What Is the Ultimate Oscillator?
The Ultimate Oscillator is a bounded momentum indicator, developed by Larry Williams in 1976, that combines buying pressure measured across three different timeframes, commonly 7, 14, and 28 periods, into a single weighted reading between 0 and 100. It was designed to reduce the false-divergence signals that can occur when a momentum oscillator relies on only one timeframe.
Direct Answer
The Ultimate Oscillator is a bounded momentum indicator, developed by Larry Williams in 1976, that combines buying pressure measured across three different timeframes, commonly 7, 14, and 28 periods, into a single weighted reading between 0 and 100. It was designed to reduce the false-divergence signals that can occur when a momentum oscillator relies on only one timeframe.
Key Takeaways
- Created by Larry Williams in 1976 to address a specific weakness of single-timeframe oscillators: divergence signals that disagree depending on which lookback period is used.
- Combines a short, medium, and long lookback (commonly 7, 14, and 28 periods) into one weighted reading, with the shortest period weighted most heavily.
- Bounded between 0 and 100, read the same general way as other bounded oscillators, extremes suggest overbought/oversold conditions.
- Divergence between the oscillator and price is one of its primary uses, alongside extreme-reading reversals.
- Because it blends timeframes, its formula is more complex and less intuitive to read at a glance than a single-period oscillator like RSI.
- Like all oscillators, it describes historical price behavior and does not predict future moves.
Why Blend Three Timeframes?
Most momentum oscillators, RSI, the Stochastic Oscillator, and similar tools, calculate their reading over a single lookback period. That works well when short-term and longer-term momentum agree, but it can produce a misleading divergence signal when they don't: the oscillator might show a bullish divergence on a 14-period setting while a 28-period version of the same indicator shows nothing unusual, simply because the two windows are looking at different slices of recent price action.
Larry Williams built the Ultimate Oscillator to address that specific problem. Rather than picking one period and living with its blind spots, the indicator calculates buying pressure across a short, medium, and long lookback (7, 14, and 28 periods in the common default) and combines the three into a single weighted average. The shortest period is weighted most heavily since it reacts fastest to price changes, with the medium and long periods weighted progressively less, intended to keep the reading responsive while still reflecting the broader trend.
How the Ultimate Oscillator Is Built
The calculation starts with two per-bar building blocks rather than working from closing price alone:
- Buying Pressure (BP): the current close minus the true low (the lesser of the current low or the prior close), a measure of how far the bar closed above its downside extreme.
- True Range (TR): the same true-range calculation used in other volatility-aware indicators, the greatest of the current high minus current low, current high minus prior close, or current low minus prior close.
For each of the three periods, the indicator sums Buying Pressure over that lookback and divides by the sum of True Range over the same lookback, producing a raw average for that timeframe. Those three averages are then combined with weights that favor the shortest period (commonly 4x for the short period, 2x for the medium period, and 1x for the long period), and the weighted sum is scaled to fall between 0 and 100. The exact weighting can vary slightly by charting platform, but the short-heavy weighting scheme is standard.
Reading Overbought, Oversold, and Divergence
Because it's bounded between 0 and 100, the Ultimate Oscillator is read the same general way as other bounded oscillators: readings near the top of the range are considered overbought, readings near the bottom are considered oversold, and there is no single universal threshold, traders typically calibrate the levels they watch to the specific instrument's own historical range rather than applying one fixed number everywhere.
Divergence is one of the indicator's most cited uses, and it's the specific behavior the three-timeframe design was meant to make more reliable. A bullish divergence occurs when price prints a lower low while the oscillator prints a higher low, a sign that downside momentum, as measured across all three timeframes at once, is not confirming the new price low. A bearish divergence is the mirror case: price makes a higher high while the oscillator makes a lower high. As with any divergence signal, it flags a potential shift in momentum, not a guaranteed reversal, and is typically combined with price confirmation before acting on it.
Limitations and Common Mistakes
- Treating divergence as a standalone trade signal, a divergence describes disagreement between price and momentum; it doesn't specify timing or confirm that price will actually turn.
- Ignoring the underlying trend, like other oscillators, extreme readings can persist through a strong trend rather than immediately reversing.
- Assuming the blended formula removes all noise, combining three timeframes reduces (it doesn't eliminate) single-timeframe false signals, and the indicator can still lag or whipsaw in choppy conditions.
