ToolsSIP check
How bad can a SIP get?
A Nifty 500 SIP a year old was below the money put in for 26% of start months; at three years, 8%. The worst start fell 50% below its own instalments. Being under water early is normal, because a young SIP has had little time to grow. What matters is how long it lasts: the longest stretch for any Nifty 500 SIP was 1 year 11 months.
Written for SIP investors deciding whether a bad patch is unusual. Data to 6 Oct 2026.
How often a SIP is under water
"Under water" here means the SIP's value on a fund statement (after fund costs, before tax and before selling) is below the money put in so far. Each cell takes every SIP that has reached that age and counts the share that was under water at that point. New money keeps buying units at the going price, so a SIP climbs out of a fall faster than the index it holds. But in its first months the money has had no time to grow, so even a small fall puts it under.
Share of SIPs under water, by how long they had been running
Statement value below the money put in, per cent of start months from Apr 2005
How to read it. Darker is more often under water. Read across a row to watch the share fall as a SIP ages. In the Nifty 500 it went from 26% at one year to 2% at five; in the Smallcap 250, 24% were under water at three years. Gold took longer to get clear: at five years 8.5% of gold SIPs were still under water, against 1.5% for the Nifty 500. The equal four-asset mix was clear of it in every start month by five years.
Share of SIPs under water, by how long they had been running. Heat table, 15 rows by 7 columns. Values from 0% to 35%.
Per cent of start months under water at each age, and below the same instalments in fixed deposits. Later ages have fewer start months: a 10-year-old SIP must have started by 2016.
| Portfolio | 6 months: under | below FD | 1 year: under | below FD | 2 years: under | below FD | 3 years: under | below FD | 5 years: under | below FD | 7 years: under | below FD | 10 years: under | below FD | More |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Nifty 50 | 34% | 40% | 23% | 33% | 16% | 25% | 8% | 20% | 1% | 9% | 0% | 3% | 0% | 1% | |
| Nifty 100 | 34% | 39% | 24% | 34% | 16% | 24% | 8% | 18% | 1% | 8% | 0% | 2% | 0% | 1% | |
| Nifty 500 | 35% | 38% | 26% | 34% | 16% | 28% | 8% | 21% | 2% | 10% | 0% | 7% | 0% | 1% | |
| Nifty LargeMidcap 250 | 34% | 39% | 25% | 34% | 17% | 29% | 9% | 21% | 2% | 8% | 0% | 4% | 0% | 0% | |
| Nifty Midcap 150 | 34% | 37% | 25% | 35% | 20% | 27% | 13% | 24% | 2% | 9% | 0% | 6% | 0% | 0% | |
| Nifty Smallcap 250 | 35% | 39% | 35% | 42% | 26% | 36% | 24% | 29% | 10% | 25% | 5% | 18% | 1% | 4% | |
| Gold | 30% | 37% | 21% | 33% | 14% | 29% | 10% | 28% | 9% | 30% | 1% | 31% | 0% | 19% | |
| G-sec, 5-year | 9% | 34% | 2% | 32% | 0% | 24% | 0% | 12% | 0% | 4% | 0% | 0% | 0% | 0% | |
| 60/40 | 28% | 38% | 18% | 31% | 10% | 21% | 4% | 16% | 0% | 4% | 0% | 1% | 0% | 0% | |
| 60/20/20 | 24% | 34% | 16% | 26% | 6% | 18% | 3% | 9% | 0% | 1% | 0% | 1% | 0% | 0% | |
| Equal thirds | 17% | 28% | 6% | 20% | 1% | 10% | 0% | 6% | 0% | 0% | 0% | 0% | 0% | 0% | |
| Four assets, equal quarters | 23% | 32% | 17% | 28% | 8% | 20% | 4% | 15% | 0% | 2% | 0% | 1% | 0% | 0% | |
| Four assets, 80% equity | 30% | 35% | 21% | 35% | 15% | 26% | 8% | 18% | 1% | 5% | 0% | 3% | 0% | 1% | |
| Four assets, 60% equity | 26% | 33% | 17% | 29% | 7% | 22% | 4% | 13% | 1% | 2% | 0% | 1% | 0% | 0% | |
| Four assets, 40% equity | 20% | 29% | 11% | 26% | 3% | 14% | 1% | 8% | 0% | 1% | 0% | 0% | 0% | 0% |
How far below
For each SIP with at least a year behind it, the deepest its statement value went below the money put in, at any session. The middle half of start months is the bar, the median the dot. A SIP's deepest point is usually shallower than its index's worst fall, because the money that went in during the fall bought cheaply.
