**The wider the window you use to define a dip, the better the strategy performs. Weekly drawdown barely works. Monthly drawdown works modestly. ATH drawdown works consistently — and with far less downside risk than either of the other two.

The premise

The idea behind all three strategies is the same. If you wait for Nifty to fall a meaningful amount before deploying your accumulated bank balance, you are buying at a discount rather than investing indiscriminately. The disagreement is over what counts as a meaningful dip. A 5% drop from this week's high is a different kind of event from a 5% drop from the monthly high, which is a completely different kind of event from a 36% drop from the all-time high. These three strategies test whether that distinction matters.

It matters enormously.

How all three strategies work

A bank account receives ₹10,000 on the first trading day of each month, compounding at 5% annualised on actual calendar days. When a signal fires, the full available balance deploys after a 0.03% transaction cost. All purchased units are held permanently with no selling or rebalancing. The benchmark is a capital-matched monthly SIP investing ₹10,000 on the first trading day of each month at 0.03% cost, no bank, no timing. SIP XIRR over the full 30-year test period: 11.9907%.

The signal logic differs only in the reference point for the drawdown calculation.

Weekly drawdown: (WeekHigh - Close) / WeekHigh >= threshold. Week high is the rolling maximum over the prior 5 trading days. Thresholds tested from 1.0% to 20.0% in 0.5% steps — 39 total.

Monthly drawdown: (MonthHigh - Close) / MonthHigh >= threshold. Month high is the rolling maximum over the prior 21 trading days. Same 39 thresholds.

ATH drawdown: (ATH - Close) / ATH >= threshold. ATH is the all-time high up to and including the current date. Thresholds tested from 5% to 40% in 1% steps — 36 total.

Weekly drawdown: four wins, one accident, one disaster

Four of 39 thresholds outperform SIP. At first glance that scorecard is poor but comparable to the daily-signal strategies from the previous posts. On closer inspection, it is worse.

The outperforming thresholds are 4.5%, 5%, 6.5%, and 11%. The first three are genuine if small edges. At 4.5% the strategy adds ₹28,370 against SIP across 61 trades. At 5% it adds ₹72,267 across 49 trades. At 6.5% it adds ₹2.37 lakh across 23 trades.

The 11% threshold is a different kind of result. It produces XIRR of 12.225%, alpha of +0.234%, and additional wealth of ₹15.90 lakh — the best single number in this entire post. It does this across exactly 4 trades over 30 years. The surrounding thresholds tell you what that means: at 10% and 10.5%, the strategy underperforms by ₹41.27 lakh each. At 11.5% and 12%, it underperforms by ₹55.47 lakh each. The 11% outperformance is not a strategy — it is the result of four specific historically unusual weeks landing at exactly the right moments. One threshold step in any direction and the result collapses.

The sweep is not monotonic. There is no clean range where the strategy consistently wins. At 5.5%, sitting between two outperformers, the strategy underperforms by ₹6.54 lakh. At 7%, between 6.5% and 7.5%, it underperforms by ₹8.75 lakh. The alpha surface is spiky, not smooth, which is itself a signal that the strategy is capturing random variation rather than a structural edge.

The real damage begins at 8.5%. Between 8.0% and 8.5%, additional wealth drops from -₹5.11 lakh to -₹27.20 lakh in a single threshold step, on a trade-count change from 14 to 11. At 9.0%, it drops further to -₹43.47 lakh. These are not gradual deteriorations — they are cliffs.

Beyond 17%, the strategy fires zero times. The bank accumulates at 5% for 30 years while Nifty compounds at 12%. XIRR drops to 5.0%. The trailing difference against SIP: ₹2.42 crore. Every rupee contributed over three decades stays in bank and never reaches the market.

Monthly drawdown: a real sweet spot

The monthly drawdown strategy produces 18 outperforming thresholds out of 39. That is a materially better outcome. The strategy works across a genuine range rather than at isolated threshold islands.

The best threshold is 7%, producing XIRR of 12.136%, alpha of +0.145%, and additional wealth of ₹9.76 lakh across 38 trades. The surrounding thresholds are nearly as strong: 7.5% adds ₹8.75 lakh and 8% adds ₹9.19 lakh, both across 32-36 trades. This is a real sweet spot, not a spike.

The outperforming range runs broadly from 2.5% to 11.5%, with one notable exception: the 9.5% threshold, which sits between two outperformers and underperforms by ₹6.07 lakh across 18 trades. No clear structural explanation for that dip — 9.5% monthly drawdown appears to select for a specific category of market conditions that subsequently performed poorly. Every other threshold in the 2.5-11.5% range outperforms.

Below 2.5% the strategy fires constantly — 96 to 129 trades — and accumulates too many instances of buying into ordinary pullbacks that then recover slowly. At 1% and 1.5%, it underperforms by ₹2.85 lakh and ₹66,762 respectively.

Above 12%, the strategy starts to break down as it gets more selective without selecting for better moments. At 14-15%, it trails SIP by ₹13.40 lakh across 6 trades. At 17.5-20%, it trails by ₹84.09 lakh across 2 trades — those 2 trades being the 2020 COVID crash and one other extreme event. Unlike weekly drawdown, monthly drawdown never reaches zero trades. Even across 30 years, extreme monthly drawdowns of 17.5%+ occur at least twice.

