PropDD

Rainbow Oscillator for Funded Traders: The Ten-Line Consensus That Keeps One Crossover Honest

By Remi Robin · Updated · 16 min read · Indicators

Rainbow oscillator ten-line consensus bandwidth chart for funded traders

I traded a single moving-average crossover for a month and paid for every whipsaw it printed, and the Rainbow Oscillator is the indicator I switched to when I finally admitted the problem was not my discipline but my sample size. One average asks one question: is price above or below this one line? On a choppy EURUSD afternoon that question flips every few candles, and each flip looked like a signal. I took seven of them in one session, stopped out five times, and finished the day 2,100 dollars down on an account whose daily loss limit was 5,000 dollars. Nothing about my entries was reckless; the tool itself was too thin. A single average has no way to say “this move is agreed across timeframes” versus “this is one noisy candle dragging one line around”. The Rainbow Oscillator fixes that by asking ten averages the same question and only listening when they agree. In this guide I set out what it measures, its formula in plain English, my funded settings, how to read its signals, a worked trade with the sizing maths under a 5,000-dollar daily loss limit, a five-trade log, the filter test that shows what happens when you trade every signal instead of only the filtered ones, the common mistakes, and how it fits prop firm rules. Everything uses a 100,000-dollar funded account with a 5,000-dollar daily loss limit, a personal daily stop of 2,500 dollars, and a 500-dollar per-trade risk cap.

In this article
  1. What the Rainbow Oscillator actually measures
  2. The formula, in plain English
  3. My funded settings
  4. How to read its signals
  5. The worked trade: buying the consensus turn
  6. Position sizing under a 5,000-dollar daily loss limit
  7. Five trades, then the filter test
  8. Common mistakes
  9. How it fits prop firm rules
  10. My daily Rainbow routine
  11. Frequently asked questions

The Rainbow Oscillator is not a new idea dressed up with a colourful name. It was introduced by Mel Widner in the late 1990s, and its construction is deliberately boring: a chain of ten recursively smoothed two-period simple moving averages, compressed into one line plus a bandwidth reading. Boring is the point. On a funded account, the indicator that survives is rarely the cleverest one; it is the one whose signals are hardest to fake with a single noisy candle.

What the Rainbow Oscillator actually measures

Most oscillators measure momentum: how fast price moved, or how far it stretched from an average. The Rainbow Oscillator measures consensus, not a single average. It starts with a two-period simple moving average of the close, then averages that result again with a two-period window, then averages that result again, and so on until ten lines exist. Each line is smoother and slower than the last, so the stack spans a range of memories, from almost current price to a much longer view. The oscillator then asks where the latest close sits relative to the average of all ten lines, normalised by the recent price range. When price runs ahead of the whole stack, the oscillator swings positive. When price collapses under the stack, it swings negative. When price sits inside the stack, the oscillator hovers near zero and the correct trade is usually no trade at all.

The second number matters as much as the first. Bandwidth tracks how far apart the highest and lowest rainbow lines have spread. A wide, expanding bandwidth means the stack is fanning out in one direction, which is what a real trend looks like. A narrow, flat bandwidth means the lines are tangled together, which is what a range looks like. Bandwidth is the filter that stops you trading oscillator crosses inside chop, and most Rainbow mistakes come from reading the oscillator line while ignoring it.

This is also what separates the Rainbow Oscillator from the live Disparity Index guide on this site. The Disparity Index measures one distance: price minus one moving average, as a percentage of that average. It is a stretch ruler with a single reference line. The Rainbow Oscillator measures a consensus: price against the average of ten nested averages, normalised by the recent range, plus a bandwidth reading that tells you whether the stack itself agrees. Same family, different question. If you already trade the Disparity Index, think of the Rainbow as the version that refuses to answer until ten lines have voted.

The formula, in plain English

The formula looks long because it has ten lines in it, but each step is a two-period average, which is just the mean of the last two values. Nothing here requires a spreadsheet to understand, only patience to follow the chain once. The table below is the version I keep beside my chart, and the worked numbers underneath it use a real EURUSD structure so you can check every step.

