Tidewatch NO EDGE

No robust edge; passes the screens only at ₹15 lakh

At ₹5 lakh nothing passed. At ₹15 lakh, where one futures lot is about 1x leverage and the drawdown limit is ₹3 lakh, 17 of the 44 rules are profitable at 3x costs in both periods within that limit, but the gains are small (about 1% to 5% of capital a year above cash), the trend rules show no timing skill, and the reality check across them gives p = 0.175.

• Data: Nifty 50 daily closes 2007-10 to 2022-12 (public Yahoo Finance index data, cross-checked against Dhan closes), India VIX, and an RBI repo-rate table for carry (estimate). Ranked on 2008-2016, checked on 2017-2022. The 2023 to 2026 hold-out was not loaded.
• Friction: about 3 basis points a round trip (about 7.6 points at 24,000: transaction tax on the sell leg, exchange fee, stamp, brokerage and 0.25 point slippage per fill), multiplied by 1x, 2x and 3x. Holding a long future also pays carry of about 5 to 7% a year (repo less dividend yield). Buy-and-hold futures after carry: +2.5% a year with a 62% drawdown in 2008-2016, +11.3% with 39% in 2017-2022.
• 73% of the 44 rules had a positive Sharpe ratio at 2x costs on the ranking period; a White reality check across the 8 best finalists on the validation period gives p = 0.097 (best t-statistic 1.94). That is suggestive at most, after 44 variants.
• Trend (SMA 100/150) and Donchian 100/50, long or flat: validation Sharpe 0.6 to 0.8 with unit-leverage drawdown 16 to 17%, but only 0.3 to 0.4 and drawdown 27 to 29% in 2008-2016, and the timing t-statistic is negative (-0.5 to -0.7): the gain comes from holding less (less carry), not from timing skill.
• Turn of month (hold the last 2 and first 2 trading days): the only rule positive in both periods (Sharpe 0.60 and 0.57 at 2x, timing t 1.46 and 0.63). On 1 lot: +₹18,600 a year (drawdown ₹73,000) in 2008-2016 and +₹45,200 a year (drawdown ₹159,000, 32% of ₹5 lakh) in 2017-2022; at 3x +₹16,400 and +₹42,100. It fails the 20% limit in the second period. Nine stop-loss and VIX-filter variants did not bring that drawdown under ₹1.5 lakh.
• Options (Black-Scholes on India VIX, a model, not a chain): long ATM calls or puts and ATM call spreads, 8% of equity in premium. Drawdowns of 19% to 90%; train Sharpe at most 0.3. The best validation result (trend call spread, Sharpe 0.9) had a train Sharpe of 0.12, so it reflects the 2017-2021 bull market, not a stable edge.
• CAPITAL RE-TEST (2026-10-20, ₹15,00,000, 1 lot of 65, drawdown limit ₹3,00,000 = 20%): the 44 rules were re-scored in rupees at 1x, 2x and 3x costs. 17 of 44 are profitable at 3x in both periods with drawdown inside the limit. Largest validation gains at 3x (excess over cash): momentum 63-day long/flat +₹77,000 a year, trend SMA100 long/flat +₹64,000, Donchian 100/50 +₹47,000, turn of month (3+3 days) +₹47,000, turn of month (2+2 days) +₹42,000; first-period (2008-2016) gains are much smaller (₹1,000 to ₹22,000 a year). Drawdowns: turn of month 2+2 ₹74,000 and ₹166,000 (5% and 11% of capital), trend SMA100 ₹127,000 and ₹201,000 (8% and 13%). Timing statistics: trend and momentum rules are near zero or negative (they win by holding less, not by timing); turn of month is +1.5 and +0.6. Reality check over the 17 passers on validation at 2x: max t 1.95, p = 0.175. Reading: the drawdown barrier is gone, but what passes is mostly low-leverage equity exposure with a market filter, earning a few percent above cash with no significant timing alpha. Cash on ₹15 lakh at about 4.5% adds about ₹67,000 a year on top.
• Reading: friction is not the problem here; capital is. A 1-lot future carries 3x leverage on ₹5 lakh, and defined-risk options give away too much theta. Realistic version: a small, unlevered holding (Nifty ETF with a trend filter) earns a few percent above cash with about 15% drawdown, but that is risk management, not a trading edge, and is outside the futures/options brief.