Technical analysis has a credibility problem, and most of it is self-inflicted. The people loudest about chart patterns tend to be the worst at calling markets. But underneath the YouTube theatrics, there is a small set of statistical tools that genuinely help — if you use them as what they are, not as oracles.

What TA is, honestly

Technical analysis is descriptive statistics applied to a price series. That's it. A 50-day moving average is the arithmetic mean of the last 50 closes. The RSI is a ratio of average gains to average losses over a window. MACD is the difference between two exponential moving averages. None of these things "predict" the future in any reliable way — they describe the present.

The mistake most beginners make is treating these descriptions as forecasts. The mistake most cynics make is dismissing them because they don't forecast. The honest middle ground: TA tells you where price is, where it has been, and how it's been behaving. That is genuinely useful — just not magical.

Moving averages

The 50-day and 200-day moving averages aren't sacred numbers. They became conventions because they're roughly two trading months and ten trading months — short-term and long-term reference points. If a stock is above its rising 200-day, the long-term trend is up. Below a falling 200-day, the long-term trend is down. That's the entire content of the indicator. Anyone selling you more than that is selling you something else.

The useful application: as a regime filter. Don't try to fight long-term trends with short-term trades. If the 200-day is falling, your default stance on that name should be sceptical — not because the chart is "bearish" but because trends, statistically, persist more often than they reverse.

RSI

The Relative Strength Index measures how much of the recent price action has been to the upside versus the downside. By convention, readings above 70 are "overbought" and below 30 are "oversold." Those labels are misleading. RSI doesn't tell you a reversal is coming — it tells you that recent moves have been one-sided. Strong trends can keep RSI in the 70s or 30s for weeks.

The useful application: divergence. When price makes a new high but RSI doesn't, the move is weakening. When price makes a new low but RSI doesn't, selling is exhausting. Neither is a buy or sell signal on its own. Both are flags that something is changing.

MACD

The Moving Average Convergence Divergence is just the gap between two exponential moving averages (typically 12 and 26 periods), with a signal line on top (9-period EMA of the gap). When the gap widens, momentum is accelerating. When it narrows, momentum is decelerating.

The useful application: catching momentum shifts earlier than raw price action shows them. The crossover signals beloved of trading courses work no better than chance on most assets; the histogram dynamics (the bars showing the gap between MACD and signal) are more informative.

Bollinger Bands

Two standard deviations either side of a 20-period moving average. Price spends roughly 95% of its time inside the bands — that's the math, not a prediction. Bands widen when volatility rises and contract when it falls. The most useful pattern is the "squeeze" — bands contracting to multi-month tight ranges — which historically precedes large moves in either direction.

What to ignore

Fibonacci retracements as anything other than psychological reference points (the math has no causal link to markets — 38.2% and 61.8% are just numbers humans like to round to). Most chart pattern theology (head-and-shoulders, cup-and-handle, etc.) — these patterns are real but the predictive backtests are weaker than salespeople claim. Anything called a "secret indicator" or "leading indicator." If it actually led, it would already be priced in.

A working process

Use TA as one of several lenses, never as the whole view. A reasonable workflow: check the long-term trend first (is the 200-day rising or falling?), then medium-term momentum (RSI direction and divergences), then short-term volatility regime (Bollinger width). Combine with a fundamental and macro view. If TA, fundamentals, and macro all agree, you have a thesis. If they disagree, you have a question — which is often more valuable than a thesis.