An indicator is a calculation made from past prices, drawn on or under the chart. It does not know anything the price does not already contain. What it does is summarise: smooth the noise, measure speed or measure how wide the swings are. That can be useful, as long as you know what each one is measuring.
Moving averages: the direction, late
A simple moving average (SMA) is the average of the last N closing prices. A 20-period SMA on a five-minute chart is the average close of the last twenty five-minute candles, recalculated as each candle closes.
An exponential moving average (EMA) works the same way but gives more weight to recent prices, so it reacts sooner.
What they show:
- Direction. A rising average suggests prices have been higher recently than before.
- Distance. Price far above its average has moved a long way from its recent norm. In a strong trend it can stay that way. In a range it often falls back.
- Crossovers. When a short average crosses a long one, traders read a change in direction. Because averages lag, the signal usually arrives after a good part of the move has happened.
What they cannot do: a moving average is always behind the price, by construction. It describes, it does not predict. In a sideways market, crossovers fire constantly and mostly produce small losses.
RSI: how one-sided the recent moves were
The Relative Strength Index compares the average size of recent up-closes with that of recent down-closes, usually over 14 periods:
RSI = 100 − 100 ÷ (1 + average gain ÷ average loss)
It ranges from 0 to 100. By convention a reading above 70 is called "overbought" and below 30 "oversold". Those words are misleading. A reading above 70 means the recent moves were mostly upward and large. It does not mean a fall is due. In a strong trend, RSI can sit above 70 for a long time while price keeps rising.
A more careful use: RSI is a speed gauge. If price makes a new high but RSI does not, momentum is weakening, which is a reason to be more cautious about new trades in the same direction, not a signal to trade the other way.
Volatility: how far price usually moves
Volatility measures the size of typical moves. Two common ways:
- Average True Range (ATR). The average of each candle's full range, including gaps from the previous close. If a market's 14-period ATR on the hourly chart is 0.80, a typical hour covers about 0.80 from high to low.
- Standard deviation of returns. The statistical spread of price changes. Bollinger Bands draw a moving average with bands two standard deviations either side, so they widen when the market is lively and tighten when it is quiet.
Volatility is arguably the most useful of the three because you can act on it directly:
- It tells you whether a target is realistic. Hoping for a move of 2.00 in ten minutes in a market whose typical hour is 0.80 is hoping for something unusual.
- It tells you how far to place an invalidation level. A level inside the market's normal noise will be hit by chance.
- It tells you when to trade smaller. Higher volatility means a wrong trade is wrong by more.
Indicators overlap
Most indicators are built from the same few inputs, so using five of them is usually one idea counted five times. A moving average, an RSI and a band indicator all react to the same closing prices. Agreement between them feels like confirmation but mostly isn't. Pick one measure of trend, one of momentum and one of volatility, learn them well and stop there.
The evidence the AI read uses
The AI analyzer's evidence list draws on measures like these: the change over the window, the share of up-ticks, short and long moving averages and recent volatility. They are the same simple tools, so you can check them yourself on the chart. Doing so is a good way to find out whether you agree with how the read weighed them.
Common mistakes
- Optimising settings (using 13 instead of 14, or 9 instead of 10) until the past looks perfect. That fits history, not the future.
- Treating "overbought" as "sell" and "oversold" as "buy".
- Adding an indicator every time one disagrees with your idea.
Tip: Pick a market, add a 20-period average and watch it for an hour on a one-minute chart. Count how many times price crossed it, and how many of those crossings were followed by a move that continued in the new direction.
