Technical analysis has fallen out of favour in the finance industry, even as millions of individuals continue to rely on charts. The Chartered Market Technician Association, a body dedicated to chart reading, counts about 2,500 active charterholders. By contrast, the CFA Institute's Chartered Financial Analyst programme, which sets a de facto industry standard, has 200,000 charterholders and has dropped technical analysis as a standalone chapter.
The received wisdom on why the practice lost relevance is that computers took over, making chart reading obsolete as statistical techniques and programming became widely accessible. But the article questions this explanation, noting that horses and cars coexisted for decades and that not every investor has access to data pipelines and teams of statisticians.
Academic work has provided mixed signals. In 1992, William Brock, Josef Lakonishok and Blake LeBaron published a paper finding that simple technical-analysis strategies based on moving averages and trading-range breakouts showed surprising support in Dow Jones data from 1897 to 1986. A year later, Narasimhan Jegadeesh and Sheridan Titman determined that momentum also matters.
However, the argument was never fully settled because of data dredging: trying enough combinations of indicators will eventually produce a strategy that back-tests perfectly. In September 2000, economist Halbert White invented a measure of data-mining bias called the Reality Check method, but it relied on an accurate record of all failed attempts, which went against self-interest to document.