ETF tracking error vs tracking difference: what each metric tells you
Two ETFs can follow the same index and still deliver different investor outcomes. Tracking difference measures the return gap; tracking error measures how consistently the ETF stays near its benchmark.
Direct answer
Direct answer
Tracking difference measures the return gap between an ETF and its benchmark over a period. Tracking error measures the variability of those return gaps. A passive ETF can have a small average gap but still track inconsistently, so the two metrics answer different questions and should be read together.
Key takeaways
- • Tracking difference is a return-gap measure; tracking error is a consistency measure.
- • For Indian index funds, NSE describes tracking error against the Total Returns Index, which includes dividends.
- • Expense ratio influences tracking, but cash, index changes, corporate actions and transaction costs also matter.
- • Low tracking error does not guarantee a small return shortfall; a fund can lag consistently.
- • ETF execution still matters: bid-ask spread and premium/discount to NAV can create costs that tracking statistics do not capture.
DeepScreen original framework
What this guide adds
The DeepScreen Tracking Matrix separates four ETF behaviours — efficient and consistent, consistently costly, erratic but near benchmark, and both costly and erratic — so two similar-looking tracking numbers become an actionable diagnostic.
Tracking difference and tracking error are not the same thing
Tracking difference asks 'how far did the ETF finish from the index?' Tracking error asks 'how much did that gap wobble along the way?'
NSE defines tracking error as the annualized standard deviation of the difference in returns between an index fund and its target index. NSE also says the calculation should use the Total Returns Index, which includes dividends. [1]
Fidelity and ETF.com distinguish tracking difference from tracking error in the same way: tracking difference is the performance gap, while tracking error describes variability in that gap. [2][3]
| Metric | What it measures | Simple interpretation | Useful question |
|---|---|---|---|
| Tracking difference | Fund return minus benchmark return over a period | Closer to zero means the realized return stayed nearer the benchmark | How much return did the ETF gain or lose relative to the index? |
| Tracking error | Annualized standard deviation of periodic fund-minus-index return differences | Lower means the relative return gap was more consistent | How reliably did the ETF stay near the benchmark? |
How do you calculate tracking difference?
Use the same period, currency and return convention for the ETF and benchmark, then subtract benchmark return from ETF return.
DeepScreen uses the sign convention: tracking difference = ETF total return − benchmark total return. If an ETF returns 11.80% while its benchmark Total Returns Index returns 12.00%, the tracking difference is −0.20 percentage point.
Always check the sign convention used by a factsheet or data vendor. Some presentations quote the shortfall as a positive number, while others preserve the negative sign for underperformance. The economic meaning is more important than the display convention.
How do you calculate tracking error?
Calculate periodic ETF-minus-index return differences, take their standard deviation and annualize it using a frequency-appropriate factor.
The important point is conceptual: tracking error does not tell you whether the ETF is ahead or behind the index. It tells you how variable the relative return has been. [1]
That is why an ETF can have very low tracking error while delivering a persistent negative tracking difference. If it trails by nearly the same amount every period, the tracking path is consistent even though the long-run shortfall is real.
The DeepScreen Tracking Matrix: read both metrics together
Plot the size of the return gap against the variability of that gap. The combination is more informative than either number alone.
This matrix is a research framework, not a rating system. 'Small' and 'high' must be judged against ETFs tracking the same or a very similar benchmark over the same measurement period.
| Tracking difference | Tracking error | What it suggests | What to investigate |
|---|---|---|---|
| Small | Low | Efficient and consistent replication | Confirm liquidity, costs and data period |
| Large negative | Low | Consistently lagging benchmark | TER, taxes, cash drag, replication structure |
| Small average | High | Ends near benchmark but gets there erratically | Rebalancing, cash flows, constituent changes, derivatives |
| Large negative | High | Both costly and inconsistent tracking | Fund structure, execution, liquidity and benchmark fit |
Example: two ETFs can have the same annual gap but different tracking quality
A similar year-end tracking difference can hide very different paths.
ETF A has the same small relative shortfall each quarter, so its tracking error would be low. ETF B finishes with a similar approximate annual shortfall but swings around the benchmark much more, so its tracking error would be higher. The example is hypothetical and exists only to show why the two metrics are not interchangeable.
| Quarter | ETF A vs index | ETF B vs index |
|---|---|---|
| Q1 | −0.05 pp | +0.10 pp |
| Q2 | −0.05 pp | −0.30 pp |
| Q3 | −0.05 pp | +0.15 pp |
| Q4 | −0.05 pp | −0.15 pp |
| Approx. annual sum | −0.20 pp | −0.20 pp |
What causes tracking error and tracking difference?
