Skip to main content

How Overlay Chart Tricks Distort Financial News

Financial news outlets love to compare two things using overlay charts. They stack two lines—or sometimes two different datasets with two different scales—on top of each other, claiming a relationship exists between them. A chart might show the stock market rising alongside unemployment falling, or compare Bitcoin price movements to Fed interest rate changes. The visual overlay creates an impression of correlation. But overlay charts are among the most deceptive visualizations in financial journalism, and understanding how they lie is essential to reading news critically.

An overlay chart places two datasets on the same visual space. If you're comparing two things measured in the same units (both prices, both percentages), this can be honest. But financial news often overlays datasets measured in completely different units—stock prices in dollars, unemployment in percentage points, interest rates in basis points. When the axes scale independently, the outlet controlling the chart can make any two things look correlated.

Quick definition: Overlay chart tricks use independent axis scaling, cherry-picked time windows, and visual misalignment to create false impressions of correlation between unrelated financial datasets.

Key takeaways​

  • Dual-axis charts with different scales can show false correlation — stretching one axis makes unrelated things look synchronized
  • Time window selection determines what a chart "proves" — choosing the right date range makes any two things appear related
  • Different measurement units without context are deceptive — comparing stock prices to unemployment rates requires careful axis scaling
  • Overlay charts work by exploiting visual pattern-matching — humans see patterns and assume causation when they see lines moving together
  • The solution is always to verify the actual numbers — screenshot the chart and check the actual correlation in the source data
  • Financial outlets use overlay tricks to suggest relationships that don't exist — supporting a narrative that feels intuitively right to viewers

The Dual-Axis Deception: Scaling One Series to Create Correlation​

The most common overlay trick uses two independent vertical axes, one on the left and one on the right. This allows the chart creator to scale each dataset separately, stretching one and compressing the other until the visual lines appear to move together.

Here's a concrete example. Imagine you want to suggest that the stock market is correlated with weather. This is absurd—weather doesn't meaningfully move stock prices. But with dual-axis scaling, you can make it look true.

Take the S&P 500 stock index, which fluctuates between roughly 3,000 and 5,000 points (a 67% range). On the left axis, scale it from 2,900 to 5,100. Now take the daily temperature in New York over the same period, which fluctuates between 32 and 85 degrees Fahrenheit (a 165% range). On the right axis, scale it from 20 to 100 degrees.

With these scales, the curves will appear visually similar even though no relationship exists between them. Stocks might be at 4,200 (center of the left scale) while temperature is at 60 degrees (center of the right scale). Both look "centered." If temperature rises from 55 to 65, the right axis expands that movement to look large. If the stock market drops 200 points, the left axis compresses that movement to look small. The visual impression is that they move together when they don't.

This is exactly what financial news does with overlay charts. A financial outlet wants to argue that "Fed rate hikes are causing stock market volatility." They take the Fed funds rate (which ranges from 0% to 5.5%, a 5.5 percentage-point range) and the S&P 500 (which ranges from 3,000 to 5,000, a 2,000-point range). By scaling the Fed rate axis from -1% to 6% and the stock market axis from 2,500 to 5,500, they can make these visually align in whatever pattern supports the narrative.

The trick exploits human visual pattern-matching. Your brain sees two lines moving together and concludes they're related. You don't think to check the axis scales, and the outlet doesn't usually highlight them clearly.

The financial media has published thousands of these misleading overlays. A chart might show oil prices rising alongside airline stock prices (true correlation, makes sense). But it might also show them with axes scaled to exaggerate the correlation, making a 2% movement in oil look as large as a 10% movement in airline stocks.

Time Window Selection: The Chart's Most Powerful Weapon​

Even with honest scaling, overlay charts can deceive through time window selection. By choosing the right date range, you can make any two things appear correlated.

Consider the S&P 500 and Bitcoin. Over certain periods, they move together (both are risk assets, so they correlate during risk-on markets). Over other periods, they diverge sharply (Bitcoin crashed 65% in 2022 while the S&P only fell 18%). A financial outlet wanting to claim they're "strongly correlated" would publish a chart from March 2020 to November 2021—a period when they genuinely moved together. A different outlet wanting to claim "Bitcoin is uncorrelated to stocks" would publish a chart from January 2022 to December 2023, when they diverged.

Both charts could be technically accurate (honest axes, real data) but tell completely opposite stories through time window selection.

Here's another example. An analyst might want to argue that "gold is an inflation hedge." Gold prices from 1980 to 1985 rose while inflation fell dramatically—no hedge. Gold prices from 2008 to 2013 rose while inflation stayed low—no hedge. But gold prices from 2021 to 2022 rose while inflation accelerated—perfect hedge! By publishing a chart starting in 2021, the analyst can prove gold is an inflation hedge, ignoring 40 years of data showing it isn't.

