Why the Smartest SIP Is the One You Never Touch

Lately, a certain kind of financial influencers have been telling retail investors that a monthly SIP will not build real wealth. The “smart” alternative sounds irresistible: invest only when the Nifty PE ratio is below its historical average, and stop when it rises above.

It sounds like common sense. Buy when markets are cheap, step aside when they are expensive. But a rule that sounds smart is not the same as a rule that works, so we decided to test it, along with every other “clever SIP” idea we have heard in the last few years.

We tested six popular SIP timing rules against roughly 30 years of Nifty 50 data. The plain monthly SIP held its own against every one of them, and the single rule that edged ahead did so by a margin too thin to trade on.

The six ideas

We backtested the following rules, each against the same baseline:

1. PE rule: invest only when the Nifty PE is below its 12 month average.
2. Below 200 DMA: invest only when the Nifty trades below its 200 day moving average, the contrarian’s version of “buy fear”.
3. 52 week low to high: start investing at a 52 week low and stop at a 52 week high.
4. Above 200 DMA: the trend follower’s version, investing only when the Nifty is above its 200 day average.
5. 52 week high to low: start at a new 52 week high, stop at a new 52 week low.
6. The boring one: the same amount every month, no charts, no PE, no news. The investor who set up a SIP in 1996 and forgot about it.

The test used ₹10,000 a month on the Nifty 50 price index from January 1996 to October 2026, which is 370 months. It ignores dividends, costs and tax, so every strategy is treated identically on those fronts.

First, the percentage return trap

On XIRR, several of the timing rules look better than the plain SIP, which returned 11.49 percent a year. Investing only below the 200 DMA scored 12.05 percent. The 52 week low to high rule scored 12.15 percent, the highest of all seven rows we ran.

If you stop reading there, the influencers appear to be right. Now look at the rupees.

The 52 week low rule invested in just 108 of 370 months. It put in ₹10.8 lakh in total and ended with ₹1.49 crore. The plain SIP put in ₹37 lakh and ended with ₹3.05 crore. The best percentage return finished with roughly half the wealth of the boring approach. The “below 200 DMA” rule did worse, ending at ₹1.37 crore, about 45 percent of the plain SIP. Even the PE rule, which invested in 208 months, ended at ₹2.26 crore, about 74 percent.

XIRR measures the return on money actually invested. It says nothing about the money that sat on the sidelines. A rule that keeps you out of the market most of the time can post a lovely percentage on a small base and still lose badly on wealth. The 52 week low rule sat out 262 months, almost 22 years of the 30 year window.

Round one: park the skipped money

That comparison is unfair, because the skipped money did not vanish. A sensible investor would put it somewhere. So we gave every strategy the same ₹37 lakh to set aside, and routed each skipped instalment into a liquid fund earning 6 percent a year.

The results:

Every rule lost, by between ₹41 lakh and ₹111 lakh. The reason is arithmetic, not mystery. Over this window the Nifty compounded at around 11 percent, while the parked money earned 6. Every month that cash waited in a liquid fund, it gave up roughly five percentage points a year against the market.

Round two: deploy the pile later

The fair objection to round one is that the influencer’s rule does not say “leave the money in a liquid fund forever.” The idea is to hold cash during expensive phases and then put the whole pile into the Nifty when the rule says buy. So we tested that version too.

This is the fairest version of the influencer’s idea, and it is close to a tie. The PE rule finishes first, ahead of the plain SIP by about ₹4 lakh, or 1.3 percent over three decades. Three of the other rules land within about 1 percent of the plain SIP, or worse.

What that 1.3 percent actually costs

To collect the PE rule’s edge, an investor would have had to do three things for 30 years without fail.

First, track the Nifty PE every month and recompute a 12 month average. Second, watch a cash pile grow for months, sometimes years, while friends and colleagues talked about rising markets. Third, deploy the entire pile at once, usually when the headlines were at their worst, because that is precisely when PE falls below its recent average.

The plain SIP investor needed none of that discipline. They needed one habit, which was to leave the mandate running.

And a 1.3 percent edge is thin. It sits comfortably inside the range that small choices in a backtest can move, which brings us to the honest part.

What this test does not prove

We would rather you hear the limits from us than discover them later.

It does not test the exact claim being made. The influencer said, “below its historical average.” Our PE rule used the 12 month average, which is a moving, short-memory benchmark. Over the full history since 1999, the Nifty’s median PE sits near 21, and for much of the last decade it has traded above that. A rule anchored to the long-run average would have kept an investor out of the market far more often than ours did. We think that makes the influencer’s version look worse, not better, but we have not run it, so treat that as a hypothesis.

PE data is shorter than the backtest. NSE has published Nifty PE data publicly only since 1 January 1999. A 12 month PE average cannot exist for the first few years of a 1996 start, so those early months need a stated treatment, such as defaulting to a normal SIP. Whatever was chosen should be disclosed.

The PE series has a break in it. NSE calculated the Nifty PE on standalone earnings until March 2021 and switched to consolidated earnings from April 2021. The two are not directly comparable, so a rolling 12 month average that straddles that month is blended from two different measures. The PE rule’s result carries that uncertainty.

The 6 percent liquid fund assumption is a simplification. Liquid fund 10 year returns across the industry have sat around 6 percent, which makes the assumption reasonable for the recent past. But the category did not exist for the first several years of the window, and cash yields in the late 1990s were higher. A higher cash rate would help every timing rule a little. It would not change the ranking by much, because the gap in round one is wide, but it could narrow the round two margins to nothing.

It is one path, not a distribution. Thirty years is a single sequence of events. Rules that would have behaved differently in a different sequence of bull and bear markets look identical to rules that were simply lucky here.

It is a price index without costs or tax. Dividends, brokerage, switching costs and tax on liquid fund gains were all ignored. Tax matters most: a high bracket investor earns well under 6 percent after tax on cash, and switching in and out of the market triggers gains more often than a plain SIP does. Each of these pushes against the timing rules, not for them.

Does valuation matter at all?

None of this means valuation is meaningless. Independent studies of Nifty PE since 1999 find that the cheapest fifth of days were followed by far stronger five year returns, around 23 percent a year at the median, compared with roughly 10 to 13 percent for the other bands. The signal is real over long horizons.

The trouble is that the same data shows no clean ladder. The most expensive fifth of days also delivered a median five year return of about 13 percent, better than the middle bands. A signal that is strong at one extreme and ambiguous everywhere else is hard to trade as a switch. Valuation is useful context for how much you might expect, not a reliable instruction for when to stop investing.

What to do with this

If you are investing for a goal ten or more years away, the evidence points to a boring conclusion.

Keep the SIP running. Do not pause it because the PE is high or the index crossed a moving average. If you cannot resist adjusting something, adjust the amount, not the timing. Our own table shows what a 10 percent annual step up in the SIP amount did over the same period: it invested ₹2.15 crore in total and ended at ₹7.78 crore. That is not a like-for-like comparison with a flat SIP, because far more money went in, and its XIRR of 11.07 per cent was slightly below the flat SIP’s. It does illustrate the real lever, which is how much you put to work and for how long, not when within a given month you do it.

If you want to act on valuation, there is a milder way to do it. Keep the base SIP untouched and direct occasional lump sums, such as bonuses, toward the market when valuations are genuinely depressed. That adds to the habit, it doesn’t replace it.

The takeaway

Thirty years of data, six clever rules, and the best result was a 1.3 percent edge that required decades of monthly monitoring and the nerve to deploy a large pile in the worst headlines. The plain SIP matched or beat the rest without a single decision.

Your SIP does not need a strategy. It needs you to keep it running.