Why mean reversion signals fail after small volatility spikes

The problem with fading every band touch

You see price close outside the lower Bollinger band and you take the long, expecting mean reversion.

The band touch looks identical to the one that worked yesterday.

But this time price keeps falling, because the volatility regime underneath was ordinary, no real spike, no exhaustion, nothing extreme enough to justify the snapback.

Most mean-reversion entries fail because the trigger does not measure whether the volatility push itself was abnormal.

What a snapback actually requires

A snapback is not just a band breach and a return.

It is the relief move after an extreme volatility event exhausts and the market reclaims controlled range.

That means three pieces have to align at the same closed candle:

• A synthetic volatility score above its 85th percentile, built from ATR, standard deviation and Bollinger bandwidth, ranked against recent history
• A breach of an adaptive band, then a reclaim back inside within 6 bars
• A momentum turn confirmed by RSI(7) crossing back through 35 for longs or 65 for shorts

Only when the volatility extreme, the reclaim and the RSI turn agree does a signal print.

The tool that enforces this is Synthetic Volatility Snapback for MetaTrader 5.

How the synthetic volatility measure works

Most volatility tools report a number but leave you to decide whether it is high enough to matter.

This one builds a chart-local percentile rank from ATR, standard deviation and Bollinger bandwidth, then uses the 85th and 90th percentile thresholds as fade triggers.

The logic: a snapback is only worth taking after a genuine spike relative to the instrument's own recent behaviour.

No external VIX, no benchmark index, the score is derived entirely from the price action on your chart.

The dashboard reports the current regime score against the 70/85/90 percentile levels.

Adaptive bands and a centerline show the mean-reversion lane the snapback is targeting.

The mistake this avoids is treating every band touch as equal, the percentile score has to confirm the volatility event was extreme first.

Reading a signal: spike, breach, reclaim

Signals are confirmed on closed candles.

A BUY arrow prints when:

• The synthetic volatility score is above its 85th percentile
• Price closed below the lower adaptive band, then back above it within 6 bars
• RSI(7) crosses above 35 and the candle closes above the previous candle high
• Band width, EMA50 slope, spread and warm-up filters all allow the bar

The strong tier tightens the rule: the score must have been above the 90th percentile, and the trigger candle range must be at least 0.8 ATR so there is room left to reach the mean.

A SELL is the mirror: a score above the 85th percentile, a close above the upper band then back inside, RSI(7) crossing below 65 and a close below the prior candle low.

What appears on the chart:

• The dashboard volatility-regime score, adaptive bands, centerline, a regime ribbon and exhaustion markers
• A BUY or SELL arrow with entry at the snapback close
• A stop beyond the exhaustion candle extreme plus 0.2 ATR
• TP1 at the adaptive mean and TP2 at the opposite half-band, with a volatility-cooldown trail after TP1
• Alerts for volatility spike armed, regular and strong snapbacks, cooldown exit, TP touches and opposite snapback

Every threshold is editable in the Inputs tab.

When to use it, and when not to

Use Synthetic Volatility Snapback if you fade exhaustion after volatility spikes and you want the entry, stop and targets attached to the signal.

It fits gold M5-M15, indices M1-M15 and crypto M15-H1.

It fits you less well if you chase breakouts or trade one-way news moves, mean-reversion snapbacks fail on trend days when the market does not return to the mean.

The synthetic score is price-derived, not a real options VIX.

Percentile thresholds need enough closed bars to stabilise, so the first signals on a fresh chart may be unreliable until the lookback window fills.

What it does not do

Synthetic Volatility Snapback does not predict the size of the snapback move or guarantee that the mean will be reached.

It does not work on one-way trend days where exhaustion is brief and the dominant force keeps pushing price in the original direction.

It is not a real volatility index, the score is a chart-local percentile rank, not an options-implied measure like VIX.

It does not tell you whether the larger trend has changed, only that a short-term volatility extreme has reclaimed range.

It does not filter fundamental news events, so a headline can invalidate the setup after the signal prints.

It should not be used on instruments with wide spreads or low liquidity where the 0.2-ATR stop buffer is routinely exceeded by normal spread variation.

Pairing with other tools

Synthetic VolatilitySnapback is a mean-reversion signal tool.

If you also trade breakouts after a squeeze, Squeeze Divergence Radar identifies compression zones and divergence setups on the same MT5 platform.

If you fade traps at round numbers, Big Figure Trap Signals marks failed breakouts at psychological levels and can run alongside snapback setups on different timeframes.

Do not pair this with another mean-reversion tool that fires on the same band-touch logic, you will see duplicate signals, not independent confirmation.

Installing and running it

Copy the .ex5 file into the MQL5/Indicators folder inside your MetaTrader 5 data folder.

Refresh the Navigator panel, then drag Synthetic Volatility Snapback onto the chart.

Every threshold, percentile levels, RSI crosses, ATR multipliers, reclaim window, is editable in the Inputs tab.

The dashboard and adaptive bands appear immediately; signals print on closed candles once the warm-up period completes.

---

Trading involves risk of loss. This indicator is provided for educational purposes only and does not constitute financial advice. Past performance does not guarantee future results.