What is confluence in trading?

A level is a price where something happened or where a calculation lands: an old high, a Fibonacci ratio, an average. Confluence is when two or more of those land in the same narrow zone. Narrow needs a number. A workable rule is that all the factors sit within about half an average true range of each other on your chart; spread wider than that, you have several separate levels, not one zone.

The reason it helps is plain. Each factor has its own group of traders watching it: swing traders the weekly level, retracement traders the Fibonacci line, trend followers the 50-day average. When those overlap, more orders are likely to meet at one price. Nothing more mysterious than that, and nothing that makes a zone hold.

Independent vs redundant factors

Confluence only counts when the factors come from different information. Two tools built from the same input are one factor wearing two hats. The families that are independent of each other are structure (prior highs, lows and support and resistance), swing math (Fibonacci levels, harmonic zones), averages (moving averages, VWAP), volume location (volume profile nodes) and round numbers.

If you use UpTickFlow, our invite-only TradingView indicator, its map already does part of this merging: levels from several timeframes at one price are drawn once and tagged with all of them, for example W·D for a weekly and a daily level. Read that tag as one structural factor that two timeframes agree on, not as two points.

  • RSI, stochastic and MACD turning up together are one factor, momentum, all calculated from recent closes. They are also triggers or warnings, not locations.
  • A 20 EMA and a 21 EMA, or a 50-day SMA and a 50-day EMA, are one moving-average factor.
  • The 61.8% retracement and a harmonic PRZ built from the same swing share their math; count them once. A PRZ from a different swing can count separately; see harmonic PRZ trading.
  • Daily and weekly levels at the same price are one structural factor with two timeframes behind it: stronger than either alone, but still one family.

Turn your phone sideways to read the image. Tap to close.

Two cards. Left, redundant: RSI, Stoch and MACD chips held by one bracket and counted as 1, with the note one factor: momentum. Right, independent: a weekly level, 61.8% and the 50-day MA, each with its own bracket and counted 1, with the note three factors.

A simple confluence score

Give each family one point at most, and write the checklist down before you look at the chart, so you do not keep adding tools until a zone qualifies. Then read the score: 0 or 1 is a line, not a zone, so skip it; 2 is worth watching; 3 or more is worth a written plan, still only if a trigger appears.

  • A higher-timeframe level: a daily or weekly high, low or old breakout level, as in multi-timeframe analysis.
  • A Fibonacci level: the 38.2%, 50% or 61.8% retracement of the last clear swing; see Fibonacci retracement.
  • A moving average or VWAP: a rising 20- or 50-period average in an uptrend, or session VWAP for day trades.
  • A harmonic PRZ, if one is complete and built from a different swing.
  • A round number: $45.00, $50.00, $100.00.
  • A volume profile node: a high-volume node or the point of control of the prior range; see volume profile.

Turn your phone sideways to read the image. Tap to close.

Left, a price ladder for the worked example: the 50-day MA at 45.40, a weekly level at 45.20, the 61.8% retracement at 45.06 and the round number 45.00, shaded as one zone 0.40 wide, or 0.33 of the 1.20 ATR. Right, a scorecard: higher-timeframe level 1, Fibonacci level 1, moving average or VWAP 1, harmonic PRZ 0, round number 1, volume profile node 0, score 4 / 6, and the rule 0–1 skip, 2 watch, 3+ plan. A note under the ladder reads RSI oversold: a trigger, not a location.

Confluence + trigger + risk

Confluence answers where. A trade needs the whole sequence, in this order:

  1. Location: a zone that scores 3 or more, in the direction of the higher-timeframe trend.
  2. Trigger: a closed candle that shows the reaction, such as a hammer or an engulfing bar back out of the zone, or a change of character on a lower timeframe. Price touching the zone is not a trigger.
  3. Risk: the stop beyond the far edge of the whole zone plus a buffer, not beyond the first line in it; the target at the next opposing level; the size from the stop.
  4. Room: if the next level is less than about twice the risk away, skip it, however high the score; see the risk-reward ratio.

Worked example: a long from a four-factor zone (hypothetical)

Hypothetical, on the daily chart of XYZ in an uptrend. The last swing ran from $42.00 to $50.00, so the 61.8% retracement sits at $45.06 ($50.00 − 0.618 × $8.00). An old breakout high from weeks ago, now a weekly support, is at $45.20, the rising 50-day moving average is at $45.40, and $45.00 is a round number. Four factors within $0.40, about a third of the $1.20 daily ATR: a score of 4.

Price pulls back into the zone, trades as low as $45.10 and closes at $45.70, a hammer back out of the zone. That close is the trigger, and you buy at $45.70.

The failure: the same zone on another day, and the next morning the stock gaps down to $44.20, below the stop, so the order fills at the open, $1.50 below the entry instead of $1.30. The loss is 192 × $1.50 = $288, about −1.15R. Four factors made the zone worth watching, not safe.

  1. Entry: $45.70, on the close of the hammer.
  2. Stop: $44.40, under the whole zone ($45.00) minus half an ATR ($0.60). Risk: $1.30 a share.
  3. Target: $49.60, just under the $50.00 swing high. Reward: $3.90, or 3R.
  4. Size: a $50,000 account risking 0.5% accepts a $250 loss. $250 ÷ $1.30 = 192 shares, rounded down, about $8,774 of stock.
  5. Result: at the target, 192 × $3.90 = $748.80, or +3R; at the stop, 192 × $1.30 = $249.60, or −1R, plus costs.

Why too much confluence causes paralysis

More lines do not make better zones. Draw enough tools and every price is next to something. Hypothetical: twelve lines in a $10.00 range, each with a $0.20 band on either side, cover up to $4.80 of it, almost half. A checklist that can call any price a zone filters nothing.

The opposite failure is waiting for perfect zones that score 5 or 6. They are rare, so you trade so seldom that a few results dominate your record, or you start lowering the bar in the middle of the session. Fix the checklist at five or six families, fix the threshold, and let your trading plan decide, not the chart in front of you.

Turn your phone sideways to read the image. Tap to close.

Two panels with the same zigzag price path. Left: twelve grey horizontal lines, each with a shaded band, with the sum 12 × 0.40 = 4.80 / 10.00 and the note every price is near a line. Right: three colored lines meeting in one shaded zone, with the note three lines that agree.