The worst case for monthly drawdown is ₹84.09 lakh of trailing wealth. That is a serious number, but it is roughly one-third of the weekly drawdown strategy's catastrophic floor.

ATH drawdown: a different kind of signal

The ATH drawdown strategy produces 31 outperforming thresholds out of 36. Almost every tested configuration beats SIP.

The reason is structural. A 36% drawdown from the all-time high is not a weekly fluctuation or a monthly pullback. It means Nifty has fallen into genuine bear market territory — the 2001-2003 dot-com correction, the 2008-2009 global financial crisis, the 2020 COVID crash. Deploying accumulated bank capital during those periods is deploying into real distress, when future returns are structurally higher.

The best threshold is 36%, producing XIRR of 12.437%, alpha of +0.446%, and additional wealth of ₹30.99 lakh across 43 trades. The second best is 37% at ₹30.47 lakh across 38 trades. The 35%, 34%, and 33% thresholds produce ₹23.46 lakh, ₹20.47 lakh, and ₹15.71 lakh respectively.

The outperforming range is broad and consistent. From the 8% threshold onward, every configuration through 38% beats SIP. At 8% it barely wins — ₹6,086 across 238 trades. At 12% it adds ₹3.03 lakh across 182 trades. At 20% it adds ₹4.43 lakh across 115 trades. The alpha grows as the threshold rises, because higher ATH drawdown thresholds progressively filter out noise and leave only genuine crash deployments.

The 30% threshold is the one anomaly in the sweep. It produces alpha of +0.129% and additional wealth of ₹8.66 lakh — lower than both the 29% threshold (₹23.72 lakh) and the 31% threshold (₹11.50 lakh). This is a result of the specific historical crash dates that fall above and below 30%. It does not change the structural story.

Only four thresholds underperform: 5%, 6%, 7%, and 39%. At 5-7%, the ATH drawdown condition is met so frequently that the strategy deploys almost constantly — 254 to 291 trades — and functions essentially as an imperfect SIP with slightly erratic timing. The idle bank balance in these configurations is zero. At 39%, the strategy holds back 97% of trading days, deploying only 32 times. The ₹31.72 lakh sitting idle at 39% earns 5% while waiting for Nifty to fall 39% from its peak — a threshold that was met only during the deepest part of the GFC and COVID. At 40%, the configuration improves slightly to -₹1.13 lakh trailing, because the 28-trade count at that threshold happens to align with slightly better entry points.

The most important number in the entire ATH drawdown dataset: the worst case is -₹11.98 lakh at the 39% threshold. Compare that to the weekly drawdown strategy's worst case of -₹2.42 crore. ATH drawdown's downside risk is structurally bounded because the strategy remains invested as long as Nifty has not crashed more than 39% from its peak. Capital stays in the market most of the time.

Reading all three together

The key variable is not the threshold — it is the reference point. Weekly drawdown measures noise. Monthly drawdown measures medium-term trend. ATH drawdown measures historical context.

A 5% weekly drawdown happens constantly on Nifty. Nifty will routinely fall 5% from a recent intraweek high and then recover in the same week or the next. Directing accumulated bank capital at those moments is not timing a real opportunity — it is reacting to normal market volatility. The signal fires prolifically at low thresholds and selects for arbitrary historical moments at high thresholds.

A 5% monthly drawdown is a more meaningful filter but still fires regularly enough that the strategy's timing adds limited value. The 18-of-39 outperformer count and the ₹9.76 lakh best result are real improvements over weekly drawdown, but the ceiling is still low.

A 36% ATH drawdown is something most market participants rarely see. Nifty spent most of the 30-year test period within 20% of its all-time high. The 43 trade executions at the best threshold cluster around genuine crashes. Every trade at that threshold represents a meaningful bear market entry.

The scorecard makes this concrete: 4/39 for weekly, 18/39 for monthly, 31/36 for ATH. The outperformance rate doubles with each wider reference window.

One number captures the asymmetry: the best result across all three strategies is ATH drawdown at 36% with ₹30.99 lakh of additional wealth. The worst result across all three strategies is weekly drawdown at 17%+ with -₹2.42 crore of trailing wealth. The same broad strategy concept, applied with a different reference point, produces a 27x difference in worst-case outcomes.

Verdict

Weekly drawdown produces 4 winners from 39 thresholds. The best meaningful result is ₹2.37 lakh of additional wealth at 6.5%. The 11% result of ₹15.90 lakh is an artifact of 4 well-timed trades, not a replicable edge. The strategy's worst case destroys ₹2.42 crore against SIP.

Monthly drawdown produces 18 winners from 39 thresholds. The best result is ₹9.76 lakh of additional wealth at 7%. The strategy's worst case destroys ₹84.09 lakh against SIP.

ATH drawdown produces 31 winners from 36 thresholds. The best result is ₹30.99 lakh of additional wealth at 36%. The strategy's worst case destroys ₹11.98 lakh against SIP.

The reference point is the strategy. Define the dip using weekly noise and you are buying randomly. Define it using all-time-high context and you are buying genuine distress. The numbers are unambiguous about which definition produces a usable edge.**