The founder of — a cryptocurrency data API for machine learning — has conducted a study that ostensibly indicates that the readability of a cryptocurrency’s ticker symbol positively correlates with higher returns. The results of the study were published by data science site Towards Data Science on May 29. founder Gerard Martinez notes that the study’s hypothesis is based on the principle of “cognitive ease,” a phenomenon defined by the Israeli-American psychologist and economist Daniel Kahneman in his book, “Thinking Fast and Slow.”

The phenomenon is based on the tenet that “easily pronounced words evoke a favorable attitude,” with implications for the prospective success of market instruments:

“Stocks with pronounceable trading symbols (like KAR or LUNMOO) outperform those with tongue-twisting tickers like PXG or RDO — and they appear to retain a small advantage over some time.”

The cryptocurrencies in the study — all listed on crypto exchange Binance — were selected on the basis of two criteria: first, that they use a three-letter long ticker symbol, and second, that they trade against bitcoin (BTC). 114 Binance-listed cryptocurrencies were thus included in the research.

Readability was defined according to two separate classifications: the first determined that a three-letter ticker symbol is readable provided that it includes at least one vowel, and the second if the vowel was placed in the middle position.

Having grouped the tickers according to both classifications, Martinez retrieved the 1-hour candlesticks for the first week after each cryptocurrency was first listed on Binance. He then statistically assessed whether the distribution of percentual returns over this time period had a correlation with the readability of cryptocurrencies’ tickers.

For both classifications, the study identified higher mean returns for the readable class of cryptocurrencies than for the unreadable, although in regard to the middle vowel classification grouping, this difference decayed over time, Martinez noted.

In regard to the vowel vs. no-vowel grouping, the difference was broadly smaller, although the study still isolated “at least three points with statistical significance,” leading Martinez to conclude that the study had corroborated the hypothesis to a certain extent.

As reported yesterday, the latest version of The Associated Press Stylebook and Briefing on Media Law has included a new section on cryptocurrencies, providing a popular standard for cryptocurrency names.