Relative Rotation Graph (RRG) Explained: The Four Quadrants, RS-Ratio and RS-Momentum
A Relative Rotation Graph plots securities against a common benchmark on two axes, relative strength and the momentum of that relative strength, sorting them into four quadrants.
A Relative Rotation Graph, usually shortened to RRG, is a chart that compares a whole set of securities or sectors against one shared benchmark at the same time, and places each of them on a two-axis grid. The horizontal axis measures how strong an item is relative to that benchmark. The vertical axis measures whether that relative strength is currently rising or falling. Split the grid at the centre and you get four quadrants, and the position and direction of travel tell you where relative performance has been concentrating.
The idea it captures is old and simple: at any moment some parts of the market are doing better than the average and some worse, and those roles change over time. An RRG is a way of seeing that whole rotation on one page instead of flipping through fifty individual charts.
What the two axes actually measure
Every RRG needs three ingredients: a list of items to plot, a benchmark to plot them against, and a lookback period.
The horizontal axis is relative strength. In most implementations it is called the RS-Ratio. Conceptually, you take the price of the item, divide it by the price of the benchmark, and then normalise that ratio so it can be compared across very different securities. The normalisation is what makes the chart work: without it, a ratio line for a small-cap name and a ratio line for a large bank would sit on completely different scales and could not share one grid. After normalisation, a value above the centre line means the item has been outperforming the benchmark over the lookback window, and below the centre line means it has been underperforming.
The vertical axis is the momentum of that relative strength. Usually called RS-Momentum, it is the rate of change of the horizontal measure, normalised the same way. Above the centre line means the item’s relative strength has been improving. Below means it has been deteriorating. Crucially, this is momentum of the relative line, not momentum of the price. An item can be falling in absolute price and still show improving RS-Momentum, if it is falling more slowly than the benchmark.
So the two axes answer two different questions:
- Horizontal: has this item been beating the benchmark?
- Vertical: is that answer getting better or worse?
The four quadrants
The centre of the chart is the benchmark itself. Everything is measured against it, so the benchmark sits permanently at the middle point and never moves.
| Quadrant | Position | Reading in plain words |
|---|---|---|
| Leading | Top right | Outperforming, and the outperformance is still building |
| Weakening | Bottom right | Still outperforming, but the pace has started to fade |
| Lagging | Bottom left | Underperforming, and still getting worse |
| Improving | Top left | Still underperforming, but the gap has started to close |
The names are descriptive labels, not verdicts. “Leading” does not mean good and “Lagging” does not mean bad. They are shorthand for a position on two relative measures over one chosen window.
The reason the chart is called a rotation graph is that items tend to travel through these quadrants in a broadly clockwise direction. A name that has been out of favour starts to fall less than the market, so it drifts up into Improving. If the recovery persists, relative strength itself turns positive and it crosses into Leading. Eventually the pace of outperformance slows, momentum rolls over, and it slips into Weakening. If the fade continues, relative strength turns negative and it lands in Lagging.
That clockwise tendency is a tendency, not a law. Plenty of items cut across the middle, reverse, or sit in one quadrant for a long time. Reading the clockwise pattern as a rule is one of the most common ways people misuse the chart.
Tails: the part that carries the information
A single dot on an RRG is close to useless, because it tells you only where something stands today with no sense of how it got there. That is why RRGs are drawn with tails: a short trail of the previous few observations, so you can see the path.
The tail carries three pieces of information a static dot cannot:
- Direction. Which way is the item heading across the grid?
- Speed. Long tail segments mean fast movement. Short, bunched segments mean the item is drifting or stalling.
- Consistency. A smooth arc suggests a persistent rotation. A tail that doubles back on itself repeatedly suggests noise rather than a trend.
Tail length is a user setting. A short tail shows recent behaviour and reacts quickly. A long tail shows a fuller cycle but makes the chart crowded and slower to change. Neither is correct in the abstract, and the choice materially changes what the picture appears to say.
What choices you are making without noticing
An RRG looks objective because it is drawn from prices and arithmetic. It is objective given its inputs. The inputs are choices.
The benchmark decides everything. The same set of sectors plotted against a broad market index and against a mid-cap index can produce very different pictures, because “relative” is only meaningful against something. If you plot large-cap sectors against a small-cap benchmark, you are largely measuring the size effect, not sector rotation.
The universe decides the story. A grid of eleven broad sectors tells you about market structure. A grid of forty individual stocks tells you about individual names and will be far noisier. Mixing the two on one chart makes the comparison meaningless.
The lookback and smoothing decide the sensitivity. Longer windows produce slower, steadier rotation. Shorter windows produce a chart that whips around. Different providers use different normalisation and smoothing methods, which is why two RRGs of the same sectors on the same day can disagree.
Point-in-time constituents matter. If you are studying rotation historically, the sector or index membership you use should be the membership that existed on each past date, not today’s list. This is exactly the trap described in why point-in-time data matters, and it quietly flatters any historical rotation study built on today’s constituent lists.
What an RRG does not tell you
This is the section that keeps the tool honest.
It is entirely relative. Every position on the chart is measured against the benchmark. In a market where everything is falling, the Leading quadrant contains the things falling least. An RRG can look healthy while absolute returns are negative. Always read it next to an absolute price or return view.
It is entirely price-based. Nothing on the chart knows anything about earnings, cash flow, balance sheet strength, valuation, or the business. A sector can rotate into Leading on a narrative that fundamentals never support. If you want the fundamental view, that has to come from filings and financial analysis, not from this chart.
It is backward-looking by construction. Both axes are computed from past prices over a chosen window. Rotation you can see has already happened. The chart describes; it does not predict.
It is smoothed, so it lags. The normalisation and smoothing that make the chart readable also delay turns. A sharp reversal shows up on an RRG after the fact, not as it begins.
It has no notion of risk or size. Two items in the same quadrant may have wildly different volatility, liquidity and drawdown behaviour. The chart treats them identically. Position on the grid is not a substitute for risk-adjusted return analysis.
Crowding and cluster effects are invisible. When many items sit in the same quadrant, that itself may be information about how concentrated positioning has become, but the chart will not flag it.
Used carefully, an RRG is a good structural map: it compresses a lot of relative performance data into one legible picture and forces you to name your benchmark. Used carelessly, it becomes a chart that seems to say “buy the top right”, which is not what it measures and not what it can support.
Related reading
- Portfolio metrics explained: the hub for the risk, return and portfolio measures referenced here.
- How to read an RRG chart: the practical reading of tails, rotation and the failure modes.
- Relative strength explained: the underlying measure that the horizontal axis is built from.
- Sector rotation strategy in India: how rotation between sectors is analysed more broadly.
- What is the momentum factor: the academic cousin of relative strength, and its known crash risk.
This article is educational. Altys Labs is not a registered research analyst or investment adviser, and nothing here is investment advice or a recommendation to buy, sell, or hold any security.
Frequently asked questions
What is a Relative Rotation Graph?
It is a chart that compares many securities or sectors to one shared benchmark at the same time. Each item is plotted using two numbers: how strong it is relative to the benchmark, and whether that relative strength is rising or falling. The two numbers become the horizontal and vertical position on a grid split into four quadrants.
What do the four quadrants of an RRG mean?
Broadly: Leading means strong and still strengthening, Weakening means strong but losing pace, Lagging means weak and still weakening, and Improving means weak but picking up. Names tend to travel between quadrants over time rather than sit still.
Is an RRG a buy or sell signal?
No. An RRG is a description of relative price behaviour that has already happened, not a forecast and not a recommendation. It tells you where attention has been flowing. It says nothing about valuation, business quality, or what happens next.