What Is Tracking Error in an ETF? A Practical Guide (2026)

Tracking error in an ETF is the annualized standard deviation of the gap between the fund’s returns and the returns of the benchmark index it is built to follow. It tells you how consistently the fund shadows that index over time, not how much return it gives up. Lower numbers mean a steadier, more predictable match.

An index fund is not an exact photocopy of its index. It holds real securities, pays real fees, and occasionally holds cash while the market moves. Those small frictions add up, and tracking error is the statistical measure of how badly the fund wanders from the line it is supposed to sit on. Most investors never check the figure, then wonder why two funds tracking the “same” index ended up in different places.

This guide covers how the number is calculated, what pushes it up, and where to find it before you buy. Tracking error is a risk and quality measure, not a performance measure, and reading it wrong is the single most common mistake I see. Fund rules, fee structures and index behaviour change over time, so treat everything here as a framework to apply to the specific fund in front of you rather than a fixed rule.

What Is Tracking Error in an ETF?

Tracking error in an ETF is the annualized standard deviation of the difference between the fund’s returns and its benchmark’s returns. In plain terms, it measures how erratically the fund strays from the index it tracks. A fund with 0.4% tracking error stays close to its index day after day. A fund with 4% can wander several percentage points away in any single stretch of bad luck.

Three things about that definition matter more than they look.

• It is a spread, not a distance. Tracking error describes how much the fund’s results bounce around its average gap, not how far the average gap sits from zero. A fund can consistently trail its index by two percentage points a year and still have a tiny tracking error.
• It is never negative. Standard deviation has no sign. If you have read that negative tracking error means a fund outperformed, that is confusing tracking difference with tracking error.
• It grows with time. Dividing the standard deviation by the number of observations leaves you with the average deviation. The reported figure is scaled up to a yearly basis.

The confusion here is worth spelling out. Your ETF’s return is the money it made. Its tracking difference is the annualised gap between that return and the index’s return. Its tracking error is the volatility of that gap. Two funds can post the same tracking difference and very different tracking errors, and one fund can have a modest tracking difference with a wild tracking error.

What Does an ETF Track?

An ETF tracks a benchmark index, and the index is just a rulebook for what to hold. A widely followed large-company index might tell a provider to hold every eligible constituent in proportion to its market weight. The fund’s job is to own those same securities so that its own performance mirrors the index’s.

Three pieces sit between that rulebook and your account:

• The index — a published list of securities with weights, rebalanced on a set schedule.
• The holdings — what the fund actually owns, which may not be identical to the list if it uses sampling.
• The fund’s returns — the outcome after fees, trading costs, taxes on dividends and everything else that happens between a trade and your account.

Many funds do not buy every single constituent. Sampling means holding a representative slice of a large index, which cuts transaction costs and can slightly lower tracking error because fewer trades are needed. Optimisation goes further, keeping the exposure that explains most of the index’s movement while trimming the rest. Both are deliberate deviations from exact replication, and both can pull the fund slightly off its benchmark line.

There is also a gap between what the fund owns and what you pay. The fund’s net asset value is calculated once a day using closing prices. The price you trade at moves continuously, so an ETF can sit a little above or below its net asset value for hours at a time. That premium or discount is a separate issue from tracking error, even though both show up as a short-term divergence from the index.

How Is Tracking Error Measured?

Measurement starts with one simple subtraction, done once per day or per month: the fund’s return minus the benchmark index’s return. Take the series of those differences and compute the standard deviation. Multiply it by the square root of the number of periods in a year to annualise. That final number is the tracking error the fund reports.

Measure	What it answers	Direction	Typical use
Tracking difference	On average, how far did the fund land from the index?	Can be positive or negative	Explaining a multi-year performance gap
Tracking error	How erratically did the fund bounce around that average gap?	Always zero or positive	Measuring fund quality, process consistency and hidden risk
Beta	On average, did the fund move with the market?	Can be below or above 1	Measuring directional sensitivity

What is tracking error in an ETF, in numbers?

Here is an illustrative monthly example. The figures are made up to show the mechanics, but the arithmetic is exactly what a spreadsheet does.

Month	Benchmark return	ETF return	Difference (percentage points)
1	2.10%	1.65%	-0.45
2	-3.40%	-3.20%	+0.20
3	1.25%	1.05%	-0.20
4	-2.80%	-2.25%	+0.55
5	0.90%	0.55%	-0.35
6	4.20%	4.25%	+0.05
7	-5.10%	-5.70%	-0.60
8	1.80%	2.15%	+0.35

The average of those eight differences is about -0.06 points a month, or roughly -0.7 points a year. That is the tracking difference, and the sign is negative because fees ate into the return.