Financial news outlets do this constantly. They'll overlay the Federal Reserve's balance sheet against stock prices, then start the chart right at the March 2020 market bottom—a moment when the Fed was expanding its balance sheet and stocks were recovering. This creates a visual impression of causation: "Fed buying = market up." But if they started the chart in 2018, or if they included the period from 2015-2019 when the Fed was shrinking its balance sheet while stocks rose, the narrative would be different.

The solution requires you to ask: Why did the outlet choose this specific time window? What would the chart look like if it started two years earlier or ended two years later? You can often test this using financial databases or charting platforms that let you change the date range.

Comparing Incompatible Units: The Ultimate Trick​

A more sophisticated overlay trick compares things measured in completely different units without meaningful conversion. A chart might show "Federal debt (trillions of dollars)" overlaid with "S&P 500 (index points)." These are incompatible units. An index point for the S&P 500 isn't a dollar. Federal debt is an absolute number; the S&P 500 is a relative measure of stock prices.

Yet by scaling the axes independently, you can make federal debt and stock prices appear to move together, suggesting causation. This is especially common in conspiracy-theory financial content. A chart might show federal debt rising alongside cryptocurrencies rising, implying that loose monetary policy is "forcing" people into crypto. But federal debt has been rising for 60 years while crypto didn't exist until 2009. The chart is pure misdirection.

Another frequent comparison: "Corporate earnings" (measured in dollars, absolute values) overlaid with "stock market volatility index" (the VIX, measured in percentage-point volatility). These are completely incompatible. Earnings are absolute dollar amounts; volatility is a measure of price fluctuation. A chart might show earnings declining 20% while volatility triples, visually suggesting they're opposite forces. But earnings and volatility measure entirely different things and shouldn't be compared using the same visual space.

The correct approach when comparing incompatible units is to convert both to percentage changes (or z-scores, or other normalized measures) before overlaying. A financial outlet that doesn't do this is either careless or intentionally misleading.

The Correlation-Causation Trap​

Even honest overlay charts—with appropriate scaling and full-period data—exploit humans' tendency to see correlation as causation. Two things can move together for completely different reasons.

Stock prices and ice cream sales move together seasonally—both rise in summer. This is correlation. But it's not causation (stocks don't rise because people buy ice cream). The cause is the season.

The yield curve (the difference between long-term and short-term interest rates) often inverts before recessions. A financial outlet might overlay these two things and suggest the yield curve "causes" recessions. But that's not quite right. The yield curve inverts because investors expect future economic weakness. The expected future weakness causes both the yield curve inversion and the recession. The yield curve is a signal, not a cause.

A concrete example: In 2021 and early 2022, the Federal Reserve began raising interest rates to combat inflation. Financial outlets published overlay charts showing Fed rate hikes rising alongside stock market declines. The visual correlation suggests "Fed hikes caused stock declines." But causality is more complex. The Fed raised rates because it expected economic weakness. Economic weakness was already driving both the Fed's decision and stock market declines. The correlation is real, but the causality is more indirect than the overlay suggests.

When you see an overlay chart, always ask: Which direction does causation flow? Or is a third factor causing both?

Real-World Examples: How Outlets Use Overlay Tricks​

Example 1: Fed Balance Sheet vs. Stock Prices

A financial outlet publishes a chart showing the Federal Reserve's balance sheet (measured in trillions of dollars, left axis) overlaid with the S&P 500 (measured in index points, right axis). The two lines rise together from 2020-2021, visually suggesting the Fed's balance sheet expansion "caused" stock gains.

The trick: The axes are scaled so that the Fed balance sheet appears as a near-straight line (it rises from $4.2 trillion to $8.9 trillion, only a 2x change), while the S&P 500 appears volatile (it rises from 2,600 to 4,700, a 1.8x change). The visual scaling exaggerates the Fed's role and minimizes the S&P's volatility. The chart ignores 2015-2019, when the Fed shrank its balance sheet while stocks rose.

Example 2: Unemployment vs. Stock Market

A political outlet publishes an overlay chart showing unemployment (left axis, ranging 3% to 13%) and the S&P 500 (right axis, ranging 2,000 to 3,500 points). The lines appear to move in opposite directions—when unemployment rises, stocks fall. This suggests a causal relationship: "Job losses cause market declines."

The trick: The relationship is real but the causality is indirect. Both unemployment and stocks decline during recessions caused by other factors (monetary shocks, demand collapse, etc.). The overlay suggests the unemployment itself causes market declines when actually both respond to the same underlying economic shock.