Fees are only one cause. Portfolio implementation, cash, flows and index events can all move an ETF away from its benchmark.
NSE specifically highlights inflows and outflows, corporate actions, changes in index constituents, liquidity cash and transaction costs as drivers of tracking error. [1]
- • Expense ratio and other recurring fund costs.
- • Bid-ask spreads and transaction costs when the fund trades holdings.
- • Cash held for liquidity or pending subscriptions and redemptions.
- • Index constituent changes and the timing of rebalancing trades.
- • Corporate actions and implementation differences.
- • Sampling or alternative replication methods instead of holding every index constituent at exact weight.
- • Taxes, securities-lending revenue or derivative implementation where relevant to the fund.
Why tracking statistics are not enough for an ETF trade
Tracking measures the fund against its benchmark; it does not fully capture the price you personally pay in the market.
Investor.gov notes that ETF shares trade at market prices that can be above or below NAV. It also explains that the bid-ask spread is a transaction cost and that more liquid, higher-volume ETFs typically have tighter spreads. [4]
This creates two separate layers of due diligence: fund efficiency and trade execution. A well-tracking ETF can still be expensive to enter or exit if the spread is wide or the market price is far from NAV.
How to compare two ETFs tracking the same index
Use the same benchmark and measurement window, then compare tracking, cost, concentration and execution in that order.
- • Confirm both ETFs really track the same index version, ideally the same Total Returns Index.
- • Compare multi-period tracking difference, not just the latest one-year number.
- • Compare tracking error over the same frequency and window.
- • Check expense ratio, but do not assume the lowest fee will always have the smallest realized tracking difference.
- • Review AUM, normal trading volume, median spread and premium/discount history.
- • Check whether one ETF uses sampling, derivatives or a different replication method.
- • Re-check after major index rebalances or when fund size/liquidity changes materially.
Common tracking-metric mistakes
Most mistakes come from comparing incompatible numbers or treating one metric as a complete quality score.
- • Comparing an ETF against a price index while the fund benchmark is a Total Returns Index.
- • Comparing one-year tracking difference for one fund with three-year tracking error for another.
- • Thinking low tracking error means low cost.
- • Ignoring the sign convention on tracking difference.
- • Ignoring spread and premium/discount because they are not part of published tracking error.
FAQ
Common questions
- Which is more important: tracking error or tracking difference?
- They answer different questions. Tracking difference shows the realized return gap from the benchmark; tracking error shows the consistency of that gap. For a passive ETF, read both together and compare them over the same period and benchmark.
- Is lower tracking error always better?
- Lower tracking error means more consistent benchmark-relative performance, but it does not guarantee a small return shortfall. An ETF can lag its benchmark by a similar amount every period and still report low tracking error.
- Can tracking difference be positive?
- Yes. Under the fund-return-minus-index convention, a positive tracking difference means the ETF outperformed the benchmark over the measured period. Securities lending, implementation timing and other effects can sometimes offset part of the fee drag.
- Does expense ratio equal tracking difference?
- No. Expense ratio is one driver of tracking difference, but cash drag, transaction costs, taxes, rebalancing, sampling and securities-lending income can make the realized return gap larger or smaller than the headline fee.
- What benchmark should an Indian index ETF use for tracking?
- NSE states that tracking error should be calculated against the Total Returns Index because it includes dividends. Always verify the benchmark named in the ETF's own scheme documents.
Continue your research
Sources
References
- [1] Tracking Error · National Stock Exchange of India (NSE) · accessed October 2026. Primary/source page
- [2] Understanding tracking error and tracking difference for an ETF · Fidelity · accessed October 2026. Primary/source page
- [3] Tracking Difference / Tracking Error · ETF.com · accessed October 2026. Primary/source page
- [4] Updated Investor Bulletin: Exchange-Traded Funds (ETFs) · Investor.gov / U.S. SEC Office of Investor Education and Advocacy · accessed October 2026. Primary/source page
Editorial disclosure
DeepScreen is a financial-research platform. This article is educational and is not personalized investment, tax or legal advice. Product data, regulations and fund disclosures can change; verify the latest issuer and regulator material before acting.
Author: Sooraj, Founder of DeepScreen. Facts were checked against the cited regulator, exchange, industry-association and issuer sources on 2 October 2026. No independent credentialed reviewer has been claimed.