The standard deviation of the same column comes out near 0.38 percentage points. Annualising with monthly data gives 0.38 multiplied by the square root of 12, which lands at about 1.3%. That is the tracking error.

The takeaway is that both numbers came from the same column of differences. One describes the average, the other describes the wobble around it. In a spreadsheet, add a row that subtracts each index return from the matching fund return, run STDEV.S over that range, and multiply the result by SQRT(12) for monthly data or SQRT(252) for daily data. Using daily data gives a cleaner read because it captures the path, not just the endpoints.

Two variants show up in fund documents. Ex-post tracking error uses realised past returns, so it describes what the fund has actually done and can spike during a crisis when everyone sells at once. Ex-ante tracking error is estimated from the portfolio’s characteristics, a forecast rather than a record. Some providers use both, which is why two data sources can quote slightly different figures for the same fund and the same period.

What Causes Tracking Error?

Roughly nine things push a fund away from its index, and they do not all hit the two measures in the same way. A cost that is constant every month drags tracking difference down without making tracking error any worse. A cost that varies with market conditions does both.

1. Management fees and fund expenses. The expense ratio is charged every day whether markets rise or fall, so it lowers the tracking difference but barely touches tracking error. It is the most predictable drag and the easiest one to compare.
2. Cash drag. Every fund holds some cash for creations, redemptions and expenses. That cash earns less than the index constituents when markets climb, so it quietly subtracts from returns. It is also a common cause of elevated tracking error, because cash moves unpredictably relative to a falling market.
3. Transaction and rebalancing costs. When an index changes its holdings, the fund must trade to match. Spreads and commissions paid during those trades vary with conditions, which shows up as both a gap and extra volatility around the gap.
4. Taxes on dividends and sales. Withholding tax on foreign dividends, and capital gains realised inside the fund, reduce returns in ways that fluctuate by period and by country. For a fund holding overseas equities this is often the largest single cause.
5. Sampling and portfolio optimisation. A fund that does not hold every constituent will not move in exact lockstep with the index. This is a deliberate trade of a small, steady deviation for lower costs.
6. Index changes. A reconstitution or reweighting alters the index itself. If the fund cannot trade at the same instant the index updates, the timing gap lands in the fund’s return.
7. Securities lending. Some lenders pass a share of the income from lending out stocks to the fund. That income can offset expenses and push the tracking difference positive. Discussion on investor forums about a fund beating its index usually traces back to exactly this line.
8. Currency hedging. A fund hedging against currency movement pays forward points and rebalances hedges as rates shift. Both costs vary over time, so hedging tends to widen tracking error rather than just lower returns.
9. Futures roll. A fund holding commodity futures must sell an expiring contract and buy the next one. If the market sits in contango, that roll quietly costs money every month; in backwardation it helps. This is the main reason a commodity fund can drift far from its own index for structural reasons rather than manager error.

Leveraged and inverse products add one more. They reset their exposure every single day to hit a daily multiple, so a multi-day result diverges from the same multiple applied over the whole period. Path dependence shows up as a large and lumpy tracking error figure.

How Do You Compare ETFs Using Tracking Error?

Comparing two funds on tracking error only works if they are chasing the same benchmark. A realistic five-step process:

1. Confirm the benchmark matches

Read the fund’s stated index and make sure the competitor tracks the same one. A fund following a real, published index is directly comparable. A fund tracking a synthetic or custom index may report a lower figure simply because the index was built to match what the fund already does.

2. Find the number and its period

Issuer fact sheets and quarterly reports usually publish it in a standard risks and performance table. Look for the label beside it, because “standard deviation”, “tracking error” and “tracking difference” get mixed up in some fact sheets. If the period is not stated, assume the worst case and assume it is a short window.

3. Compare like for like on fees

Two funds on the same index with very different expense ratios will have very different tracking differences even if their tracking errors look similar. Compare the fee first, then use tracking error to judge everything the fee does not explain.

4. Read the figure in context

Big inflows and outflows distort tracking error. When money floods into a fund, it holds a lot of cash until it can deploy, which widens the gap and the wobble around it. Crisis periods do the same by forcing sales. A single quarter after a shock tells you about market stress, not about the fund’s normal behaviour.

