Log vs Linear Chart Scale: Which to Use and How It Changes Your Levels

Plenty of traders never think about the scale setting until a trendline they've watched for a week suddenly looks broken after an accidental scale switch, which on most platforms is a single shortcut or button away. Price didn't move. The scale did. That is the whole topic in one sentence: the scale doesn't change any price, but it changes the picture you draw your lines on, and lines are what you trade off.

Quick Answer

What Is the Difference Between a Linear and a Log Chart?

It comes down to what the vertical axis treats as "the same size." The StockCharts ChartSchool primer on chart types and price scales puts it plainly: an arithmetic scale shows 10 points as the same vertical distance regardless of the price level, while a log scale measures price movements in percentage terms. TradingView calls its linear mode "Regular" and describes it as using absolute values in price changes.

The numbers that make it click

Take two moves: $10 to $20 and $100 to $200. Both are +100%. On a linear chart, the first is a $10 move and the second is a $100 move, so the second one is ten times taller. On a log chart they are exactly the same height, because the percentage is the same.

Flip it around. $50 to $51 and $100 to $101 are both $1 moves, so they look identical on linear. On log they don't. The first is +2%, the second is +1%, and the $50 move draws almost exactly twice as tall. A log axis shows this in its gridlines too: 10, 20, 40, 80 sit at equal spacing, because each step is a doubling.

Percent vs absolute: two ways to read size

This matters most when you eyeball volatility. A $2 candle on a $20 stock is a 10% bar. The same $2 candle on a $200 stock is 1%. Linear draws them the same height, so a chart that spans $20 to $200 makes the late candles look as violent as the early ones, which they weren't in percentage terms. Log draws the $20 candle ten times taller, which matches how it felt to hold it. Neither reading is wrong. Linear answers "how many dollars did it move," log answers "how big was that move relative to the price." Your stop is in dollars and your account cares about percent, so you end up needing both questions answered at different times. If the basics of reading the axis and candles are still new, start with this walkthrough of reading a stock chart and come back.

When Should You Use Log Scale vs Linear?

Linear: intraday, short-term, tight ranges

StockCharts describes arithmetic scales as useful when price is confined to a relatively tight range and for short-term charts and trading, where you want to see movements dollar for dollar. That covers most day trading. On a 5-minute chart where a stock trades between $40.00 and $41.50 all morning, the whole range is under 4%, and log and linear draw nearly the same picture. Linear wins on readability because your levels, stops and targets are all prices in dollars and cents, and the gridlines match them. Low-volatility, large-cap names on daily charts over a few weeks fall in the same bucket.

Log: long-term, wide ranges, big multipliers

Log earns its place once price has moved a lot. StockCharts says log scales are useful when price has moved significantly, over a short or extended timeframe, and for gauging percentage moves over a long period. A stock that went from $15 to $150 over three years is the classic case. On linear, everything below $40 is a flat smear along the bottom and the last year looks like a rocket. On log you can actually see the early base, the first breakout and the pullbacks in proportion.

A rule of thumb that gets repeated a lot: once price has roughly doubled within the visible chart, switch to log. It isn't a law, but it lines up with the math. Under a 2x range, the two scales draw similar shapes. Past it, they start disagreeing about trendlines, and the disagreement grows with the range. High-volatility names hit that threshold faster, so small caps, biotechs and recent IPOs often deserve log even on a few months of daily bars.

The exception for momentum traders: a low-float runner that goes from $3 to $9 in one session has tripled on the intraday chart. The 1-minute chart of that day covers a range where the scale choice really does move your trendlines. Most days aren't like that, and a default of linear intraday is still right. On the days that are, pick a scale on purpose. The multi-timeframe breakdown covers stacking a daily read over an intraday entry, which is often where you'll want log on the higher timeframe and linear on the lower one.

Here is everything above side by side.

The rows that cost money are the trendline and the Fibonacci ones. Gridlines and candle sizes are cosmetic. A line that moved is a trade you took or skipped.

Why Does the Same Trendline Break on One Scale and Not the Other?

Because a straight line means something different on each axis. On linear, a straight line rises by the same number of dollars every bar. On log, a straight line rises by the same percentage every bar. Pick the same two anchor points and you get two different lines. They agree at the anchors and drift apart everywhere else.

Here's a made-up example you can check with a calculator. A stock closes at $10 in month 0 and $20 in month 6. Draw a line through those two points.

On linear, the line adds $10 every six months, so it projects $30 at month 12. On log, the line doubles every six months, so it projects $40 at month 12. Now say the stock closes at $32 in month 12. On the linear chart, price is $2 above the line and the trend looks fine. On the log chart, price is $8 below the line and the trendline broke a few months back. Same stock, same closes, same anchors. One chart says hold, the other says the trend is over.

The StockCharts ChartSchool lesson on trend lines makes the same point from the other side: highs and lows tend to line up better on a semi-log scale, especially for long-term lines or after a large price change, and what looks like a break on arithmetic can be an intact line on semi-log. Which line is "right" depends on the trend. A stock compounding at a steady percentage rate draws a straight line on log and a curve that keeps steepening on linear. A stock grinding up by roughly fixed dollar amounts does the opposite.

The practical rule: decide the scale before you draw, and don't judge a break on a scale you didn't draw on. On many charting platforms, a drawn line keeps its two anchors and gets redrawn straight when you toggle the scale. When that happens, the line now passes through different prices everywhere except those anchors. The basics of picking anchors and confirming touches are in the guide to drawing trendlines that hold up; this is the extra step on top of it.

