A Linear Regression Line is a best fit line drawn through the last N bars of price, most commonly the close. It is built to summarize the recent direction of price by finding the straight line that minimizes total squared distance from the data points. In practical terms, it turns a noisy sequence of candles into a single line with a slope you can read as trend direction and trend pace.
This indicator is often grouped with terms like least squares, linear regression moving average, or least squares moving average. The important idea is that it is not an average of prices like an SMA, it is a best fit trend estimate over a fixed window. Because it is fit to only the last N bars, it continuously updates and its slope can change quickly when the newest bars differ from the older ones in the window.
How the Linear Regression Line is calculated in a simple way
The calculation treats time as an index and price as the value to fit. Let x be bar positions 0 to N minus 1 and y be the chosen price for each bar such as close. The regression finds the slope m and intercept b of the line y = m x + b that best fits those N points in the least squares sense.
A simple formula version looks like this. Compute the average of x and the average of y over the window, then compute m = sum((x – xbar)(y – ybar)) / sum((x – xbar)^2) and b = ybar – m xbar. Once you have m and b, the line value plotted on the chart is the fitted value at each x in the window, and most platforms display the current value as b + m (N minus 1). The slope is the key output because it expresses direction and rate of change per bar in the units of price.
Most used settings and why traders choose them
Common periods cluster around the same horizons traders use for other trend tools: 20, 50, 100, and 200 on daily charts. A shorter window like 20 reacts faster to recent changes, which can help if you are timing breakouts and pullbacks in active moves. The tradeoff is that the slope can flip more often in choppy conditions because the fit is dominated by the newest bars.
Longer windows like 50 or 100 make the line steadier and reduce rapid slope flips. This can be useful when you want a higher level trend filter that does not change its message on every pullback. The cost is lag in recognizing a fresh turn because the older bars in the window still influence the fit. A practical way to think about period choice is to align it with your holding time, then check how often the slope changes sign during ranges versus trends.
How it behaves on charts and what signals look like
On the chart, the line often looks like a smoother trend line that sits near the middle of recent price action. When price trends steadily, the line tends to slope cleanly and price frequently oscillates around it on pullbacks and pushes. When price is accelerating, the line can lag behind the move at first and then steepen as the new bars enter the window and older bars roll off.
The most common signal interpretation is slope based. A rising slope is treated as bullish trend context and a falling slope is treated as bearish trend context, while a near flat slope is treated as range risk. Some traders also use distance from the line as a way to describe extension, where price far above the line suggests a stretched move relative to the current fitted trend. Others pair the line with a regression channel, which adds bands based on deviation around the fitted line so you can see whether price is trending with contained swings or thrashing with wide dispersion.
When it tends to work and why
This tool tends to be most useful when the market shows directional persistence over the chosen lookback. In that environment, the best fit line captures a meaningful slope and the slope stays consistent through normal pullbacks. That consistency is what makes it usable as a trend filter, because you are not trying to predict the next candle, you are trying to stay aligned with the direction that has been persistent enough to dominate the window.
It also tends to work better when you treat it as context and combine it with a simple regime check. For example, using a baseline trend line such as a moving average can help you avoid over reacting to small slope shifts that occur inside normal pullbacks, and you can compare how your slope filter behaves versus your baseline in the internal guide on Simple Moving Average SMA. In practice, the line is most helpful when it reduces decision noise and supports consistency, not when it is used as a standalone trigger.
When it tends to fail and why
The main failure mode is mean reverting chop where price rotates around a central value. In that regime, the best fit line has little stable slope, and the slope can flip sign frequently as the newest bars alternate direction. If you trade every slope change, you can end up taking a sequence of low quality entries that are really just reactions to noise inside a range.
Another failure mode is volatility spikes and gaps that distort the fit for a while. Because the regression minimizes squared errors, a few large moves can strongly influence the slope and intercept even if the broader structure is not truly trending. This can make the line look decisive right when conditions are least stable, especially if the move is quickly reversed. A practical defense is to require additional confirmation of trend strength and avoid interpreting a single sharp burst as a new regime.
Practical rules for entries, exits, stops, and filters
A workable approach is to use the line and its slope as a gate, then use price action for timing. For example, you can require the slope to be positive for long setups and negative for short setups, then take entries on breakouts, pullbacks, or consolidation resolution that match your style. This keeps the regression line in a supporting role where it filters direction rather than tries to provide precise entry points.
Here is a simple rule set you can test without adding many parameters:
- Use a 50 bar Linear Regression Line on your main timeframe as a direction filter, long only when slope is rising and short only when slope is falling
- Enter on a close beyond a clear breakout level or after a pullback that holds above the line in an uptrend and below the line in a downtrend
- Place a stop where the trade idea is invalidated, often below the most recent swing low for longs or above the most recent swing high for shorts
- Add a regime filter so you avoid flat slope conditions, and consider a strength check like the process described in How to use ADX Average Directional Index
For exits, keep it consistent with your entry logic. If you entered because a trend was strong and aligned, an exit can be triggered when slope flattens and price fails to make progress, or when price closes across the line and does not reclaim it quickly. You can also separate exit logic from the line by using structure based trailing stops, which often reduces the chance of exiting on a normal pullback. The key is to decide whether you want the regression line to manage the trade or simply to filter which trades you take.
Summary
A Linear Regression Line is a least squares best fit line applied to the last N bars of price, usually the close. Its core output is slope, which describes direction and the pace of change over the window. It is best used as a trend context tool and filter, not as a prediction engine.
Common periods like 20 and 50 balance responsiveness and stability, while longer windows like 100 can reduce slope flipping at the cost of slower turns. The indicator tends to be most useful in persistent trends where slope stays consistent through pullbacks, and it tends to fail in mean reverting ranges and during volatility spikes that distort the fit. If you keep it simple, use slope as the gate, use price structure for timing, and add one regime filter to avoid flat slope conditions.