The deepest point below the money put in
Per cent below cumulative instalments at the worst session, across start months
How to read it. Further left is deeper. The median Nifty 500 SIP's deepest point was −19%; the worst, −50%, was a SIP begun in Oct 2007 at the bottom on 27 Oct 2008. The Smallcap 250's worst was −56%. The equal four-asset mix's worst was −28%, and gold's −16%.
The deepest point below the money put in. 1 horizon, 15 bands. Nifty 50: median −16%; Nifty 100: median −17%; Nifty 500: median −19%; Nifty LargeMidcap 250: median −20%; Nifty Midcap 150: median −22%; Nifty Smallcap 250: median −29%; Gold: median −5%; G-sec, 5-year: median −1%; 60/40: median −8%; 60/20/20: median −6%; Equal thirds: median −3%; Four assets, equal quarters: median −7%; Four assets, 80% equity: median −15%; Four assets, 60% equity: median −8%; Four assets, 40% equity: median −4%.
Start months with at least a year of history. 'Ever under' is the share that dipped below the money put in at least once.
| Portfolio | Ever under | Median deepest point | Worst | Worst start | On | More |
|---|---|---|---|---|---|---|
| Nifty 50 | 97% | −16% | −48% | Oct 2007 | 27 Oct 2008 | |
| Nifty 100 | 97% | −17% | −49% | Oct 2007 | 27 Oct 2008 | |
| Nifty 500 | 96% | −19% | −50% | Oct 2007 | 27 Oct 2008 | |
| Nifty LargeMidcap 250 | 95% | −20% | −52% | Oct 2007 | 27 Oct 2008 | |
| Nifty Midcap 150 | 96% | −22% | −55% | Oct 2007 | 27 Oct 2008 | |
| Nifty Smallcap 250 | 97% | −29% | −56% | Oct 2007 | 27 Oct 2008 | |
| Gold | 89% | −5% | −16% | Sept 2012 | 17 Apr 2013 | |
| G-sec, 5-year | 60% | −1% | −6% | May 2013 | 19 Aug 2013 | |
| 60/40 | 94% | −8% | −30% | Oct 2007 | 27 Oct 2008 | |
| 60/20/20 | 90% | −6% | −31% | Nov 2007 | 27 Oct 2008 | |
| Equal thirds | 74% | −3% | −17% | Dec 2007 | 24 Oct 2008 | |
| Four assets, equal quarters | 87% | −7% | −28% | Nov 2007 | 27 Oct 2008 | |
| Four assets, 80% equity | 94% | −15% | −43% | Oct 2007 | 27 Oct 2008 | |
| Four assets, 60% equity | 90% | −8% | −33% | Nov 2007 | 27 Oct 2008 | |
| Four assets, 40% equity | 79% | −4% | −22% | Dec 2007 | 27 Oct 2008 |
The longest stretches
The longest unbroken run of month-ends a single SIP spent in each state, over all start months. Below inflation is the hardest bar: the instalments grown by consumer prices. A stretch marked still running belongs to a SIP that has not come out yet, so its length is a floor.
The longest a SIP stayed below the money put in, FDs and inflation
Longest run of consecutive month-ends, any start month from Apr 2005
How to read it. The longest any Nifty 500 SIP stayed under water was 1 year 11 months (from a Oct 2010 start); below the FD SIP, 2 years 1 month; below inflation, 3 years 4 months. For the Smallcap 250 the stretch below FDs was 3 years 5 months.
The longest a SIP stayed below the money put in, FDs and inflation. Bar chart of 15 items. Below the money put in: highest Nifty Smallcap 250 at 34, lowest G-sec, 5-year at 6.
