An indicator transforms data into a simpler view. A moving average smooths prices; a momentum measure compares recent gains and losses; a volatility measure summarizes movement. The transformation can make a question easier to answer, but it does not create new observations.
Before using an indicator, explain its inputs and what its number would do under a simple price path. That is often more useful than memorizing a conventional buy or sell threshold.
A line can move while price stands still
Take five closes: $98, $99, $100, $101 and $102. Their five-period simple moving average is $100. If the next close is again $102, the oldest $98 observation leaves the window. The new average is $100.80.
($99 + $100 + $101 + $102 + $102) ÷ 5 = $100.80.
The average rose despite no change in the latest price. Its movement came partly from an old low observation dropping out. Understanding the window helps distinguish new information from the arithmetic of the indicator.
CME’s moving-average lesson explains the distinction between a simple average and an exponential average, which assigns greater weight to recent prices. Greater responsiveness reduces some delay but also reacts more to short-lived fluctuations. The best setting depends on the question, rather than on one universal period length.
Multiple indicators can repeat the same evidence
A rising moving average, positive momentum and a bullish crossover may all be consequences of the same recent price rise. Treating them as three independent confirmations can exaggerate confidence.
Ask what each addition contributes. Volume might describe participation, although its coverage matters. A calendar identifies an approaching event that price history alone does not contain. Another differently colored transformation of the same closes may contribute less.
Thresholds also require context. A high momentum reading can accompany a persistent trend or precede a reversal. Calling it “overbought” does not specify which outcome is expected, on what horizon, or why anyone must sell.
When the signal becomes tradable
“Buy a bullish crossover” needs a definition. Which series crosses which? Is the rule evaluated during the candle or after its close? At what price can the order first be submitted? What closes the trade, and what happens if the signal reverses before a fill?
If a rule uses the completed daily close, it cannot assume that close was known earlier in the day. Even an order placed immediately afterward faces a bid-ask spread and potentially a different price. An appealing historical signal can disappear when information timing and execution are made consistent.
Information is not yet profit
Lo, Mamaysky and Wang’s research found incremental information in some formally defined chart patterns in a historical stock sample. The authors did not equate that with a guaranteed profitable strategy. A pattern can shift the distribution of returns slightly while transaction costs consume the benefit.
Testing many settings adds another difficulty. If you try hundreds of combinations and report only the best, that result benefits from the search itself. Sullivan, Timmermann and White study this data-snooping problem in technical trading rules.
Keep a record of the alternatives tried. Reserve later data for a test after the rule is fixed, compare against a simple baseline and include costs. Look for a result that remains sensible across nearby settings, rather than one unusually profitable parameter.
The five-period average rose because an old low price left the sample. That can help describe a trend, but it is not a new observation about demand. A trading rule needs a reason that this transformation of past prices helps predict, execute, or size the next position.