SIP vs lump sum: time in the market

A friend sold a flat last month. For the first time he had a genuinely large number sitting in one account, and it made him nervous. So he asked the question that sounds responsible. "Should I drip it into the market through a SIP over a year, so I don't accidentally buy at the top?"

Spreading it out feels safe. I pulled the math, and safe turns out to have a price.

Run the same Rs 12 lakh two ways. Put it all in today, or feed in Rs 1 lakh a month for a year while the rest waits in a liquid sleeve earning about 6 percent. In a year where the market climbs, going all in ends around 13.52 lakh and spreading it out ends around 13.15 lakh. The gap is not magic. It is just that money waiting on the sidelines is money that is not compounding yet.

Then flip the year. Let the market fall first and recover later. Now all-in ends around 13.26 lakh and spreading ends around 14.73 lakh, because the later instalments quietly bought cheaper units. So staggering wins, but only in the year that drops first. And here is the thing the brochures skip: markets rise roughly two years in three. That single fact is the whole reason putting it in at once beat spreading it across twelve months in about two of every three historical windows.

Which leads to the part most people get backwards. A SIP is not a clever way to buy low. It buys on a calendar, not on a chart. The averaging only helps you in the years the calendar happens to walk you through a dip. What a SIP actually does, for someone earning a monthly salary, is put each rupee to work the day it arrives. That is not timing the market. That is the opposite. It is the cleanest way to stop trying.

Time in the market beats timing the market. A SIP is not how you beat that rule. It is how you stop fighting it

I ran it on the last two years

The opening's two in three comes from long histories. I wanted to see the same test on the most recent stretch of real prices, so I took NSE's month end closes for the Nifty 50 from Oct 2024 and ran the friend's choice on every 12 month window I could: Rs 12 lakh all at once at a month end, against Rs 1 lakh at each of the next 12 month ends with the waiting cash earning 6 percent a year. Thirteen windows, each scored a year after it started.

That is the reverse of the long run result, and the reason is in the prices. The Nifty 50 closed Oct 2024 at 24,205 and was at 22,520 on 9 Oct 2026, with a deep fall in Mar 2026. Two years that ended lower are exactly the minority case the opening describes, the year that drops first. Over long histories those years are fewer. Over this particular two years they were most of the sample.

Feb 2025 was a low month end, so putting everything in then won by over a lakh. Oct 2025 was near a high, so spreading won by about Rs 92,000. Nobody knew either on the day. That is the honest summary of the choice: the lump sum wins when the year goes up, and you only find out afterwards which kind of year it was.

Your own version

The calculator below simplifies the drip to one number: the average price your twelve instalments end up paying. Change it to see when spreading wins.

What the SIP is really for

For a lump sum, both choices are rules, and that is the point the free Learn module on why rules beat gut feeling makes in general. All at once is a rule. Twelve equal monthly instalments is a rule. The one choice that has no rule behind it is waiting for the market to feel right, which is how lump sums end up sitting in a savings account for a year. The lump sum against SIP calculator compares the two at any return. Related: lump sum against drip and the cost of waiting.