Small Cap funds are known for sharp swings. A useful question is how that changes as the holding period gets longer. We looked at lump-sum rolling windows for Small Cap Direct Growth schemes and counted how often the annualised return was negative.
Key Takeaways
- Loss odds fell at each longer horizon: 19.46% (1Y), 5.73% (3Y), 2.79% (5Y), 0.25% (7Y) and 0.0% (10Y).
- The zero-loss horizon in this dataset is 10 years. No shorter horizon tested was loss-free.
- Median CAGR was similar at long horizons: 19.28% (7Y) and 19.13% (10Y).
- Worst cases improved with time: -2.72% for the worst 7Y window and +3.37% for the worst 10Y window.
- The 10Y sample is narrow: 14,686 daily rolling windows across 12 Direct Growth schemes, with start dates only from 2013 to 2016.
What We Measured
Each statistic is a lump-sum, point-to-point annualised return (CAGR) for a rolling window. A rolling window starts on a given day and ends after 1, 3, 5, 7 or 10 years. The windows step forward daily, and the results are pooled across schemes.
The data covers Direct plan, Growth option series, with window start dates from 2013-01-01. That is the date from which Direct plans have NAV history. The universe is currently active schemes only. Closed or merged schemes are excluded, which is a survivorship-bias limitation. The data is internally validated.
Loss Odds by Holding Period
| Horizon | Daily rolling windows | Direct Growth schemes | Loss probability | Median CAGR |
|---|---|---|---|---|
| 1Y | 79,852 | 31 | 19.46% | 14.71% |
| 3Y | 59,156 | 24 | 5.73% | 22.01% |
| 5Y | 43,163 | 21 | 2.79% | 20.2% |
| 7Y | 29,217 | 17 | 0.25% | 19.28% |
| 10Y | 14,686 | 12 | 0.0% | 19.13% |
The scheme count shrinks as the horizon grows because fewer schemes have enough history for longer windows. The window count is the number of daily rolling windows pooled across those schemes.
At 1 year, 19.46% of the 79,852 windows ended with a negative annualised return. By 3 years the figure was 5.73%, and by 7 years it was 0.25%. At 10 years, none of the 14,686 windows did.
How Bad Were the Worst Windows?
| Horizon | Worst window | 5th percentile | Median | 95th percentile |
|---|---|---|---|---|
| 1Y | -44.61% | -13.44% | 14.71% | 92.37% |
| 3Y | -23.62% | -1.19% | 22.01% | 41.96% |
| 5Y | -13.99% | 2.7% | 20.2% | 36.36% |
| 7Y | -2.72% | 6.17% | 19.28% | 26.34% |
| 10Y | 3.37% | 8.12% | 19.13% | 26.38% |
Over 1 year, outcomes ranged widely, from -44.61% in the worst window to 222.83% in the best. At 10 years, the range was narrower, from 3.37% to 31.8%. The standard deviation of returns was 31.27 at 1Y and 4.77 at 10Y.
Loss probability counts how often a window lost money, not how much. A 1Y window that lost money could still be a deep loss, as the worst-window figure shows.
Did Returns Beat Inflation?
The snapshot uses 6.0% as an inflation threshold, the RBI CPI upper tolerance band. The share of windows with annualised returns above that threshold was 68.1% at 1Y, 88.91% at 3Y, 91.27% at 5Y, 95.27% at 7Y and 98.25% at 10Y.
Start Year Matters
Pooled numbers hide how results depended on when a window began. For 1-year windows, loss probability by start year was 0.0% for 2020 and 2023, but 79.51% for 2018 and 39.38% for 2024. For 3-year windows starting in 2017, it was 51.23%.
At 10 years, windows starting in each of 2013, 2014, 2015 and 2016 showed 0.0%. Daily windows overlap heavily, so these are not independent outcomes.
Why the 10-Year Result Needs Care
A 0.0% loss probability is a statement about this sample, not a prediction. Several limits apply:
- Few start dates. The 10Y windows begin only between 2013-01-01 and 2016-10-01.
- Small sample. 12 Direct Growth schemes contribute to the 10Y result.
- Survivorship bias. Only schemes still active are included.
- Lump-sum only. The study says nothing about other ways of investing over time.
- Other holding periods. Other holding periods were not tested.
- Category label. It reflects current SEBI classification, and earlier windows use today's membership.
Past performance is not indicative of future results.
Frequently Asked Questions
What was the loss probability for Small Cap funds over 1 year?
In 79,852 daily rolling windows across 31 Direct Growth schemes, 19.46% ended with a negative annualised return. The worst 1Y window returned -44.61%.
Which holding period had no losses in the data?
The 10-year horizon. Across 14,686 daily rolling windows across 12 Direct Growth schemes, loss probability was 0.0% and the lowest return was +3.37%. The 7-year horizon was close, at 0.25%.
Does 0.0% loss probability at 10 years mean a loss is impossible?
No. It describes the observed sample only. The 10Y windows start between 2013 and 2016, include 12 schemes and cover currently active schemes only, so the result may not hold in other periods.
What was the median return at long horizons?
The median annualised CAGR was 19.28% at 7 years and 19.13% at 10 years. Those are medians of lump-sum, point-to-point windows, not expected future returns.
Do these figures apply to periodic investing?
No. Every statistic is a lump-sum, point-to-point return. The study did not test other investing patterns.
How were closed or merged schemes handled?
They are excluded. The universe is currently active schemes only, so survivorship bias applies and loss probabilities may look better than they would with those schemes included.
This is educational content, not investment advice; consult a SEBI-registered investment adviser.