6. Indicator Combinations — Confluence That Actually Works

In Chapter 5 you met the indicators one by one. In the wild, no single indicator is enough. Pros stack 2–3 indicators from different families so each one answers a different question. That overlap is called confluence, and it’s the difference between a coin-flip signal and a setup with an edge.

⚠️ This chapter is a toolkit, not a list of “buy buttons.” Every combo below has failure modes. Always pair with the rules in Risk Management.

The two rules of combining indicators

Rule 1 — One per family, max

The four families from Chapter 5 each answer a different question:

Family	Question	Examples
Trend	Which direction?	EMA, MACD, Supertrend, ADX
Momentum	How strong? Overbought?	RSI, Stochastic, CCI
Volatility	How much movement?	Bollinger Bands, ATR
Volume	Is the move real?	VWAP, OBV, MFI

Stacking RSI and Stochastic and CCI is not confluence — they all measure momentum and will agree (and be wrong) at the same time. That’s called multicollinearity, and it gives a false sense of certainty.

Rule 2 — Two or three. Never five.

Each extra indicator narrows your signals — fast. Five indicators that “must all align” produce maybe 2 trades a year, half of which still lose. Stick to trend + momentum, or trend + volume, or volatility + momentum. That’s it.

The 7 combos worth learning

Each combo below follows the same structure: what it shows → entry/exit logic → when it fails.

1. EMA(20) + EMA(50) + RSI(14) — the “hello world” combo

Trend filter: price above both EMAs and 20 > 50 = uptrend regime.
Momentum trigger: RSI dips to 40–50 zone (a pullback, not oversold) and turns up.
Entry: next candle close above the prior high.
Stop: below the recent swing low or below 50 EMA, whichever is closer.
Fails when: market is choppy/sideways — EMAs cross repeatedly and RSI oscillates around 50. Use ADX > 20 as a regime filter to avoid this.

2. MACD + 200 EMA — trend-aligned momentum

The classic “only trade with the major trend” filter.

Long only when price > 200 EMA.
Entry: MACD line crosses above signal line while histogram turns positive.
Stop: below recent swing low.
Fails when: strong reversals — the 200 EMA is slow, so it’ll keep you long into the start of a bear market for weeks. Combine with a hard stop, not “wait for 200 EMA cross.”

3. Bollinger Bands + RSI — mean reversion

For ranging stocks, not trending ones.

Setup: price tags the lower Bollinger Band and RSI < 30.
Trigger: next candle closes back inside the band (rejection of the extreme).
Target: middle band (20 SMA); aggressive target is the upper band.
Stop: a defined % below the low of the rejection candle.
Fails when: the stock is in a strong downtrend — price walks down the lower band for days and RSI stays oversold. Filter: only take it when price > 200 SMA.

4. VWAP + Volume — the intraday workhorse

Used by institutions and Indian retail intraday traders alike.

Bias: price > VWAP = bullish for the day; price < VWAP = bearish.
Entry (long): pullback to VWAP with declining volume, then a bounce with rising volume.
Stop: below VWAP by 1× ATR(14, 5-min).
Exit: end of day (intraday-only tool — VWAP resets at 9:15 IST every session).
Fails when: gap days or news-driven moves — VWAP gets dragged by the open and gives weak signals. Skip VWAP trades in the first 15 minutes.

5. Supertrend + RSI — the Indian retail favourite

Hugely popular on Zerodha Kite and TradingView in the Indian community because Supertrend gives an unambiguous green/red regime.

Regime: Supertrend (10, 3) green = long bias only.
Entry: RSI(14) crosses up through 50 while Supertrend is green.
Stop: the Supertrend line itself (it trails — that’s the point).
Fails when: whipsaw markets — Supertrend flips colour every 2–3 candles and you get chopped. Add a higher-timeframe Supertrend filter (e.g. only take 15-min signals when the 1-hour Supertrend is also green).

6. Donchian 20-day high + ADX — the breakout filter

Donchian channels = highest high / lowest low of last N periods. The 20-day high is the original “Turtle Trader” breakout signal.

Entry: close above 20-day high and ADX(14) > 20 (trend strength is real).
Stop: 2× ATR below entry, or the 10-day low (Turtle exit).
Fails when: ADX is low — most “breakouts” in low-ADX environments are fakeouts that revert in 1–2 days. The ADX filter is what separates this from naive breakout chasing.

7. Moving Average Crossover + Volume — classic with a guard

The textbook “20 crosses 50” with one critical addition.

Signal: 20 EMA crosses above 50 EMA.
Confirmation: the crossover candle (or next one) closes with volume > 1.5× the 20-day average volume.
Stop: below the 50 EMA.
Fails when: crossover happens on light volume — these are the textbook fakeouts everyone warns about. Without the volume filter this combo is barely better than random.

Combos to avoid

These show up everywhere on YouTube and they’re traps:

RSI + Stochastic + CCI — three momentum indicators saying the same thing. Looks confluent, isn’t.
5 EMAs of different periods — it’s still one indicator. You’ll just pattern-match the spaghetti.
MACD + RSI + Stochastic + Bollinger + Ichimoku — analysis paralysis. By the time everything aligns, the move is over.
Any combo without a regime filter — every momentum signal needs trend context, every mean-reversion signal needs a “is this stock actually ranging?” check.

How to actually deploy a combo

Pick one combo. Just one. From this chapter.
Define the rules in writing — exact entry, exact stop, exact exit. If you can’t write it as code, you can’t execute it.
Backtest by hand on 50 historical setups (Chapter Backtesting covers this properly later). Count wins/losses and average R:R.
Paper trade for 1 month. No real money.
Then trade ¼-size for another month. Only scale up if the live numbers match the backtest.

If you skip steps 3–5, you’re not using a strategy — you’re using vibes with extra steps.

Common mistakes

Curve fitting — tweaking parameters until past data looks great. The market doesn’t care about your optimised RSI period of 11.7.
Ignoring market regime — a momentum combo in a ranging market loses every time, and vice versa. Always check: trending or ranging? (ADX > 25 = trending.)
Adding indicators after losses — “if I add MACD too, this loss wouldn’t have happened.” It also wouldn’t have caught your last 6 wins. Don’t reverse-engineer indicators onto losing trades.
Trading every signal — even a great combo signals 3× more often than you should trade. Filter by trend, by sector strength, by your own daily loss limit.

TL;DR

Confluence = indicators from different families agreeing.
Two or three indicators. Never five.
Each combo has a market regime where it works and one where it dies — know both.
A combo without an exit rule and a stop-loss isn’t a strategy.