The basics nobody bothers to explain
A stock chart shows the price of a stock over time. The X-axis is time. The Y-axis is price. The line goes up when the stock goes up.
If that were all there was to it, this article would end here. But charts have hidden traps. The same stock can look like a "must buy" on one chart and a "stay away" on another, depending on:
- The time window
- Whether the Y-axis is linear or logarithmic
- Whether dividends are included
- What benchmark it's plotted against
Get those wrong and you'll draw the wrong conclusions about what the stock has actually done.
Step 1: Always check the time window
The single biggest source of misreading a chart is choosing the wrong time window.
A stock that looks like a screaming buy over the past 6 months might look like a structurally broken business over the past 5 years. Both views can be true simultaneously. The honest analyst looks at multiple time windows.
A useful default: 5-year, 10-year, and since-inception views. The 5-year captures the recent narrative. The 10-year captures full business cycles. The since-inception view tells you what the long-run owner has actually experienced.
Be especially careful with:
- 1-day or 1-week views: almost always pure noise
- YTD views: biased by where the year happens to start
- "All time" on a recent IPO: the chart starts at a price that may have been engineered for the IPO
Step 2: Pick the right Y-axis scale
This is the trap that fools more beginners than any other.
A linear Y-axis treats every dollar of price the same. A move from $10 to $20 looks the same as a move from $100 to $110, even though the first is a 100% gain and the second is a 10% gain.
A logarithmic Y-axis (also called "log scale") treats every percentage move the same. A 10% gain looks the same whether the stock is at $10 or $1,000.
For long time periods, always use log scale. A 10-year chart of a stock that went from $5 to $500 is unreadable on linear scale, it looks like the stock did nothing for 8 years and then exploded. On log scale, you can actually see the rate of growth across the full window.
For short time periods, linear is fine. Over 6 months, the percentage moves are small enough that linear and log charts look nearly identical.
The Advising Alpha portfolio detail pages all use log scale for the same reason, see the Core 20 chart.
Step 3: Total return, not price return
When you look at a stock chart on most websites, the line you see is the price return. what the price did, ignoring dividends.
For most stocks, that's fine. Apple doesn't pay much in dividends, so its price chart and total-return chart are nearly identical.
For dividend-heavy stocks (utilities, REITs, consumer staples), it matters enormously. A stock paying a 4% dividend has a total return roughly 4 percentage points per year higher than its price return. Over 30 years, that compounds into a multiple-of-2 or 3 difference in ending value.
When comparing a stock to the S&P 500, make sure you're comparing total return to total return. The S&P's price index is misleading; the S&P 500 Total Return Index (SPXTR or SP500TR) includes reinvested dividends. Real returns work this way. Pretending dividends don't exist understates everything.
Step 4: The reference benchmark matters
A stock can be up 50% in 5 years and have lost to the S&P 500. It can be up 20% and have crushed it. The chart of the stock alone tells you nothing about whether the company outperformed.
Always plot a stock against a relevant benchmark:
- Large-cap US stocks: vs the S&P 500 Total Return
- Small-cap US stocks: vs the Russell 2000 Total Return
- Tech stocks: vs the Nasdaq 100 or the S&P 500 Tech sector
- International: vs the appropriate regional index
This is why our portfolio pages always show the portfolio plotted against the S&P 500, you should be able to see at a glance whether the strategy actually beat the alternative.
Step 5: What to ignore
A list of chart elements that mostly waste time for long-term investors:
Volume bars. Useful for traders. Largely irrelevant if you hold for years.
Moving averages (50-day, 200-day). Useful for traders. The 200-day is occasionally useful as a rough trend indicator, but it's a lagging signal, by the time it tells you anything, the move is mostly over.
Bollinger Bands, MACD, RSI, and most other indicators. Useful for traders looking for short-term patterns. Largely useless if your time horizon is years.
Drawing trend lines. A trend line is a story you're telling yourself about the chart. The chart will look like the story until it stops looking like the story, at which point you'll redraw the line.
Patterns with names. Cup-and-handle. Head-and-shoulders. Double-bottom. There's some academic evidence that some chart patterns have predictive value over very short time windows. There is essentially no evidence they help long-term investors.
What to actually look at
Five things on a stock chart genuinely matter for long-term investors:
- The slope of the line on log scale. This is the actual rate of growth.
- The depth of the worst drawdown. How far did it fall from peak in bad years?
- The recovery time. How long did it take to make new highs after a major drop?
- The relative performance vs the benchmark. Is the line above or below the index over the full window?
- The trend in fundamentals over the same window. Did the chart go up because earnings went up, or because the multiple expanded? (You need a separate chart for this, usually a chart of P/E ratio or earnings over the same period.)
A worked example
Imagine you're looking at a stock chart for Apple over the past 10 years. The price went from roughly $25 to $200, which sounds like a 700% gain.
On a linear-scale chart, the first 7 years look almost flat compared to the recent run. On a log-scale chart, you can see the stock compounded at a remarkably steady rate the whole time.
On a price-return chart, dividends are ignored. On a total-return chart, the actual experience of owning the stock includes ~50% more wealth from reinvested dividends and the small amount of share buybacks shifted to net repurchase.
Plotted against the S&P 500 Total Return on log scale, you can see Apple beat the index over the full window, by what margin, and during which years. That comparison is the one that matters.
What this means for you
When you're evaluating a stock or portfolio:
- Use log scale for any window over 2-3 years
- Use total return, not price return
- Always plot against a relevant benchmark
- Look at the worst drawdown, not just the average return
- Don't draw lines or patterns over the chart
Charts are useful when you read them carefully. Most beginners get burned because they're reading the wrong scale, the wrong return type, or the wrong window.