I lost 400 dollars on a Tuesday because my moving average lied to me in the politest possible way. GBPUSD chopped sideways all morning, my 21 EMA kept crossing price every hour, and every cross looked like a trend starting. I took three of them. All three reversed within twenty pips, and by lunch my daily loss was nearly used up on trades that were never trends at all.
In this article
That evening I went looking for an average that behaves differently in chop than it does in a trend, and I found one built by Tushar Chande: VIDYA, the Variable Index Dynamic Average. It does not use a fixed smoothing speed. It asks the market how strongly it is trending, and then it adjusts its own speed to match the answer. In a strong trend it tracks price almost like a normal EMA. In chop it barely moves at all, which is exactly what I needed on that Tuesday.
This guide is the full VIDYA setup I now use on a funded account: what it measures, the formula worked with real numbers, the settings I use, the four signals I trade, one fully sized trade, and the ten-trade test that convinced me to keep it.
What VIDYA actually measures
Most moving averages smooth price with one fixed speed, whatever the market is doing. VIDYA changes its speed bar by bar, using a second indicator called the Chande Momentum Oscillator, or CMO, as its speed control. When the CMO says the market is trending hard, VIDYA speeds up and hugs price. When the CMO says the market is chopping, VIDYA slows to a crawl and stops printing fake crosses.
That makes VIDYA a trend filter first and a signal line second. I do not use it to predict reversals or to catch exact tops and bottoms. I use it to answer one funded-account question: is this move strong enough to be worth risking part of my daily loss limit on, or is it just noise that will eat the limit for nothing?
Two distinctions matter before you use it, because the site already covers both cousins. VIDYA is not the Kaufman Adaptive Moving Average: KAMA changes speed using an efficiency ratio built from net price change over total movement, while VIDYA changes speed using the CMO, which compares up-day and down-day momentum over a lookback. And VIDYA is not the Chande Momentum Oscillator itself: the CMO is the speedometer, VIDYA is the average that reads it. My full CMO guide and my KAMA guide are linked below if you want the family tree.
The VIDYA formula, worked with real numbers
VIDYA needs three ingredients: a CMO value, a base smoothing factor, and yesterday’s VIDYA value. The formula looks technical, but every step is simple arithmetic.
| Step | What it means | My worked numbers |
|---|---|---|
| 1. CMO over 9 bars | 100 x (up moves minus down moves) divided by (up moves plus down moves) | See the nine-bar example below |
| 2. Base factor F | 2 divided by (CMO period plus 1) | 2 / 10 = 0.20 |
| 3. Effective speed k | F x absolute CMO, as a decimal | 0.20 x 0.4545 = 0.0909 |
| 4. New VIDYA | Close x k, plus previous VIDYA x (1 minus k) | 1.2809 (worked below) |
Here is the nine-bar example, using GBPUSD daily moves in pips: up 40, down 20, up 30, up 25, down 15, up 35, up 10, down 25, up 20. The up moves add to 160 pips and the down moves add to 60 pips, so the CMO is 100 x (160 minus 60) divided by (160 plus 60), which is 100 x 100 / 220 = 45.5. The market is trending, but only moderately, so VIDYA should move at less than half its base speed.
With a previous VIDYA of 1.2800 and a close of 1.2900, the new VIDYA is 1.2900 x 0.0909 plus 1.2800 x 0.9091, which is 0.1173 plus 1.1636 = 1.2809. Compare that with a standard 9-period EMA on the same bar: 1.2900 x 0.20 plus 1.2800 x 0.80 = 1.2820. The EMA jumps 20 pips toward price; VIDYA moves only 9 pips, because the CMO told it the trend was only moderate.
Now watch the speed change. In a strong trend with a CMO of 90, the effective speed becomes 0.20 x 0.90 = 0.18, and the new VIDYA is 1.2818, nearly matching the EMA. In dead chop with a CMO of 10, the speed collapses to 0.02 and the new VIDYA is 1.2802, barely a flicker. One line, three speeds, driven by the market itself. That is the whole idea, and it is why VIDYA survives funded-account chop better than a fixed average.
