Support, Resistance & Trend: Does TA Actually Work?

Technical analysis (TA) is the study of price itself — support, resistance, trend lines — to decide where to act. It is also one of the most contested ideas in finance. So before you draw a single line, it helps to be honest about what the evidence actually supports.

What the research really says

Academics have argued about TA for over thirty years. A famous 1992 study found that simple moving-average and breakout rules appeared to beat a random market on a century of Dow data. But later work using stricter statistics showed a problem: if you test thousands of rules and report only the best one, you will always find something that looks good in the past. When researchers corrected for that "data-snooping," most of the apparent edge vanished — and what survived was often eaten by trading costs.

The honest summary from that literature: most ad-hoc TA rules are noise fitted to history. A narrow set of trend signals holds up better, but even those have a modest edge that shrinks as more people trade them.

So a chart line is not a crystal ball. Treat TA as a way to define rules and levels — not a guarantee about the next candle.

Support, resistance and trend, defined

• Support is a price area where buying has previously been strong enough to stop a fall. It is a zone, not an exact number.
• Resistance is the mirror image: an area where selling has stopped a rise.
• Trend is the overall direction — a series of higher highs and higher lows (up), or lower highs and lower lows (down).

These work partly because they are self-fulfilling: enough traders watch the same obvious levels that orders cluster there. That makes them useful reference points, not magic. A level that "held three times" can break on the fourth — and the break itself can be the signal.

Turning a level into a rule you can test

The real value of a level is that it lets you write down, in advance, exactly when you are right and when you are wrong:

• Validation: what must happen to enter — for example, price closes back above a prior support level on rising volume.
• Invalidation: the price that proves the idea wrong — usually just beyond the level. This becomes your stop.
• Because the stop sits a measurable distance from your entry, you can size the trade so a loss is a fixed 1R, and judge every outcome in R-multiples.

This is the move that turns "I think it'll bounce here" into something you can actually evaluate. A level without an invalidation price is just a feeling.

Why honesty beats conviction

If TA's edge is small and decays, then which setups you keep trading matters enormously. The only way to know whether your support-bounce trades or your breakout trades actually make money is to log them, attach a stop so R is defined, and look at the expectancy of each setup over many trades. The chart suggests; the journal decides.

Put it to work in FSP: save a level-based setup as a Strategy with explicit validation and invalidation rules, attach a stop so every trade has a clean R, then check in Analytics whether that setup actually carries positive expectancy.