Momentum indicators

Chande Momentum Oscillator (CMO): Formula, CMO vs RSI, and Every Signal on Bitcoin

Tushar Chande’s CMO compares the sum of up moves with the sum of down moves over 20 periods. How it is calculated, why it is the RSI rescaled yet does not match it on a chart, and what its ±50 signals gave on Bitcoin.

· Founder of PatternsWizard
Checked 6 min read
CMO(20) on Bitcoin, daily, August 30, 2026 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
On this page
  1. How the CMO is calculated
  2. CMO vs RSI
  3. Signals: every ±50 entry on Bitcoin
  4. Does it work?
  5. Frequently asked questions

The Chande Momentum Oscillator (CMO) compares the sum of the up moves with the sum of the down moves over the last 20 periods. It reads +100 when every close was higher than the one before, −100 when every close was lower, and 0 when gains and losses cancel out. Tushar Chande introduced it in 1994 in The New Technical Trader, written with Stanley Kroll. It looks like the RSI, and on paper it is almost the same thing; in practice the two lines differ because of how they average. Below: the formula, the link with the RSI, and every ±50 signal it gave on Bitcoin over the last 250 days.

Chande Momentum Oscillator (20) and the rescaled RSI (20) on Bitcoin, daily

CMO(20) on Bitcoin, daily, February 21, 2026 to October 28, 2026. Illustration of how the indicator reads, not a backtest.
  • Price, daily candles (hollow up, filled down)
  • CMO(20), plain sums of up and down closes (solid line)
  • Wilder RSI(20) on the same scale, 2 × RSI − 100 (dashed line)
CMO(20) on Bitcoin, daily, February 21, 2026 to October 28, 2026. Illustration of how the indicator reads, not a backtest.

How the CMO is calculated

For each period, take the change in the close. Add up the gains over the last n periods (Su) and the losses, as positive numbers (Sd). The CMO is their difference divided by their sum, times 100. TA-Lib notes that this version, with plain sums rather than smoothed averages, is the one Chande defined in his book and the one TradingView uses.

CMO vs RSI

With plain sums, the RSI would be 100 × Su / (Su + Sd), so the CMO is exactly 2 × RSI − 100: same information, on a −100 to +100 scale. The RSI that charting platforms show, though, is Wilder’s, which replaces the sums with a smoothed average that keeps a trace of every past move. That is why the two lines in the chart above do not overlap. On Bitcoin over the window, with both set to 20 periods, the gap between the CMO and the rescaled RSI reached 36 points. The CMO swung from −74.10 on June 5, 2026 to +75.57 on August 21; the rescaled RSI, from −54.70 to +58.77. The CMO reacts faster and reaches its extremes more often.

Lowest and highest readings on Bitcoin, same 250 days

Lowest and highest readings on Bitcoin, same 250 days. Lowest reading: CMO(20) −74.10, Wilder RSI(20), as 2 × RSI − 100 −54.70; Highest reading: CMO(20) +75.57, Wilder RSI(20), as 2 × RSI − 100 +58.77. Our computation; the inputs and the method are in the text.
  • CMO(20)
  • Wilder RSI(20), as 2 × RSI − 100
Our computation; the inputs and the method are in the text.Daily Bitcoin, February 21 to October 28, 2026, both on the same −100 to +100 scale. Lows on June 5 (CMO) and June 6 (RSI), highs on August 21 for both: the plain sums of the CMO reach further in both directions than Wilder’s smoothed averages.

The plain sums have a second effect: the CMO forgets a move as soon as it leaves the 20-period window. A big up day lifts the sum of gains for exactly 20 bars; on the 21st, it drops out, and the CMO can fall on a day when the price barely moves. Wilder’s average lets old moves fade gradually instead, so the RSI has no such jump.

Why the CMO can fall on a quiet day

Why the CMO can fall on a quiet day: a price path with one big up day followed by quiet days. A first bracket marks today's window of closes, which includes the big up day, so the CMO stays high. A second bracket, a few bars later, no longer includes it: the sum of up moves shrinks and the CMO falls although price barely moves.
Schematic, not market data.

Signals: every ±50 entry on Bitcoin

TradingView describes the usual reading: above +50 is overbought, below −50 oversold, and a moving average of the CMO, such as a 10-period one, serves as a signal line. Over the 250 daily bars from February 21 to October 28, 2026, the CMO(20) spent 19 days above +50 and 11 below −50, and crossed zero 32 times. An entry is the first close beyond ±50 after a close inside; these are all four, with Bitcoin’s close 10 and 20 days later.

EntryZoneClose10 days later20 days later
June 2, 2026Below −5066,760.83−4.76%−4.11%
June 12, 2026Below −5063,580.01+0.69%−3.18%
August 20, 2026Above +5073,025.15+6.38%+7.23%
October 5, 2026Above +5085,766.87−4.11%−6.54%

Read as reversal signals, three of the four were wrong after 20 days: Bitcoin kept falling after both oversold entries in June, and kept rising after the August one. Only the October overbought entry was followed by a decline. At the stricter ±70 threshold, there was one entry each way (August 21 above, June 3 below) and both were followed by a lower close 20 days later. Read as momentum, the August entry was the start of a rise, not the end of one.

On the S&P 500 ETF over its last 250 sessions, the CMO(20) went above +50 on 18 days and never below −50; its lowest reading, −47.03, came on March 30, 2026, the day of the fund’s lowest close of the year.

Does it work?

We have no long-run measurement of the CMO: our bench tests chart patterns detected by rule against random entries, not indicators (see our methodology). Four entries in eight months are far too few to judge, and on this sample they did not work as reversal signals. TradingView itself warns that overbought and oversold readings can be less effective in strong trends.

Related pages: the momentum indicator, the Stochastic RSI, also by Chande and Kroll, the Chande Kroll stop, RSI divergence, and the list of momentum indicators.

Every pattern's results on one sheet

The only pattern cheat sheet with tested stats: 152 patterns, what the textbook says and what really happened on 6 markets. Free PDF, sent to your inbox.

The PDF by email, plus our newsletter: new tests and results that change. Unsubscribe anytime. Privacy.

Frequently asked questions

What is the Chande Momentum Oscillator?

A momentum oscillator by Tushar Chande that divides the difference between recent gains and losses by their total, over usually 20 periods. It runs from −100 to +100.

What is the difference between the CMO and the RSI?

The CMO uses plain sums of gains and losses and is centered on zero; the usual RSI uses Wilder’s smoothed averages and runs from 0 to 100. With the same sums, CMO = 2 × RSI − 100; with Wilder’s averages the two lines differ, sometimes by a lot.

What are the overbought and oversold levels of the CMO?

+50 and −50 are the usual levels. Some traders use ±70 to get fewer signals: on Bitcoin over 250 days, ±70 gave one entry each way against two each way for ±50.

What is the best period for the CMO?

20 is the usual setting and TA-Lib defaults to 14. None has been shown to work best; a shorter period gives more and noisier signals.

Sources

  1. TradingView, Chande Momentum Oscillator (CMO)
  2. TA-Lib, Chande Momentum Oscillator (Unsmoothed) (CMOU)
  3. StockCharts ChartSchool, Relative Strength Index (RSI)
, Founder of PatternsWizard

Passionate about finance, quantitative trading and data analysis. Amaury built the PatternsWizard test bench, which detects each pattern on full price history and measures what happened next against random entries.

PatternsWizard tests chart and candlestick patterns on full market history, against random entries. How we test patterns.

179 patterns testedRandom-walk self-test passed

Related guides

All momentum indicators articles