Tracking error is the standard deviation of the difference between your portfolio’s returns and a benchmark’s returns over time. It tells you how consistently your portfolio follows (or deviates from) a benchmark like the S&P 500. A tracking error of zero would mean your portfolio perfectly mirrors the benchmark every period. In practice, even index funds have some tracking error, and actively managed funds have significantly more.
The Core Formula
The calculation boils down to one line:
Tracking Error = Standard Deviation of (P − B)
P is the portfolio return for a given period, and B is the benchmark return for that same period. The difference between the two, P minus B, is called the “active return” or “excess return.” You calculate this difference for every period in your dataset (monthly returns are the most common), then take the standard deviation of those differences. The result is your tracking error, expressed as a percentage.
If your portfolio returned 1.2% in January and the benchmark returned 1.0%, the active return for January is 0.2%. You repeat that for every month in your timeframe, collect all those active return figures, and compute their standard deviation. That final number captures how much the gap between your portfolio and the benchmark bounces around from period to period.
Step-by-Step Calculation
Here’s how to work through it manually or on paper before moving to a spreadsheet:
- Step 1: Gather return data. You need a series of periodic returns for both your portfolio and the benchmark. Monthly returns over at least 12 months give you a reasonable sample. The longer the series, the more reliable the result.
- Step 2: Calculate active returns. For each period, subtract the benchmark return from the portfolio return. If your portfolio returned 2.5% in a given month and the benchmark returned 2.1%, the active return is 0.4%.
- Step 3: Find the mean active return. Add up all the active returns and divide by the number of periods.
- Step 4: Compute the standard deviation. For each period, subtract the mean active return from that period’s active return, then square the result. Sum all the squared differences, divide by the number of periods minus one (for a sample standard deviation), and take the square root. That number is your tracking error.
Suppose you have 12 months of active returns: 0.3%, −0.1%, 0.5%, 0.2%, −0.4%, 0.1%, 0.6%, −0.2%, 0.3%, 0.0%, 0.4%, −0.3%. The mean is about 0.12%. You’d subtract 0.12% from each value, square the results, sum them, divide by 11 (12 minus 1), and take the square root. The result might be roughly 0.31%, meaning your portfolio’s monthly deviation from the benchmark fluctuates by about 0.31 percentage points.
Annualizing Tracking Error
Monthly tracking error is useful, but the industry standard is to express it on an annual basis so you can compare across funds and strategies. To annualize a monthly tracking error, multiply it by the square root of 12 (approximately 3.46). If your monthly tracking error is 0.31%, the annualized figure is about 1.07%.
If you’re working with weekly returns, multiply by the square root of 52. For daily returns, use the square root of 252 (the approximate number of trading days in a year). The key is matching the scaling factor to the frequency of your return data.
How to Calculate It in a Spreadsheet
A spreadsheet makes this straightforward. Set up three columns: one for portfolio returns, one for benchmark returns, and one for active returns.
In the active return column, each cell simply subtracts the benchmark return from the portfolio return. If your portfolio return is in cell B2 and your benchmark return is in C2, the formula is =B2-C2. Drag that down for every period.
Once you have the full column of active returns, use the STDEV.S function (sample standard deviation) on that range. If your active returns are in cells D2 through D13, the formula is =STDEV.S(D2:D13). That gives you the tracking error for whatever period your data covers. To annualize monthly data, wrap it: =STDEV.S(D2:D13)*SQRT(12).
Use STDEV.S rather than STDEV.P when you’re working with a sample of returns rather than the entire population. In most real-world situations, you’re working with a sample (12 months out of a fund’s life, not every month it will ever exist), so STDEV.S is the appropriate choice.
What the Number Tells You
A low tracking error means the portfolio closely follows its benchmark. Index funds and ETFs that replicate a major index typically have annualized tracking errors well under 0.5%, often in the range of 0.05% to 0.20%. A passively managed total market fund, for instance, deviates very little from its target index.
Actively managed funds, where portfolio managers deliberately pick stocks or time sectors differently from the benchmark, commonly have tracking errors between 4% and 10% or higher. That wider range is the whole point: the manager is intentionally deviating to try to outperform.
The number does not tell you whether the portfolio is outperforming or underperforming. A fund could have low tracking error and still trail its benchmark slightly every month (due to fees, for instance). Tracking error only measures the consistency of the gap, not its direction. To evaluate whether deviations are rewarded, you’d pair tracking error with the information ratio, which divides average active return by tracking error.
Why Tracking Error Exists
Even a fund designed to perfectly replicate an index will have some tracking error. Several structural factors create it:
- Management fees and expenses. Every fund charges an expense ratio, which drags returns slightly below the benchmark. This creates a small, consistent negative active return.
- Cash drag. Funds hold a small cash buffer to handle redemptions. That cash earns a different return than the index, creating a gap. In rising markets, holding cash hurts; in falling markets, it helps slightly.
- Sampling and optimization. Some index funds don’t hold every security in the benchmark. Instead, they hold a representative sample. This approach keeps trading costs down but introduces tracking differences.
- Dividend timing. An index may assume dividends are reinvested immediately, while a real fund receives dividends on specific dates and reinvests them with a short lag.
- Rebalancing costs. When an index adds or removes securities, the fund must trade to match. Transaction costs from those trades create small return differences.
For actively managed portfolios, tracking error is intentional. It reflects the manager’s active bets: sector overweights, individual stock picks, or timing decisions that differ from the benchmark. Higher tracking error in an active fund means the manager is making bolder bets, for better or worse.
Choosing the Right Benchmark
Tracking error is only meaningful if the benchmark actually represents what your portfolio is trying to do. Measuring a small-cap growth fund against the S&P 500 will produce a large tracking error, but that number tells you very little because the comparison is mismatched. The fund was never trying to replicate the S&P 500.
Match your benchmark to the portfolio’s investment universe and strategy. A U.S. large-cap fund should be measured against a U.S. large-cap index. A bond portfolio targeting intermediate-term government debt should use an intermediate government bond index. When the benchmark is appropriate, tracking error becomes a genuine measure of how tightly the portfolio sticks to its intended target, or how aggressively it departs from it.