Rolling Returns: The Consistency Test Every SIF Investor Should Know
Rolling returns usually give a more reliable view of SIF performance than a single point-to-point CAGR because they reduce start-date luck and show how outcomes varied across market conditions. For India’s new SIF category, where strategies may use long-short, derivatives and tactical positioning, consistency, downside behaviour and regime sensitivity matter at least as much as a headline annualised return.
Key Takeaways
- 1CAGR can change sharply with the chosen start date.
- 2Rolling windows show consistency across many entry points.
- 3Percentile bands reveal best, median and weak outcomes.
- 4Risk-adjusted metrics matter more for tactical SIF strategies.
- 5Regime context is essential when interpreting rolling charts.
Why point-to-point CAGR can mislead
A point-to-point CAGR compresses the entire return experience between one start date and one end date into a single annualised figure. That is useful as a summary, but it can be highly sensitive to **where the clock starts**. If the starting point happens to be just after a market correction, the CAGR may look unusually strong; if it begins near a market peak, the same strategy may appear weaker despite little change in underlying process quality. This start-date sensitivity is particularly relevant for **Specialised Investment Funds (SIFs)** because the strategies permitted under SEBI’s February 2024 framework can include long-short exposure, derivatives and tactical allocation. Such approaches may have more path dependency than plain-vanilla long-only funds. A single CAGR can therefore overstate or understate what an investor was realistically likely to experience. In practice, an investor allocating the minimum **INR 10 lakh** to a SIF is rarely interested only in the outcome for one lucky start date. The more relevant question is: *how did the strategy behave for investors who entered in different months or quarters?* That is the gap rolling returns help fill.
What rolling 1-year, 3-year and 5-year windows show
Rolling returns measure the annualised return earned over repeated overlapping periods. A **1-year rolling return** asks: if an investor entered on each possible date and held for 12 months, what return would they have earned? A **3-year rolling return** repeats that exercise for 36-month holding periods, and **5-year rolling returns** do the same for 60 months. This matters because consistency often improves as the holding period lengthens. A tactical equity SIF, for example, may show wide variation in 1-year outcomes because shorter windows are heavily influenced by market noise, hedging costs, and timing decisions. Over 3-year or 5-year windows, one can see whether the process has historically produced a narrower and more dependable range of outcomes. For SIFs, this is often more informative than asking whether the trailing 1-year or trailing 3-year number looks attractive today. A fund with a lower headline CAGR but tighter rolling outcomes may be better suited to investors seeking process stability. Conversely, a strategy with very high upside but highly dispersed rolling outcomes may require stronger risk tolerance and more patience.
Use the distribution of rolling returns, not just the average
Looking only at the average rolling return still misses an important layer: **distribution**. A robust review should examine percentile bands such as the **10th percentile, 25th percentile, median, 75th percentile and 90th percentile** of rolling returns. These bands help answer practical questions. - **Median rolling return**: what a typical holding-period experience looked like. - **Lower percentiles**: how painful weaker entry points were. - **Upper percentiles**: how much upside came from favourable timing. - **Spread between percentiles**: how predictable or variable outcomes were. Consider an illustrative 3-year rolling return profile for a hypothetical equity long-short SIF. Suppose the median 3-year rolling CAGR is **11.8%**, the 25th percentile is **8.1%**, and the 10th percentile is **4.6%**. That tells an investor that while the strategy often compounded in low double digits, weaker entry periods still produced materially lower results. If another strategy has a similar median but a 10th percentile of **-1.5%**, that second strategy may demand a different suitability assessment. For Indian investors and distributors, this percentile approach is especially useful because the SIF category is still new and track records may initially be limited or inherited from pre-existing team processes rather than identical live products. Distribution analysis encourages a more careful reading than a single average return line.