- Applying one fixed overbought/oversold threshold across every instrument, a level that works for a low-volatility index can be routinely breached by a higher-volatility stock or crypto asset.
- Using it in isolation, as with most single-indicator approaches, it's typically paired with price action, trend context, or another confirming signal rather than traded alone.
Whether the Blended Formula Is Worth the Complexity
The Ultimate Oscillator was built to solve one specific complaint: a divergence that appears on a 14-period oscillator and disappears on a 28-period version of the same tool. If that problem has never bothered you, blending three lookbacks buys little, and a simpler bounded oscillator will be easier to interpret under pressure. If it has, this is a considered answer to it, with the shortest window weighted most heavily so the blend stays responsive.
What the blend does not do is remove noise. Combining 7, 14 and 28 periods reduces the odds that a signal is an artefact of one arbitrary lookback. It does not make the reading immune to choppy price action, and it does not stop the oscillator from staying extended through a strong trend the way any bounded oscillator can.
The mistake to avoid is borrowing thresholds. Because the formula starts from buying pressure and true range rather than from closes alone, its distribution differs from RSI, and a level that rarely trips on a low-volatility index can be routine on a volatile stock or crypto asset. Calibrate to the instrument own history before deciding what counts as an extreme on it.
There is also a practical cost worth naming. A reading assembled from three weighted timeframes is harder to reason about at a glance than a single-period oscillator, which matters when you are trying to explain to yourself why you are acting. If you cannot say what moved the number, the extra sophistication is working against you.
Ultimate Oscillator FAQs
Why does the Ultimate Oscillator use three timeframes instead of one?
A single-timeframe oscillator can show a false divergence when short-term price action briefly disagrees with the broader trend. Larry Williams designed the Ultimate Oscillator to weight a short, medium, and long lookback together so a reading is less likely to be dominated by noise on any one timeframe.
What periods does the Ultimate Oscillator use?
The commonly used periods are 7, 14, and 28, with the shortest timeframe weighted most heavily and the longest weighted least. Charting platforms let these periods be adjusted, but 7/14/28 is the standard default most traders reference.
Is the Ultimate Oscillator better than a single-timeframe oscillator like RSI?
It is not universally better. It is designed to address one specific weakness (single-timeframe false divergence) at the cost of a more complex, less intuitive formula. Many traders use it alongside, not instead of, other oscillators.
How is overbought and oversold defined on the Ultimate Oscillator?
Because it is a bounded oscillator between 0 and 100, extremes near the top and bottom of that range are generally read as overbought and oversold, similar to how RSI or the Stochastic Oscillator are read. There is no universal fixed threshold; traders typically calibrate levels to the specific instrument's historical range.
What is bullish or bearish divergence on the Ultimate Oscillator?
Bullish divergence is when price makes a lower low but the oscillator makes a higher low, suggesting weakening downside momentum. Bearish divergence is the mirror case: price makes a higher high while the oscillator makes a lower high, suggesting weakening upside momentum.
What is buying pressure in the Ultimate Oscillator formula?
Buying pressure is the close minus the true low, where the true low is the lower of the current low and the previous close. It measures how far price closed above the lowest point that mattered for the period, including any overnight gap. Dividing the sum of buying pressure by the sum of true range over each window is what produces the three component ratios the oscillator blends.
Why is the shortest period weighted most heavily?
Larry Williams weighted the components so that the fastest window has the most influence and the slowest the least, on the reasoning that recent behaviour should dominate while the longer windows supply context. The weights are part of the published definition rather than a result of optimisation. Changing them produces a different indicator that will not match any reference material written about the original.
Was the Ultimate Oscillator designed to reduce false divergences?
That was the stated motivation. A single-period oscillator can show a divergence that exists only because of the specific window chosen, and blending three windows was intended to make such artefacts less likely to survive. Whether it succeeds is an empirical question that the design does not settle. What the construction does guarantee is that a divergence must hold across three timescales rather than one.
How much history does the Ultimate Oscillator need?
At least as much as its longest window, plus one extra bar because the true range and true low calculations reference the previous close. In practice a little more than the longest period is needed before the readings settle, since the first values are computed from a partially filled window. That makes it more demanding than a single-window oscillator with the same nominal fast setting.
References
Disclaimer
This page is for educational purposes only and does not constitute investment advice. The Ultimate Oscillator, like any technical indicator, describes historical price behavior and does not predict future performance. Nothing on this page is a recommendation to buy or sell any security or asset.