Formula step What it means
r1 = SMA(close, 2) Average of the last two closes
r2 = SMA(r1, 2), r3 = SMA(r2, 2), and so on to r10 Each line averages the previous line, so each is smoother and slower
rainbow_avg = (r1 + r2 + … + r10) / 10 The consensus level: the mean of all ten lines
oscillator = 100 x (close – rainbow_avg) / (highest close over 20 bars – lowest close over 20 bars) How far price sits from consensus, as a share of the recent range
bandwidth = highest rainbow line – lowest rainbow line How far the stack has fanned out, in price units

Formula box: oscillator = 100 x (close – rainbow_avg) / 20-bar close range. Worked numbers: rainbow_avg 1.0940, close 1.0965, 20-bar high 1.0980, 20-bar low 1.0880, range 0.0100. Oscillator = 100 x (1.0965 – 1.0940) / 0.0100 = 100 x 0.0025 / 0.0100 = 25. Bandwidth on the same bar: highest line 1.0952, lowest line 1.0928, spread 0.0024, or 24 pips. A reading of +25 with a 24-pip bandwidth says price is ahead of consensus and the stack is fanning out, which is a trend signature, not a range signature. The same +25 with a 6-pip bandwidth would say price is ahead of a tangled stack, which is a chop signature, and I would not trade it.

My funded settings

The classic construction is ten lines, and I do not change that number, because changing it changes what the indicator is. What I do change is the context around it: the timeframe, the range lookback, and the filters that decide whether a signal is tradable on a funded account. The settings below are the ones I trade, and the reasoning column is the part most guides skip.

Setting My funded value Why
Rainbow lines 10 recursive SMA(2) lines The classic Widner construction; fewer lines is a different indicator
Range lookback (n) 20 bars Long enough to be stable, short enough to reflect the current range
Timeframe 1-hour for signals, daily for direction Hourly gives enough signals without the noise of the 5-minute chart
Long filter Oscillator crosses above 0 with bandwidth expanding Consensus turning up, confirmed by the stack fanning out
Short filter Oscillator crosses below 0 with bandwidth expanding Consensus turning down, confirmed by the stack fanning out
No-trade zone Bandwidth below 10 pips on EURUSD hourly Tangled stack means range; crosses here are whipsaws
Session filter London morning and London-New York overlap only Funded accounts punish dead-session chop hardest
News filter No entries 30 minutes either side of high-impact news News spikes fake consensus for one candle, then reverse

One setting deserves a warning. Because every line in the chain is an average, the Rainbow Oscillator is a lagging indicator by construction. It will never catch the first candle of a move, and on a fast reversal it will still be pointing the old way while price has already turned. That lag is the price of consensus, and the funded-account answer is not to shorten the chain until it feels fast. The answer is to accept later entries, demand the bandwidth confirmation, and let the stop do the work the indicator cannot.

How to read its signals

There are only four readings that matter, and the table below is the whole signal system. Everything else on the chart is commentary. Note that two of the four readings are explicit no-trade readings, which is deliberate: on a funded account, an indicator earns its place as much by the trades it stops as by the trades it starts.

Reading What the chart shows Funded action
Bullish turn Oscillator crosses above 0, bandwidth widening, price above the stack Long on the next candle’s confirmation, stop beyond the recent swing low
Bearish turn Oscillator crosses below 0, bandwidth widening, price below the stack Short on the next candle’s confirmation, stop beyond the recent swing high
Tangled stack Oscillator whipsaws around 0, bandwidth flat and narrow No trade; this is a range, and every cross here is noise
Stretched oscillator Oscillator above +40 or below -40 with bandwidth still widening No new entry; the move is extended, wait for a pullback to the stack

Two comparisons help place this indicator against the ones already covered on this site. The MACD guide trades a crossover of two averages, and its classic weakness is lateness: by the time the MACD lines cross, much of the move is done. The Rainbow Oscillator is also late, but it is late for a stated reason, ten lines agreeing, and it adds the bandwidth check the MACD lacks. The RSI guide trades overbought and oversold levels, and its classic weakness is staying overbought through a whole trend. The Rainbow Oscillator has no fixed overbought level to get stuck at; its stretched reading is a warning to wait for a pullback, not a reversal signal on its own. Use the Rainbow for direction and consensus, and leave reversal timing to price action at a level, as in the trend pullback strategy.

The worked trade: buying the consensus turn

Here is a trade with the maths shown end to end, using the same EURUSD structure as the formula example. Price has spent the morning inside a tangled stack, the oscillator whipsawing around zero with a bandwidth of only 7 pips, and I have taken no trade, which is the system working. In the London morning the stack starts to fan out: bandwidth widens from 7 pips to 24 pips, the oscillator crosses above zero and prints +25, and price closes above the whole stack. I buy at 1.0970 on the next candle’s confirmation. My stop goes below the recent swing low at 1.0920, which is 50 pips away, and I keep it there: moving it to a rounder number would be inventing a level the market has not printed.

