ICT Fair Value Gap FVG Honest Stats MT5
- Indicatori
- Versione: 1.0
Most FVG indicators draw boxes. This one also checks whether the boxes work.
ICT Fair Value Gap FVG Honest Stats draws every fair value gap on your chart and measures, on that chart's own history, how often price came back to a gap and how often the gap held, next to fake boxes of the same size. Free, part of the Honest Stats series.
MT4 version | ICT Killzones Honest Stats (free) | ICT Liquidity Levels Honest Stats (free) | ICT SMT Divergence Honest Stats (free) | TrueRisk Trade Panel (free) | TrueRisk Account Guard (daily loss limit)
What it draws
- Bullish gap: the low of a candle is above the high of the candle two bars before it. The empty space between them is the gap
- Bearish gap: the same, upside down
- Open gaps reach to the right edge of the chart. A gap is filled when price trades through its far side; filled gaps turn grey (or can be hidden)
- Gaps smaller than 0.1 x ATR(14) are ignored, so tiny gaps don't clutter the chart (ATR = the average size of one candle)
The honest stats table
The table uses every closed bar loaded on the chart.
- Gaps found: bullish and bearish
- Price came back to the gap: how often price returned to the gap within 100 bars
- Real gap held on that return: when price first came back, did it turn and move 0.5 ATR away from the gap before it went 0.5 ATR through the other side?
- Fake box of the same size held: the same test for three fake boxes per gap, each placed right under (bullish) or over (bearish) a random earlier candle. Same size, but no gap behind them
- Difference: real minus fake, in percentage points
Under the table you see whether the difference is bigger than what luck alone could produce (95% range). Only a difference bigger than that means something.
Why the fake boxes matter
Price moves up and down all the time, so almost any box under the price "holds" fairly often. On our charts real gaps and fake boxes both held about 60% of the time. A 60% hold rate sounds good, but when a random box does the same, the gap is not what made price turn.
What we found
- Gold, 1 hour, September 2011 to October 2026 (52,452 bars): bullish gaps held 61% of the time, fake boxes 60%; bearish gaps 60%, fake boxes 58%. Both differences are within luck: no edge
- EURUSD, 15 minutes, November 2022 to October 2026 (98,312 bars): bullish gaps 58% against 58%; bearish gaps held slightly less often than fake boxes (60% against 61%)
- Our research test on 18 years of gold (Dukascopy, 2008 to 2026) gave the same picture. On Bitcoin, bullish gaps held 3 to 5 points more often than fake boxes on the 1-hour and 4-hour charts, more than luck explains; bearish gaps did not
The result depends on the market. Check yours before you trade gaps.
How to use it
- Works on any symbol and any timeframe
- More history on the chart gives more reliable numbers (Tools > Options > Charts > Max bars in chart)
- The table counts closed bars only, so it doesn't jump with every tick
- Hover over a box to see whether it is bullish or bearish and whether it is filled
Limitations
- The stats only cover the history loaded on the chart
- The test uses bar highs and lows. Inside one bar the order of moves is unknown, so a bar that both turns and breaks is left out
- Past behaviour does not guarantee future behaviour. This is a measuring tool, not a trading signal
Inputs
- Smallest gap, in ATR(14): default 0.1 (0 = every gap)
- Draw gaps from the last N bars: default 300
- Also draw gaps that are already filled: default on
- Colours of bullish, bearish and filled gaps; fill the boxes
- Turn or break distance, in ATR(14): default 0.5
- Bars to wait for price to come back: default 100
- Bars to decide held or broke: default 50
- Show the honest stats table, table corner, font size
Support: if something doesn't work as expected, send me a private message first. I answer every message and fix problems quickly.