How Does Scale Change Support, Resistance and Fibonacci Levels?

Horizontal levels: same price, different spacing

A horizontal support at $25 is $25 on either scale. The price doesn't move. What moves is the visual gap between levels, and that gap is how most people judge whether there's room to a target. On a chart running from $10 to $40, the gap between $10 and $15 is half the height of the gap between $30 and $40 on linear, since $5 is half of $10. On log it's taller, because $10 to $15 is +50% and $30 to $40 is about +33%. If you eyeball "plenty of room overhead," the scale is doing some of that eyeballing for you. Read the prices instead. The fuller method for picking which levels matter is in the support and resistance breakdown.

Fibonacci retracements: the numbers actually change

Fibonacci is where scale changes the actual price, not just the look. A linear retracement splits the dollar range. A log retracement splits the percentage range. Take a swing from $10 up to $40 (a 4x move) and compare.

The 50% level is $25 on linear (halfway in dollars) and $20 on log (halfway in percent: $10 doubles to $20, $20 doubles to $40). That's a $5 disagreement on the most-watched level of the tool, and the 61.8% level is almost $4.50 apart. On a small swing, say $40.00 to $41.20, the two versions land within a couple of cents of each other and it doesn't matter. On a multi-month swing it matters a lot. Check whether your Fibonacci tool calculates on the scale you're viewing, and keep the tool and the chart on the same one. The Fibonacci retracement guide for day trading covers drawing the swing itself.

Where Each Scale Misleads

Linear on long-term charts

Linear distorts long histories. Early price action gets crushed into a flat line, and any stock that compounded looks parabolic at the end even if its percentage growth was steady the whole way. That parabolic look triggers bad instincts: "it's gone vertical, it has to top." Sometimes it does. Sometimes the log chart shows the exact same slope it has had for five years.

Log on small absolute moves

Log has the mirror problem. At high prices, a move that is small in percent looks tiny, even when it is a lot of dollars per share. On a chart spanning $20 to $400, a $12 drop at $380 is about 3% and barely registers, yet on a 500-share position that's $6,000. Log also makes intraday ranges feel calmer than your P&L will. Your risk is in dollars per share times shares, so when you size the trade, read the prices, not the picture. The reward-to-risk walkthrough does that math on a real setup.

How Do You Switch to Log Scale in TradingView?

Two ways. Per TradingView's own post on how to use log charts, you can press Alt + L on a PC or Option + L on a Mac, or click the log button near the bottom right of the chart to toggle it. TradingView's support doc on configuring Supercharts lists the modes as Regular, Logarithmic, Percent and Indexed to 100, and warns that turning log off returns the chart to the regular scale even if you had Percent or Indexed to 100 selected before. That last part catches people. The keyboard shortcut is also easy to hit by accident, which is exactly the "broken" trendline from the intro. If you're still deciding whether the paid tiers are worth it for drawing tools like these, there's a plain look at what TradingView's plans get you.

Which Scale to Screenshot Before an AI Read

If you grade setups from chart screenshots, the screenshot is the whole input. Whatever scale the chart was on when you hit capture is what gets read. With SnapPChart specifically, the AI reads prices off the price-axis labels first, as they are drawn, and builds the entry, stop and targets from those numbers. It does not detect whether the axis is log or linear, and it does not convert or normalise between them. You choose the scale. The AI reads what is drawn.

That splits the effect into two parts. Prices that sit right on a visible axis label, or show up in the OHLC header or the last-price tag, read the same on either scale. A support or resistance level that falls between two labels is different. It gets read by where it sits relative to the labels around it, and on a log axis that interpolation is exactly what goes wrong, because the spacing between labels isn't even. The geometry doesn't hold either. Say a stock sits at $40 with a stop at $38 and a target at $46. In dollars that is $2 of risk for $6 of reward, a clean 3:1. On a linear chart, the target sits exactly three times as far from entry as the stop. On a log chart, it sits about 2.7 times as far, because +15% to the target is less than three times the roughly 5% drop to the stop when both are measured in log terms. Same trade, different picture. Now shrink it to an intraday setup: entry $40.00, stop $39.80, target $40.60. Log puts the target about 2.97 times as far as the stop, so the two scales are basically identical. That's the math behind "intraday, just use linear."

Here's what I'd do before uploading:

Screenshot on the scale you actually trade. If you drew your lines on log, capture on log. If you trade intraday on linear, capture on linear.
Keep the price axis labels in frame and legible. On a log chart the ticks bunch together near the top, so zoom until there are labels close to your entry and stop. A price that sits on a label or in the header gets read directly. A level that falls between labels has to be estimated from its position, and log spacing makes that estimate drift. Better still, mark your entry, stop and key levels with horizontal lines so their price tags show on the axis.
Don't mix scales. A trendline drawn on log and then captured on linear passes through different prices than the one you were watching, and the read of that line will describe the line in the image, not the one in your head.
Default to linear intraday. Tight ranges make the scales agree, and dollar gridlines match how you place stops.

None of this is specific to one tool. Any AI that reads a chart image works from what the image contains, which is the main idea in the guide to how AI reads a chart. A neutral description of what one screenshot read covers is on the AI chart analysis page, and scale sits alongside indicators, structure and volume in the technical analysis overview.