Best settings for a funded account
I use VIDYA on H1 and H4, never below H1, because on M15 the CMO flips too fast and VIDYA starts behaving like the fixed average it is supposed to replace. The settings below are the ones that survived my testing; change the CMO period and you change the personality of the line completely.
| Setting | My funded-account value | Why |
|---|---|---|
| CMO period | 9 bars | Chande’s standard; short enough to react, long enough to ignore one wild bar |
| Base factor F | 0.20 (from 2 / 10) | Matches the 9-period base; do not raise it to chase entries |
| Timeframes | H1 for entries, H4 for direction | H4 VIDYA slope sets the side; H1 VIDYA gives the entry |
| Trend gate | Absolute CMO of 30 or higher | Below 30, VIDYA is telling you the market is not trending |
| Skip zone | Absolute CMO below 20 | No new entries at all; this is the chop that cost me 400 dollars |
| Pair choice | EURUSD and GBPUSD only | Clean CMO readings; exotic pairs whipsaw the oscillator |
One practical note: most platforms do not ship VIDYA as a built-in indicator, so you will add it as a custom indicator or a script. That is a one-time setup cost, and it is worth it, because no built-in average gives you this speed behaviour. Check the script’s formula against the worked example above before you trust it with real size: feed it the same nine moves and confirm it prints a CMO near 45.5. For an independent statement of the formula, see the TradingPedia guide to Chande’s Variable Index Dynamic Average.
How to read its signals
I trade exactly four VIDYA signals, and I ignore everything else the line does. The slope of VIDYA matters more than its level, and the CMO gate matters more than both.
| Signal | What I see | What I do |
|---|---|---|
| Trend long | Price above a rising VIDYA, absolute CMO 30 or higher | Buy pullbacks to VIDYA; stop below the recent swing low |
| Trend short | Price below a falling VIDYA, absolute CMO 30 or higher | Sell rallies to VIDYA; stop above the recent swing high |
| Flat VIDYA | VIDYA nearly horizontal for 5 or more bars, CMO below 20 | Stand aside; no new entries until the CMO wakes up |
| Price crossing a flat VIDYA | Repeated crosses while VIDYA is flat | Ignore every cross; these are the fake signals from my Tuesday |
The rule that saves the most money is the last one. A cross of a flat VIDYA is not a signal at all; it is the indicator telling you it has slowed down because there is no trend to follow. Traders who treat every cross as an entry will take ten chop trades a week. Traders who wait for a rising VIDYA plus a CMO above 30 will take two or three real ones.
A worked trade, sized to a daily loss limit
Here is a real-shaped GBPUSD trade from my journal, with the sizing done the way a funded account demands: risk first, lots second, and never the other way round.
| Trade detail | Value |
|---|---|
| Setup | H4 VIDYA rising, H1 price pulls back to a rising VIDYA, CMO 42 |
| Entry | 1.2850 long on the H1 close back above VIDYA |
| Stop loss | 1.2800, below the pullback low (50 pips) |
| Risk budget | 500 dollars (10 percent of a 5,000 dollar daily loss limit) |
| Position size | 500 / (50 x 10) = 1.00 lot |
| First target | 1.2900, half the position, plus 250 dollars |
| Second target | 1.2950, remaining half, plus 500 dollars |
| Total result | Plus 750 dollars = 1.5R, with 500 dollars risked |
Notice what VIDYA contributed and what it did not. It did not pick the entry price; the pullback low and the 50-pip stop did that. What VIDYA did was keep me out of the three chop trades earlier that week, so the 500-dollar risk budget was still intact when a real trend arrived. On a funded account, the indicator that stops bad trades is worth more than the indicator that finds good ones.
If your daily loss limit is different, scale the same maths through the max daily drawdown calculator before you trade: the stop distance and the risk budget decide the lots, never your feeling about the setup.
The sizing grid and the ten-trade test
VIDYA stops change with volatility, so I keep a sizing grid taped to my monitor. Every row risks the same 500 dollars; only the stop distance and the lots change.
| Stop distance | Lots at 500 dollars risk | Actual risk |
|---|---|---|
| 25 pips | 2.00 | 500 dollars |
| 40 pips | 1.25 | 500 dollars |
| 50 pips | 1.00 | 500 dollars |
| 60 pips | 0.83 | 498 dollars |
| 80 pips | 0.62 | 496 dollars |
Then I tested the four-signal rules over ten filtered trades: 7 winners and 3 losers. The winners were 3 trades at 2R and 4 trades at 1R, for 10R in total, and the losers cost 3R, leaving a net of plus 7R, or plus 3,500 dollars at 500 dollars of risk per trade. The best single trade made 1,000 dollars, which is 28.6 percent of the net profit, inside the 30 percent consistency ceiling most firms enforce. You can read how that ceiling works in my consistency rule guide.
