How SmartHold Works & Its Limits

1. The method, in plain terms

SmartHold's core strategy has three steps, applied the same way to every stock, every day, with no human discretion in the loop:

1. Fundamentals screen — stocks are checked against public fundamentals (profit growth, ROE, debt, valuation) sourced from screener.in before being considered at all.
2. Momentum ranking — eligible stocks are ranked by long-term relative momentum, with a trend filter (an approach with real academic backing, tuned for the Indian market) so only stocks whose own price trend is currently intact are considered.
3. Gold-hedge rule — each holding's own price is compared against gold using a price-channel breakout rule on the stock-vs-gold ratio: once that ratio breaks down, the rule calls for shifting that position into gold; once it recovers, the rule calls for shifting back. This runs every single day, automatically, against live NSE prices — not a one-time decision.

This same mechanism is also what any ticker YOU add to your own list gets checked against — SmartHold isn't only a picker of its own selections, it's a daily check on every holding in every list.

2. Backtests: what the numbers mean, and their real limits

Every backtest and "Total Return" figure shown in the app is time-weighted (chained day-over-day from whatever was actually held that day) and includes an explicit transaction-cost assumption (0.3% round-trip) on every simulated entry, exit, and gold-hedge flip — it is not a frictionless, cost-free simulation.

The specific headline statistics quoted on our landing page and marketing pages — currently "95.3% win rate vs. plain buy-and-hold, across 1,030 Indian mutual funds backtested" and "+10.3pp average excess return from the gold-hedge switch" — describe one specific historical backtest, run against every mutual fund scheme searchable in SmartHold's own Mutual Funds tab (Direct, Growth-option, currently listed — this carries the usual survivorship bias of any "currently listed" universe: closed or merged schemes aren't in the test). 1,030 of 1,031 attempted schemes produced a usable result; each was tested from 1 January 2020 (or its own launch date, if later) through 22 September 2026, applying the same gold-hedge switch described in §1 against GOLDBEES, with transaction costs included on every switch. "Win rate" is the share of schemes where the strategy's CAGR beat that same scheme's own plain buy-and-hold CAGR over the identical window (982 of 1,030); "average excess return" is the mean of that gap (strategy CAGR minus buy-and-hold CAGR) across all 1,030 schemes. Not every scheme won — the worst case lost 18.87 percentage points to the switch, the best case gained 42.53 points. This window (2020–2026) included the COVID-19 crash and an unusually strong multi-year rally in gold, both of which favoured a gold-hedge rule; results over other periods may be materially weaker.

Every backtest has the same fundamental limitation: it shows what a strategy would have done on data that has already happened. It cannot account for every real-world friction (slippage beyond the assumed 0.3%, liquidity constraints on illiquid names, taxes, your own entry/exit timing relative to a signal), and a strategy that performed well over a historical period is not guaranteed to perform similarly in the future.

3. Where the data comes from

None of these are first-party sources we control — SmartHold depends on each of them continuing to publish accurate, timely data.

4. Known limitations

• Prices may be delayed. Live signals and prices shown in the app are sourced from third-party providers and are for informational purposes only — they are not a real-time exchange feed.
• Third-party data can be incomplete or temporarily unavailable. Coverage for a specific exchange or a specific security can occasionally gap or lag for reasons outside our control (an upstream provider outage, a data-vendor change, a newly listed or thinly traded security).
• Mutual fund NAV data depends on a single unofficial source (mfapi.in) with no first-party fallback if it ever becomes unavailable.
• Fundamentals data is only as current as its own refresh cycle and depends on screener.in continuing to publish it.
• Past performance — backtested or real — is never a guarantee of future results.
• SmartHold never places trades. Any buy/sell entries you log are your own personal record-keeping, not a connection to any real brokerage account of yours.

5. Not a SEBI-registered adviser or research analyst

Neither SmartHold nor its operator, Praveen Baid, is registered with the Securities and Exchange Board of India (SEBI) as an Investment Adviser under the SEBI (Investment Advisers) Regulations, 2013, or as a Research Analyst under the SEBI (Research Analysts) Regulations, 2014. Nothing in SmartHold constitutes personalized investment advice. Please consult a SEBI-registered Investment Adviser or Research Analyst before making any investment decision.

6. Questions

Compliance or methodology questions: [email protected]. See also our Terms of Use and Privacy Policy.