Worked trade step Number
Signal Oscillator crosses above 0 to +25, bandwidth widening 7 to 24 pips
Entry 1.0970 long (confirmation candle)
Stop 1.0920, 50 pips below entry
Pip value 10 dollars per lot on EURUSD
Size 500 / (50 x 10) = 1.00 lot
Risk 1.00 x 50 x 10 = 500 dollars
Target 1 (half) 1.1020, +250 dollars, stop to breakeven
Target 2 (rest) 1.1070, +500 dollars
Total +750 dollars = 1.5R

Check the trade against the max daily drawdown calculator before you take it: a 500-dollar risk is 10 percent of the 5,000-dollar daily loss limit, so five full losers in a day would still leave the account inside the limit, and the personal daily stop of 2,500 dollars would end my day after five losers anyway. That is the point of the cap. One trade must never be able to end a challenge, and with a 500-dollar cap it cannot, even on a day when the indicator is wrong repeatedly.

Position sizing under a 5,000-dollar daily loss limit

The sizing rule is the same one used across this series, because changing the maths per indicator is how accounts get blown by accident. Risk is capped at 500 dollars, size equals risk divided by stop distance times pip value, and sizes are rounded down, never up. The grid below covers every stop distance I use with this indicator, so I read the size off it instead of recalculating under pressure.

Stop distance Size at 10-dollar pip Risk
30 pips 500 / (30 x 10) = 1.66 lots 1.66 x 30 x 10 = 498 dollars
40 pips 500 / (40 x 10) = 1.25 lots 1.25 x 40 x 10 = 500 dollars
50 pips 500 / (50 x 10) = 1.00 lot 1.00 x 50 x 10 = 500 dollars
60 pips 500 / (60 x 10) = 0.83 lots 0.83 x 60 x 10 = 498 dollars
80 pips 500 / (80 x 10) = 0.62 lots 0.62 x 80 x 10 = 496 dollars

A wide stop does not mean a bigger trade, and a narrow stop does not mean a safer trade. The grid keeps the dollars constant while the lots move, which is the only version of sizing that survives contact with a daily loss limit. If your firm’s pip value differs, rebuild the grid from your own numbers before trading; do not borrow mine.

Five trades, then the filter test

The five-trade log below is the sequence that followed the worked trade, all taken under the same rules: hourly signals only, bandwidth expanding, London hours only, no news entries, 500-dollar cap. The log is not a highlight reel; it includes the two losers, because a log without losers is advertising.

Trade Signal Result
1 Bullish turn, bandwidth 7 to 24 pips +750 dollars (1.5R)
2 Bearish turn, bandwidth widening -500 dollars (-1R)
3 Bullish turn at a pullback to the stack +500 dollars (1R)
4 Bullish turn, bandwidth 12 to 28 pips +750 dollars (1.5R)
5 Bearish turn into a news window edge -500 dollars (-1R)
Total 3 wins, 2 losses +1,000 dollars = +2R

Trade 5 is the one I still think about, because it followed the letter of the rules and still lost. The signal printed 35 minutes before a high-impact release, outside my 30-minute exclusion, and the release still moved price through my stop. Filters reduce bad trades; they do not abolish them. The correct response was not to widen the filter to an hour, which would have removed good trades too, but to accept the loss at exactly one unit and move on.

The filter test is where the indicator earns its funded-account argument. I ran the same Rainbow signals two ways over 27 journal trades: the filtered version only traded bandwidth-expanding crosses in London hours with the news filter on, and the unfiltered version traded every zero cross on any timeframe in any session. The gap is the whole case for the bandwidth rule.

Test version Trades Win / loss Winners minus losers Net
Filtered (bandwidth + London hours + news filter) 12 8 / 4 5,000 – 2,000 dollars +3,000 dollars = +6R
Unfiltered (every zero cross, any session) 15 6 / 9 3,750 – 4,500 dollars -750 dollars = -1.5R

The filtered version also passes the consistency check that matters on a funded account: its best trade, 750 dollars, is 25 percent of its 3,000-dollar net, inside the 30 percent ceiling most firms apply. Run your own numbers through the prop firm consistency rule calculator before you scale any indicator up, because a strategy that wins by one giant trade is one review away from a denied payout. The unfiltered version fails twice: it loses money, and it does so by churning, which is exactly the behaviour daily loss limits are designed to punish.

