound level (shouting) reflect a need for immediacy in dealing. This demand gets stronger when traders become more fearful of possible increases in the costs of executing their trades.

On electronic exchanges, even if traders could somehow communicate their eagerness to trade, such signals are useless. This is because “priority” is always observed in electronic markets. If two traders submit orders of identical pricing, the one who submits initially will always be dealt with first.

But in the open outcry setting, traders often deal with the first order they notice, instead of the placement order. As a result, how loudly a trader yelled influenced the odds and speed with which an order got filled. Such efforts to push an order’s priority required physical energy. Traders only exerted themselves when they were particularly eager to deal.

The overall noise level of a physical exchange revealed how anxious traders were to trade. Anxiety in the pit conveyed useful information regarding current and future trading conditions.

Elevated trading pit sound levels were highly significant in forecasting increased price volatility several minutes into the future. Trader anxiety, which was reflected in heightened exchange sound, strongly forecasted future declines in market depth.

A key implication of this research could be that technology may not be a perfect psychological substitute for old-fashioned, face-to-face trading. Electronic trading doesn’t convey the kinds of emotional signals embedded in sound levels from trading pits.

Technology has limited ability to transmit the fear in a trader’s voice. As trading volumes migrated to electronic exchanges, has psychological information useful in trading been lost? Have anxious traders compensated by looking elsewhere for this kind of information, perhaps in rumour and gossip on the internet?

Does Twitter provide a new opportunity to hear the “buzz” of the market or trading pit?

An academic study, entitled “Twitter mood predicts the stock market”, claimed an accuracy of 87.6 per cent in predicting the daily closing values of the Dow Jones Industrial Average by using emotional words such as “calm”, “alert”, “sure”, “vital”, “kind” and “happy” in Twitter feeds.

That study is from 2011, but Jermain Kaminski from the MIT Media Lab published a different one in 2016 investigating whether Twitter can help predict bitcoin market movements.

In “Nowcasting the Bitcoin Market with Twitter Signals”, 161,200 tweets from 57,727 unique users were collected within 104 days. Twitter posts containing “bitcoin” and a positive, negative, or uncertainty related words were analysed.

Words such as “happy”, “love”, “fun”, “good”, “bad”, “sad” and “unhappy” represent positive and negative emotional signals, while “hope”, “fear” and “worry” are indicators of uncertainty. A day with a low amount of negative tweets correlated with a higher close price. A higher ratio of positive to negative tweets is accompanied by a higher close price.

The study concludes that “nowcasting” (predicting the present) might be applicable to bitcoin daily markets, and that Twitter has become bitcoin’s virtual trading floor, emotionally reflecting its trading dynamics.

Clearly it’s harder than it looks to take the emotion out of trading.

Let's block ads! (Why?)