How to read tracking error

By Navprabhat K·Updated 28 July 2026·8 min read

Two index funds tracking the same index are not the same product. Tracking error is how you tell them apart, as long as you read it alongside the number most factsheets bury.

What tracking error measures

Tracking error is the standard deviation of the difference between a fund's returns and its index's returns, usually computed from daily data and annualised. In plain terms, it measures how consistently a fund follows its index.

A fund with near-zero tracking error moves in lockstep with the index day after day. A fund with high tracking error wanders: ahead one week, behind the next. Because it is a standard deviation, it is always positive, and it treats overshooting the index and undershooting it as equally undesirable. That last point matters more than it sounds, and it is where most readings go wrong.

Tracking error versus tracking difference

These two get used interchangeably in fund commentary. They are not the same thing, and the one that affects your money is usually the one that gets less attention.

Consider two funds tracking the same index over a year, where the index returned 15.0%:

• •Fund A returned 14.85%. It lagged by a steady, predictable amount every day. Tracking difference is 0.15%. Tracking error is very low.
• •Fund B also returned 14.85%, but it swung between 0.6% ahead and 0.6% behind through the year before landing in the same place. Same tracking difference, much higher tracking error.

You ended up with identical money in both. Tracking error alone would have told you Fund A was clearly better, which is only true in the sense that Fund B's process is less controlled and its future outcome is therefore less predictable. Tracking difference is what hit your account.

What SEBI requires of passive funds

SEBI's 2022 framework for passive funds put real limits on how far an Indian index fund or ETF is allowed to drift, and required AMCs to disclose the numbers rather than leave investors to work them out.

• •A cap on tracking error. For equity index funds and ETFs, annualised tracking error based on rolling one-year data is required to stay below 2%. This is a regulatory outer limit, not a target. A well-run Nifty 50 fund should be nowhere near it.
• •Mandatory disclosure. AMCs must publish tracking error and tracking difference for their passive schemes on their websites and on AMFI, so you can check rather than guess.
• •Tracking difference limits for debt passive products, where tracking error is a less meaningful statistic because of how debt indices are constructed.
• •A 1.00% TER cap on index funds and ETFs, which indirectly constrains tracking difference, since cost is its largest single component.

Rules do get revised. Treat these as the shape of the framework and confirm the current specifics on sebi.gov.in if you are making a decision that turns on them.

Why a fund drifts from its index

An index is a mathematical construct with no costs. A fund is a real portfolio that has to trade, settle and hold cash. The gap between the two comes from a handful of identifiable sources.

Expense ratio

The largest and most predictable contributor. A fund charging 0.20% starts every year 0.20% behind the index, before anything else happens. This is why tracking difference for a well-run fund tends to land close to its TER, and why the expense ratio is the first thing to check.

Cash drag

Funds hold some cash to meet redemptions and to handle incoming SIP flows before they are deployed. Cash does not participate in a rising market, so in an up year cash drag pulls the fund below the index. In a falling market it does the reverse, which is one reason a fund can briefly appear to beat its index.

Dividends and reinvestment timing

A Total Return Index assumes dividends are reinvested the instant they go ex-dividend. A real fund receives the cash days later and then has to deploy it. In a rising market, that delay costs a little return every time.

Rebalancing and index changes

When an index adds or drops a constituent, the fund must trade. Everyone tracking that index trades in the same direction on the same day, so execution is expensive. Larger, better-run funds manage this transition more skilfully, and it is one of the clearest places where operational quality shows up in the numbers.

Flows

Large inflows or redemptions force trading that the index never has to do. A fund with lumpy, unpredictable flows will track less tightly than one with a stable investor base.

Sampling

Broad indices with hundreds of illiquid constituents are sometimes tracked by holding a representative subset rather than every name. This lowers trading costs and raises tracking error. It is a legitimate trade-off, but it is a reason a Nifty 500 fund will never track as tightly as a Nifty 50 fund.

What a healthy number looks like

International feeder funds deserve a note. They tend to show large tracking error for structural reasons: currency movement, different market hours, local regulatory limits on overseas investment, and an extra layer of fees in the underlying fund. A high figure there is not automatically a sign of a badly run fund, but it does mean your return can diverge substantially from the index you thought you were buying.

How to compare two funds properly

1. Same index. Comparing a Nifty 50 fund's tracking error with a Nifty Next 50 fund's tells you about the indices, not the funds. The Next 50 is harder to track by construction.
2. Same period. Tracking error rises when markets are volatile. A figure from a calm year and one from a turbulent year are not comparable.
3. Same plan. Direct and regular plans of the same scheme have different TERs and therefore different tracking differences. Make sure you are reading the one you would buy.
4. Same data frequency. Tracking error computed from daily returns is a larger number than the same fund's figure computed monthly. AMCs generally follow the prescribed method, but aggregator sites are less consistent.
5. Check tracking difference alongside it. Then sanity-check it against the TER, as described above.

The ETF costs tracking error does not capture

ETFs often advertise lower expense ratios than equivalent index funds. That comparison is incomplete, because an ETF adds costs that never appear in tracking error, since tracking error is calculated from NAV rather than from the price you actually paid.

• •Bid-ask spread. You buy at the ask and sell at the bid. On a thinly traded Indian ETF this spread can be wide enough to wipe out several years of TER savings in a single round trip.
• •Premium and discount to iNAV. An ETF's market price can drift from the value of its underlying holdings, particularly when liquidity is poor or markets are moving fast. Buying at a premium is an instant, invisible loss.
• •Brokerage, STT and demat charges on every transaction, which matter a lot if you are investing monthly.
• •Liquidity risk at exit. A low-volume ETF can be difficult to sell in size at a fair price precisely when you want out.

For a monthly SIP, a plain index fund is usually the better instrument even at a slightly higher TER, because you transact at NAV and none of the above applies. ETFs make more sense for large lump sums in genuinely liquid products where you can control your execution.

A five-minute check before you buy

1. Open the AMC's page for the scheme and find the direct plan's TER.
2. Find the disclosed tracking difference over one year, and over three years if the fund is old enough.
3. Compare tracking difference against TER. Roughly similar is healthy. Much larger means something else is leaking.
4. Check tracking error against the ranges above, and against other funds tracking the same index over the same window.
5. Check fund size. Very small passive funds tend to track worse and cost more.
6. If it is an ETF, check average daily traded volume and the current spread before assuming the low TER is real.

Related reading: expense ratio, explained covers the largest component of tracking difference, and index funds versus active funds in India works through where passive makes sense in the first place. Fund Explorer surfaces tracking error next to cost and returns on every fund it ranks, so you are not comparing one number in isolation.