Each cell: the longest stretch and the start month of the SIP that had it.
| Portfolio | Below the money put in | Below FDs | Below inflation | More |
|---|---|---|---|---|
| Nifty 50 | 1 year 8 months, from a Oct 2010 start | 1 year 10 months, from a Jul 2010 start | 3 years 2 months, from a Aug 2010 start | |
| Nifty 100 | 1 year 11 months, from a Oct 2010 start | 2 years 1 month, from a Oct 2010 start | 2 years 11 months, from a Dec 2009 start | |
| Nifty 500 | 1 year 11 months, from a Oct 2010 start | 2 years 1 month, from a Oct 2010 start | 3 years 4 months, from a Aug 2010 start | |
| Nifty LargeMidcap 250 | 1 year 11 months, from a Oct 2010 start | 2 years 1 month, from a Oct 2010 start | 3 years 4 months, from a Jul 2010 start | |
| Nifty Midcap 150 | 1 year 10 months, from a Jul 2010 start | 2 years 2 months, from a Feb 2017 start | 3 years 4 months, from a Apr 2010 start | |
| Nifty Smallcap 250 | 2 years 10 months, from a Dec 2017 start | 3 years 5 months, from a Jul 2010 start | 3 years 6 months, from a Apr 2010 start | |
| Gold | 1 year 11 months, from a Feb 2012 start | 5 years 11 months, from a Jul 2011 start | 6 years 8 months, from a Sept 2011 start | |
| G-sec, 5-year | 6 months, from a Feb 2008 start | 3 years 5 months, from a Jan 2009 start | 6 years 8 months, from a Apr 2005 start | |
| 60/40 | 1 year 4 months, from a Sept 2007 start | 1 year 11 months, from a Oct 2010 start | 3 years 4 months, from a Jun 2009 start | |
| 60/20/20 | 1 year 4 months, from a Oct 2007 start | 1 year 11 months, from a Oct 2010 start | 2 years 1 month, from a Oct 2010 start | |
| Equal thirds | 1 year 1 month, from a Nov 2007 start | 1 year 3 months, from a Jan 2008 start | 1 year 3 months, from a Jan 2008 start | |
| Four assets, equal quarters | 1 year 4 months, from a Nov 2007 start | 1 year 9 months, from a Jan 2018 start | 2 years, from a Aug 2010 start | |
| Four assets, 80% equity | 1 year 10 months, from a Sept 2010 start | 2 years, from a Aug 2010 start | 3 years 4 months, from a Jul 2010 start | |
| Four assets, 60% equity | 1 year 4 months, from a Nov 2007 start | 1 year 10 months, from a Sept 2010 start | 2 years, from a Aug 2010 start | |
| Four assets, 40% equity | 1 year 3 months, from a Jan 2008 start | 1 year 4 months, from a Dec 2007 start | 1 year 4 months, from a Dec 2007 start |
When equity lost to gold, G-sec and FDs
The same instalments on the same dates in another asset, compared by after-tax XIRR over whole windows. Each cell is the share of start months in which the row's SIP finished behind. The losing windows bunch together in time, so a share of a third can be one long episode rather than one bad year in three.
How often each SIP finished behind gold, G-sec or an FD
Share of common start months, after tax, at 5 and 10 years
How to read it. Darker is more often behind. A 10-year Nifty 500 SIP finished behind gold in 17% of start months (41% at 5 years), behind G-sec in 4%, and behind FDs in 1% (8% at 5 years).
How often each SIP finished behind gold, G-sec or an FD. Heat table, 15 rows by 6 columns. Values from 0% to 67%.
When the Nifty 500 SIP was ahead, and when it was not
10-year SIP by start month: Nifty 500 after-tax XIRR minus the same SIP in gold, G-sec and FDs, points
How to read it. Above zero, equity won. Against gold it fell below zero for starts in Mar 2006; Apr 2010 to Aug 2010; Mar 2015 to Oct 2016: ten-year windows that ended in gold's run-up of recent years. Against FDs it stayed above zero in every start month but Apr 2010.
When the Nifty 500 SIP was ahead, and when it was not. Line chart, Apr 2005 to Oct 2016. Against gold: latest −8.5 pts, range −9.8 pts to +10.7 pts. Against G-sec: latest +5.3 pts, range −4.5 pts to +10.5 pts. Against FDs: latest +6.9 pts, range −1.5 pts to +12.7 pts.
The FD is 364-day T-bill primary yield (91-day before April 2005), after slab tax and cess each year.