Example 3: Bitcoin vs. Fed Rates

A crypto-focused outlet publishes a chart comparing Bitcoin prices to Fed interest rates. The chart starts in March 2020 (when the Fed cut rates to zero and Bitcoin surged) and ends in November 2021 (when Bitcoin hit all-time highs and the Fed was still at zero rates).

The trick: The time window is cherry-picked. If the chart continued to December 2023, it would show the Fed raising rates to 5.25% while Bitcoin surged from $16,000 to $42,000—the opposite relationship. Or if the chart went back to 2013-2018, it would show the Fed raising rates from 0% to 2.5% while Bitcoin surged from $100 to $20,000. Neither the positive correlation (2020-2021) nor the negative correlation (2022-2023) proves causation, but the time window selection determines which story the outlet tells.

Common Mistakes: How Readers Misinterpret Overlay Charts​

Many investors misunderstand overlay charts in predictable ways.

They assume that if two lines visually move together, they're meaningfully related. Visual correlation is real, but it doesn't prove the relationship you think it does.

They trust the outlet's interpretation without checking the axes. Always verify axis ranges. If the two axes don't have clear labels and values, the outlet is being deceptive.

They confuse correlation with causation. Two things can move together for unrelated reasons.

They ignore time window selection. A chart that "proves" a relationship over 2 years might tell a completely different story over 20 years.

They assume professional outlets are honest. The outlets are often honest about the data itself, but they're frequently deceptive in their axis scaling and time window selection, especially when they have a narrative to support.

FAQ: Understanding Overlay Chart Deceptions​

How can I tell if an overlay chart's axes are manipulated?​

First, check the axis ranges explicitly. A chart showing the Fed balance sheet (trillions of dollars) alongside stocks (thousands of index points) should scale appropriately or use percentage changes instead. If the left axis goes from 3 to 9 and the right axis goes from 2,000 to 4,500, they're scaling differently by necessity, but the outlet should note this. Second, take a screenshot and verify the actual correlation in the source data using a spreadsheet or statistical tool. If the visual correlation looks stronger than the actual data correlation, you've found a trick.

Should I ever trust overlay charts?​

Overlay charts can be honest, but they're inherently prone to deception. They're useful for exploring whether two things might be related, but they shouldn't be your evidence that they are related. Always verify with actual statistical correlation coefficients and multiple time periods. Honest financial analysis uses overlay charts as a starting point for investigation, not as proof of anything.

What's the correct way to compare two different datasets?​

The most honest approach is to convert both datasets to percentage changes or z-scores (standard deviations from the mean) so they're on comparable scales. This removes the axis-scaling trick. Then test the correlation over multiple time periods to account for time-window selection bias. Finally, consider causality separately from correlation—even a strong correlation doesn't prove one thing caused the other.

Why do financial outlets use overlay charts if they're deceptive?​

Because they work. Overlay charts are visually compelling. They create an instant impression of relationship that resonates with viewers. Even when the correlation is weak or spurious, the visual overlay feels convincing. A business outlet that published a boring statistical analysis ("the correlation coefficient is 0.3 over this period, but 0.1 over that period") would lose viewers. But publish an overlay chart showing two lines rising together, and viewers share it widely.

Can I create my own honest overlay charts?​

Yes, but it requires discipline. Start with compatible units or convert to percentage changes. Use axis ranges that represent the actual data variation, not manipulated scales. Show the full time period, not a cherry-picked window. Include correlation statistics with the chart—not just a visual impression. Most importantly, consider whether overlaying two things is even the right visualization. Sometimes showing two separate charts is more honest than forcing them into one.

What about overlaying the same indicator in different time periods?​

This is often honest and useful. You might overlay the S&P 500 price from 2020-2021 and 2022-2023 to compare the recovery patterns. Or compare the Fed funds rate during different inflationary episodes. When both datasets use the same units and are measured the same way, overlaying is typically fine as long as you acknowledge any time-window differences.

Summary​

Overlay charts are the most deceptive visualization tool in financial journalism. They use independent axis scaling, cherry-picked time windows, and comparisons of incompatible units to suggest correlations that don't exist or exaggerate weak correlations into apparent relationships. The human brain is primed to see visual patterns and assume correlation implies causation. By controlling the chart's axis ranges and time window, an outlet can make virtually any two things appear related. The solution is to verify the actual correlation coefficient in the source data, examine the full historical period (not just a selected window), and ask whether a visual correlation truly implies the causal relationship the outlet claims. Always remember that two things can be correlated for completely unrelated reasons, and even a strong visual correlation doesn't prove one caused the other.

Next​

→ Tables vs. charts: misleading uses