5. Decide whether the drift matches the fund type

Broad equity index trackers, commodity futures funds and leveraged products have completely different normal ranges. Judging a futures-based commodity fund against a large-company index tracker is like comparing a delivery van to a sports car.

If you want to go further, screen several funds side by side in a research terminal or a screening tool rather than pulling fact sheets one at a time. Investors who do this tend to find that fund size, spreads and lending income explain as much as the tracking error figure itself.

Tracking Error vs. Expense Ratio: What Is the Difference?

The expense ratio is a price you pay. Tracking error is a measure of how unevenly the fund delivers what you bought. They can move independently, which is why looking at only one of them is a mistake.

Aspect	Expense ratio	Tracking error
What it is	Annual charge taken from fund assets, shown as a percentage	Annualised standard deviation of the fund’s gap to its index
Direction	Always a cost, never negative	Always positive
Main effect	Lowers tracking difference almost one-for-one	Adds risk and unpredictability around the gap
What drives it	Management fee, custody, admin, index licence	Cash, taxes, trading costs, sampling, hedging, roll
How to read it	Compare across any two funds	Compare only within the same fund type

A fund with a high fee and tiny tracking error has one clear problem: a predictable cost. A fund with a low fee and wide tracking error may be cheaper on paper but deliver its exposure less consistently. Which of those matters more depends on the job the fund is doing for you, and on whether you are trying to match an index or to beat one.

Is a Low Tracking Error Always Better?

Not always. A fund designed to deviate from its index should show a higher figure, and an investor who bought that deviation should not treat it as a defect. Active funds run a portfolio on purpose, so a wide tracking error is the honest signature of the manager’s decisions. Penalising those funds for having active risk would mean penalising the reason you bought them.

The bands below are rough, for orientation rather than for judging a specific fund. Individual funds sit outside them all the time, which is why the fund’s own published figures win.

ETF type	Rough annual tracking error band	Why it sits there
Broad large-company index	Well under 1%	Deep liquidity, tight spreads, straightforward replication
Bond and fixed income	Often lower than equity equivalents	Smoother index returns mean smaller deviations
International equity	Moderate, often wider than domestic	Withholding taxes, currency effects, time-zone gaps
Sector and thematic	Wider than broad market	Fewer constituents, more sampling, heavier rebalancing
Commodity futures based	Several percent or more	Roll cost in contango or backwardation, collateral returns
Leveraged and inverse	Substantially higher, and uneven over time	Daily reset, compounding, path dependence

Investors raised this on finance forums repeatedly and the recurring points are worth keeping. Tracking error is a risk measure, never negative, and a fund can post a wide figure for one quarter for reasons that have nothing to do with its manager. The same ETF listed in another country can show different numbers because currency and withholding tax sit outside the fund.

So the practical answer: low tracking error matters most when your plan depends on getting a specific exposure at a predictable cost. It matters far less when you deliberately want deviation. Either way it should never be read on its own, and it should never be the first thing you look at.

Key Takeaways for ETF Investors

If you do nothing else, do these four things when you next open a fund fact sheet or issuer page.

• Check the benchmark index first. Everything else is meaningless until you know what the fund is chasing.
• Look for both figures side by side. The tracking difference explains why the fund’s returns differ, and the tracking error tells you how erratic that gap has been.
• Confirm the measurement period and the date. A figure from a short window that ended during a market shock tells you almost nothing about normal behaviour.
• Read the fee and the tracking error together. A predictable cost and an unpredictable one are different problems with different fixes.

Where the fund holds futures, lends securities or hedges currency, expect the number to be higher and check the prospectus for how it replicates. Those are decisions the fund has made on purpose.

Frequently Asked Questions

Conclusion

Start with the index, then read the tracking difference and the tracking error side by side, then check the measurement period. Most broad index ETFs will show a small, steady figure and nothing worth worrying about.

The figure that deserves attention is a wide one on a fund that claims to be a plain tracker, or a sudden jump that persists across several reporting periods with no explanation in the prospectus. That combination usually points at cash management, lending policy or a change in how the index itself is built, and it is worth a conversation with the issuer before you add to the position.

This is general information about fund statistics, not personal investment advice. Fees, index rules and fund structures change over time, so verify the current figures on the issuer’s own materials before you decide anything.

Source: https://www.pgm-blog.com/what-is-tracking-error-in-an-etf/