Share of common start months the row finished behind, and the median gap at 10 years (row minus rival; negative means the rival was ahead at the median).
| Portfolio | Against | Behind at 5 years | Behind at 10 years | Median gap, 10 years | More |
|---|---|---|---|---|---|
| Nifty 50 | Gold | 40% | 19% | +3.9 pts | |
| Nifty 50 | G-sec | 13% | 4% | +4.8 pts | |
| Nifty 50 | FD | 7% | 1% | +6.9 pts | |
| Nifty 100 | Gold | 41% | 17% | +4.2 pts | |
| Nifty 100 | G-sec | 11% | 3% | +5.4 pts | |
| Nifty 100 | FD | 5% | 1% | +7.5 pts | |
| Nifty 500 | Gold | 41% | 17% | +4.8 pts | |
| Nifty 500 | G-sec | 16% | 4% | +5.7 pts | |
| Nifty 500 | FD | 8% | 1% | +8.1 pts | |
| Nifty LargeMidcap 250 | Gold | 39% | 12% | +6.3 pts | |
| Nifty LargeMidcap 250 | G-sec | 12% | 2% | +7.3 pts | |
| Nifty LargeMidcap 250 | FD | 7% | 1% | +9.7 pts | |
| Nifty Midcap 150 | Gold | 38% | 9% | +8.4 pts | |
| Nifty Midcap 150 | G-sec | 14% | 2% | +9.8 pts | |
| Nifty Midcap 150 | FD | 9% | 0% | +12.4 pts | |
| Nifty Smallcap 250 | Gold | 42% | 16% | +6.5 pts | |
| Nifty Smallcap 250 | G-sec | 27% | 11% | +7.8 pts | |
| Nifty Smallcap 250 | FD | 23% | 3% | +10.1 pts | |
| Gold | G-sec | 33% | 45% | +0.3 pts | |
| Gold | FD | 31% | 18% | +3.0 pts | |
| G-sec, 5-year | Gold | 67% | 55% | −0.3 pts | |
| G-sec, 5-year | FD | 0% | 0% | +2.4 pts | |
| 60/40 | Gold | 44% | 19% | +2.9 pts | |
| 60/40 | G-sec | 12% | 3% | +3.9 pts | |
| 60/40 | FD | 4% | 0% | +6.2 pts | |
| 60/20/20 | Gold | 41% | 17% | +3.1 pts | |
| 60/20/20 | G-sec | 5% | 3% | +4.1 pts | |
| 60/20/20 | FD | 2% | 0% | +6.5 pts | |
| Equal thirds | Gold | 45% | 17% | +1.9 pts | |
| Equal thirds | G-sec | 12% | 3% | +2.4 pts | |
| Equal thirds | FD | 1% | 0% | +5.2 pts | |
| Four assets, equal quarters | Gold | 42% | 17% | +3.9 pts | |
| Four assets, equal quarters | G-sec | 11% | 3% | +4.8 pts | |
| Four assets, equal quarters | FD | 2% | 0% | +7.3 pts | |
| Four assets, 80% equity | Gold | 40% | 12% | +5.4 pts | |
| Four assets, 80% equity | G-sec | 12% | 3% | +6.7 pts | |
| Four assets, 80% equity | FD | 5% | 1% | +9.0 pts | |
| Four assets, 60% equity | Gold | 40% | 15% | +4.3 pts | |
| Four assets, 60% equity | G-sec | 6% | 3% | +5.3 pts | |
| Four assets, 60% equity | FD | 2% | 0% | +7.8 pts | |
| Four assets, 40% equity | Gold | 42% | 15% | +3.2 pts | |
| Four assets, 40% equity | G-sec | 6% | 2% | +3.9 pts | |
| Four assets, 40% equity | FD | 1% | 0% | +6.4 pts |
What I predicted
Written into the specification on 8 October 2026, before any of these numbers existed.
S10 Half held
More than half of all Nifty 500 SIPs (by start month) were under water at some month-end, and none was still under water five years after it started.
96% of start months with at least a year behind them dipped below the money put in at some point. Five years in, 2% were still under water, so the second half failed: a few SIPs begun in the run-up to a fall took longer than five years to get clear. Recent starts are still young, so the first share will move as they age.
S11 Failed
A 10-year Nifty 500 SIP had a lower after-tax XIRR than the same SIP in gold in more than a quarter of common start months.
It lost to gold in 17% of 139 ten-year windows (starts in Mar 2006; Apr 2010 to Aug 2010; Mar 2015 to Oct 2016). Those windows overlap heavily: they are close to one episode, not many.
Historical research on index portfolios, not investment advice. Returns are after fund costs; 'after tax' applies Indian capital-gains rules by date for a 30% slab investor and sells everything at the end. Most factor, mid-cap and small-cap index history before each index's launch date is back-tested by NSE. One 21-year sample, domestic assets only.
The workings
Data
Every chart has a table view, a CSV download and a link to its published JSON. Bundles are listed in data status.