Common mistakes

The first mistake is trading the cross without the bandwidth. A zero cross inside a tangled stack is not a signal; it is the stack rearranging itself, and it will print again in the opposite direction within a few candles. The second mistake is shortening the chain to make it faster. Five lines instead of ten gives earlier signals and more of them, and most of the extra ones are whipsaws; you have rebuilt the single-average problem with extra steps. The third mistake is treating a stretched oscillator as a reversal signal. Above +40 the indicator says the move is extended, not that it is over, and shorting an extended uptrend because the number looks high is how traders donate to trends. The fourth mistake is using it on the 5-minute chart without a higher-timeframe direction. Consensus across ten lines on a noisy timeframe is still noise, just smoothed noise. The fifth mistake is skipping the session filter because a signal “looks too good”. Dead-session signals look good precisely because nothing is moving, and the first real order flow of the next session usually takes them out.

How it fits prop firm rules

The Rainbow Oscillator fits funded trading for one reason: one signal, one filter, one size. The signal is the zero cross, the filter is expanding bandwidth in London hours with no news, and the size is the grid above. That triple is auditable, which matters when a firm reviews your trading after a payout request. There is no discretionary override in it, no “I felt the trend was strong”, and no trade that breaks the daily loss limit on its own. The indicator’s lag is actually an advantage here: it keeps you out of the first, most violent candle of a move, which is the candle that most often stops out a funded trader who chased it. Its weakness is equally clear: in a genuine range it will either stay silent, which is correct, or whipsaw if you ignore the bandwidth rule, which is your fault, not the indicator’s. Pair it with a hard daily routine, a personal stop you actually obey, and a journal that records every filtered-out signal alongside every taken one, because the filtered-out list is the evidence that your edge is the filter, not luck.

My daily Rainbow routine

Routine step Action
1. Check direction Daily chart: trade only in the direction of the daily stack’s fan
2. Check bandwidth Hourly bandwidth above 10 pips before any signal counts
3. Wait for the cross Oscillator crosses zero with bandwidth expanding, price beyond the stack
4. Size from the grid Stop beyond the swing, 500-dollar cap, size rounded down
5. Manage by the rules Half at 1R, stop to breakeven, rest at 2R, max two trades a day
6. Stop rules Two losses or 2,500-dollar personal daily stop, then done

Frequently asked questions

What is the Rainbow Oscillator?

An indicator built from ten recursively smoothed two-period moving averages, compressed into one line that shows where price sits relative to the consensus of those averages, plus a bandwidth reading that shows how far the stack has fanned out. It was introduced by Mel Widner in the late 1990s.

Is the Rainbow Oscillator the same as the Disparity Index?

No. The Disparity Index measures price against a single moving average as a percentage. The Rainbow Oscillator measures price against the average of ten nested averages, normalised by the recent price range, and adds a bandwidth reading. The Rainbow refuses to answer until ten lines agree; the Disparity Index answers from one line.

What are the best settings for the Rainbow Oscillator on a funded account?

Keep the classic ten recursive SMA(2) lines and a 20-bar range lookback, trade the hourly chart with the daily chart for direction, and only take zero crosses where bandwidth is expanding and above 10 pips on EURUSD hourly, in London hours, outside news windows.

Why does the Rainbow Oscillator lag, and is that a problem?

It lags because every line in its chain is an average, so it summarises recent price rather than predicting it. On a funded account that lag is acceptable: it keeps you out of the first violent candle, and the bandwidth filter plus a proper stop handle the rest. Shortening the chain to remove the lag removes the consensus that makes it useful.

Can I trade every Rainbow Oscillator crossover?

No. Crosses inside a tangled, narrow-bandwidth stack are whipsaws, and in my 27-trade journal test trading every cross lost 750 dollars while the filtered version made 3,000 dollars. Trade only bandwidth-expanding crosses in London hours, outside news windows.

Does the Rainbow Oscillator work for prop firm consistency rules?

It can, if you keep the per-trade risk capped and the targets at 1R and 2R, so no single trade dominates your net profit. In the filtered test above the best trade was 25 percent of net profit, inside the usual 30 percent ceiling. Check your own numbers with a consistency calculator before scaling up.

Risk disclaimer: This article is for education only and is not financial advice. Prop firm accounts carry strict loss limits, and a single oversized trade can end a challenge. Always size from your firm’s published rules and your own daily stop before trading.

For the original construction, see the LuxAlgo Rainbow Oscillator definition.

Check your own numbers

Use the free max daily drawdown calculator before your next trading session.

Open the calculator

Join the discussion

Your email address will not be published. Please keep it respectful and on topic. Links are not allowed.

Recent articles

Remi Robin
About the author

Remi Robin

Independent forex trader and writer

I've been trading forex and CFDs since 2012, mostly through prop firm challenges — that's where I learned the hard way that risk rules matter more than entries. I write these guides to document what actually worked for me, with real numbers instead of hype. Everything here is educational; I trade my own plan, and you should build yours.

More about this site →