The unfiltered version of the same test is the warning label. Taking every VIDYA cross, including the flat-line crosses, over twenty trades produced 8 winners and 12 losers for a net of minus 4R, or minus 2,000 dollars. Same indicator, same market, opposite result. The difference was not the line; it was the CMO gate and the discipline to skip flat VIDYA entirely.
Common mistakes with VIDYA
| Mistake | What it costs | The fix |
|---|---|---|
| Trading every cross | Chop losses that eat the daily limit | Trade only crosses of a sloping VIDYA with CMO above 30 |
| Confusing VIDYA with KAMA | Wrong expectations about speed changes | VIDYA uses CMO; KAMA uses an efficiency ratio; read my KAMA guide for the contrast |
| Using it on M5 or M15 | CMO noise turns VIDYA into a fixed average | H1 minimum, H4 for direction |
| Ignoring the CMO value | Entries in dead markets that look like trends | No entry when absolute CMO is below 20, ever |
| Skipping the sizing maths | One wide stop breaches the daily limit | Size every trade from the stop distance, as in the grid above |
The KAMA confusion deserves one more line, because the two indicators look identical on a chart. If your adaptive average speeds up when price moves efficiently in one direction with few pullbacks, you are looking at KAMA logic. If it speeds up when up-day momentum dominates down-day momentum over nine bars, you are looking at VIDYA logic. Same family, different engine, and my CMO guide explains the engine in full.
My daily routine with VIDYA
My routine is deliberately boring, because boring is what keeps funded accounts alive. Before London opens, I check the H4 VIDYA slope on EURUSD and GBPUSD and write down the side it allows, long or short, or “no side” if VIDYA is flat. During the session I only take pullbacks to a sloping H1 VIDYA with the CMO above 30, sized from the grid, and I stop after two trades or after one loss, whichever comes first. After the session I log every VIDYA cross I skipped and why, because the skipped trades are the evidence that the filter is working.
That routine pairs naturally with a trend structure: when VIDYA gives me the side, I still want price structure to agree, so I check my trend pullback rules before entry and my EMA guide when I want a second opinion on the trend. VIDYA is the gatekeeper, not the whole strategy, and treating it as the whole strategy is how traders turn a good filter into a bad system.
Frequently asked questions
Is VIDYA better than a normal EMA for prop firm trading?
For funded accounts, yes, in one specific way: it filters chop. A fixed EMA crosses price constantly in sideways markets and each cross tempts an entry that risks the daily limit. VIDYA slows down in chop and produces far fewer fake crosses, which protects the limit. In a clean trend it behaves almost like an EMA, so you give up very little trend performance for that protection.
What CMO period should I use with VIDYA?
Nine bars, which is Chande’s standard and the setting used in every worked example here. A shorter CMO period makes VIDYA twitchy and reintroduces the fake crosses you bought it to avoid. A much longer period makes it so slow that it lags real trends badly. Nine is the balance point for H1 and H4 funded trading.
Can I use VIDYA on its own as a complete strategy?
No, and you should be suspicious of anyone who tells you otherwise. VIDYA is a trend filter and a pullback reference. It does not set your stop logic, your targets, your sizing, or your daily loss budget. Pair it with a pullback structure, a fixed risk percentage, and a daily stop, and it becomes a strong component of a funded-account system.
Why does my VIDYA look different from another trader’s VIDYA?
Because the CMO period, the base factor, and the starting seed value all change the line, and most platforms implement VIDYA through different custom scripts. Two traders using different CMO periods are effectively using different indicators. Always compare settings before comparing results, and verify any script against the nine-bar example in this guide.
Does VIDYA work on gold or indices for funded accounts?
It can, but I only use it on EURUSD and GBPUSD, because their CMO readings are the cleanest and their spreads are the tightest. Gold’s violent ranges swing the CMO hard and make VIDYA speed up and slow down too violently for a fixed daily loss limit. If you trade gold, backtest VIDYA on gold separately and size from a wider stop grid.
What is the single biggest VIDYA mistake on a funded account?
Trading crosses of a flat VIDYA. Every backtest I have run says the same thing: the filtered rules make money and the unfiltered crosses lose it, on the same indicator and the same market. If you remember one line from this guide, remember this: a flat VIDYA is a stop sign, not a signal.
Risk disclaimer: This article is for education only and is not financial advice. Trading leveraged products on a funded account can lose your challenge fee and your account. Always size from your stop distance and your firm’s daily loss limit, and never risk money you cannot afford to lose.