Why drawdown-adjusted metrics matter alongside rolling returns
Rolling returns show consistency across entry points, but they do not fully capture **how returns were earned**. Two SIFs can have the same median rolling return and very different downside profiles. That is why rolling analysis should sit beside drawdown-aware measures such as **Sharpe ratio, Sortino ratio and Calmar ratio**. The **Sharpe ratio** evaluates excess return relative to total volatility. It is widely used, but for SIFs that use hedging or asymmetric payoff structures, total volatility may not fully reflect investor discomfort. The **Sortino ratio** can be more useful because it focuses on downside volatility rather than all volatility. The **Calmar ratio** compares return to maximum drawdown, making it especially relevant for tactical or leveraged exposures where drawdown depth matters. A practical way to combine these tools is straightforward. First, look at rolling 1-year, 3-year and 5-year distributions. Then ask: were the returns achieved with shallow or severe drawdowns? An illustrative example: Strategy A and Strategy B both show a median 3-year rolling CAGR near **12%**. But if Strategy A’s maximum drawdown was **-9%** and Strategy B’s was **-22%**, the Calmar profile would likely favour Strategy A. If Strategy B also exhibits more negative months below the minimum acceptable return threshold, the Sortino ratio may further weaken the case. For SIF evaluation, especially given the wider strategy toolkit permitted under SEBI’s framework, these adjusted metrics are not optional add-ons. They help distinguish between skill, leverage, timing luck and simple risk-taking.
Regime matters: rolling windows are not regime-neutral
Rolling windows improve on point-to-point CAGR, but they are not magic. They still reflect the market regimes embedded in the sample. A long-short equity SIF may look strong if most 3-year windows capture broad market trends with contained volatility. The same strategy may look less impressive if the sample includes prolonged sideways markets, sharp bear phases, or abrupt factor reversals. This is why investors should map rolling periods to **regimes**: bullish, bearish, high-volatility, low-volatility, liquidity-driven rallies, earnings-led markets, and rate-shock periods. In India, shifts in domestic flows, global risk appetite, currency conditions and interest-rate expectations can materially alter the opportunity set for tactical and derivative-heavy strategies. A practical implication follows. If a SIF’s strongest rolling returns came mainly from one favourable regime, investors should be cautious about extrapolating them. Conversely, if rolling outcomes remain respectable across multiple distinct environments, that suggests a more adaptable process. For a young category like SIFs, where live product histories may still be developing, regime attribution is a critical part of due diligence. Investors should also remember that overlapping rolling windows are statistically related to each other. They improve interpretability but do not create independent observations. So the goal is not false precision; it is a fuller understanding of **range, persistence and regime dependence**.
How to read a rolling return chart in practice, with an illustrative example
A rolling return chart is best read in layers rather than as a single line. Start with the **central tendency**: where does the median or main line spend most of its time? Next, look at the **bandwidth**: are the gaps between upper and lower percentile bands narrow or wide? Then identify **stress zones**: when did lower percentile outcomes compress or turn negative? Finally, connect those periods to market regime and drawdown history. Here is an illustrative example using hypothetical numbers for a SIF strategy versus a simple long-only benchmark: - **Point-to-point 5-year CAGR ending March 2026** - SIF Strategy: **13.2%** - Benchmark: **12.4%** On the surface, the SIF appears modestly better. But now examine rolling periods: - **1-year rolling returns** - SIF: range **-8% to 24%**, median **10%** - Benchmark: range **-15% to 28%**, median **11%** - **3-year rolling returns** - SIF: 10th percentile **5%**, median **11.8%**, 90th percentile **15.4%** - Benchmark: 10th percentile **1%**, median **11.5%**, 90th percentile **18.2%** - **5-year rolling returns** - SIF: range **8.7% to 13.6%** - Benchmark: range **6.1% to 14.8%** - **Maximum drawdown** - SIF: **-10%** - Benchmark: **-19%** This hypothetical pattern suggests the SIF did **not** necessarily maximise upside in the best periods; the benchmark had a higher 90th percentile over 3-year windows. But the SIF offered tighter dispersion, a stronger lower tail, and materially lower drawdown. That may make it more attractive for investors who value smoother compounding over headline peak returns. In practical reading, a good rolling chart for a SIF is not simply one with the highest line. It is one where: the lower percentile bands hold up reasonably well, the median remains competitive, negative rolling periods are limited or understandable, and risk-adjusted metrics confirm that performance was not bought through disproportionate drawdowns. That is why rolling returns are usually the better lens for evaluating SIFs than a single point-to-point CAGR.
Glossary
FAQ
Educational content. Not investment advice. Matched based on your profile inputs.