ALGORITHMIC TRADING
STRATEGY- BREAK OUTS
MARKET INEFFICIENCY & HYPOTHESIS
• The Problem:
• Intraday Index markets are range-bound most of the time, especially on 1-minute resolution.
• Breakouts are rare and explosive, while false breakouts are frequent.
• Most Traders overtrade during low-volatility periods, leading to drawdowns.
• The Hypothesis (EDGE)
• Short-Term price extremes(local highs/local lows) reflect temporary order-flow imbalance.
• While most breakouts fail, directional persistence increases once price accepts beyond recent extremes,
especially when exposure is adjusted dynamically .
• The Edge comes from adapting exposure based on realized market feedback, not from predicting directions.
2
STRATEGY OVERVIEW (1/2)
• Strategy Overview:
• Intraday systematic strategy designed to capture short-term directional moves while remaining market neutral
and adaptive to regime change
• Core Logic
• Operates on high frequency intraday price data (1 min).
• Continuously monitors recent price extremes(LL/LH).
• Initiates positions when price expands beyond short term ranges.
• Scales exposure dynamically in the direction of realized price movement.
3
STRATEGY OVERVIEW (2/2)
• Key Characteristics:
• Trades both long and short, no directional bias.
• Multiple trades per session, no overnight exposure.
• Designed to perform across trending and volatile market phases
• Fully rule based with predefined entry, exit, and risk logic
• Risk Philosophy
• Small initial risk per trade.
• Exposure increases only when market confirms direction.
• Losses are cut shortly, winners are allowed to run
4
MARKET REGIMES & ADAPTIVE BEHAVIOUR
• Intraday markets exhibit multiple regimes including trending, range-bound, and transition phases.
• Traditional breakout strategies perform well in directional regimes but suffer during range bound
conditions due to repeated false signals.
• To address this, the strategy incorporates an adaptive exposure mechanism that dynamically adjust
directional participation based on recent trade outcomes.
• During periods of low directional efficacy, position allocation is reduced, limiting drawdowns. As signal
effectiveness improves, exposure is gradually increased, allowing the strategy to capitalize on emerging
trends.
• This adaptive framework enables robustness across changing intraday regimes rather than reply on static
assumptions
5
RISK MANAGEMENT
• Fixed capital of 100 Cr. Ensures results are comparable across regimes.
• Two Stage Exposure Control:
• Trades start with 25% capital. Full capital is deployed only after trades moves in favor(>=1R)
• Reduces loss during false breakout.
• Capital is scaled after validation, not before conviction.
• Hard-loss limit of 2Cr in a single day, 10 Cr in a month( Gives time to reflect on strategy ).
• No overnight trades, helps avoid gap-up/gap-down holding risk under control.
• Strategy prioritizes capital survival first , convexity second & return 3rd
6
TRADE EXECUTION LOGIC & MARKETS ADAPTIVITY
• Price only breakout detection:
• Signal based of last 10-bar high/low. No indicators no curve or fitted oscillators.
• Designed to respond directly to order-flow expressed via price.
• Symmetric Long & Short Framework:
• Same logic applied to Upside breakouts & Downside breakdowns. No directional bias.
• Performs best on volatile days, trending days and looses less on choppy days.
• Tight initial risk, winner are let loose to trail using evolving HH/LL. Results in cutting losers short and letting
winners expand.( capitalizing breakout )
• Exposure is state-dependent, system behavior changes based on: price confirmation, PnL state,
day/month risk limits. Prevents over trading during unstable conditions.
• The strategy is purposefully designed to stay aligned with market movements, maintaining agility and
remaining completely unbiased about directional trends
7
TRADE STATISTICS & DISTRIBUTION
• Time Period: Jan’15 –Jul’25 | Total trades: 15,794 | Total P&L: ₹854 Cr | Win rate: 49.0% | Profit factor: 3.77.
• Avg / Median P&L per trade: ₹5,40,844 / ₹0 | Std dev: ₹20.8 L (high dispersion) | Skew: ~5.02 | Kurtosis: 52.7.
• Avg win / Avg loss: ₹15L / −₹3.82 lakh | Max drawdown: ~₹1.19 Cr | Longest streaks: Win 13, Loss 13.
• Long Trades : 7,844 | Total P&L: ~386 Cr || Short Trades: 7,950 | Total P&L: 469 Cr.
• Max profit trade: ~173 mins | Max loss trade : ~11mins
• Top 10 profit: 31.5 Cr & Top 10 losses: -2.9 Cr ( winner & loser dominate during high Vol periods: mar’20 & June’24)
• Max loss in a day : -0.54 Cr | Max profit in a day: 8.8 Cr.
• Quantiles (in Cr): 1%: −0.0153 | 5%: −0.0109 | 10%: −0.0073 | Median: 0 |75%: +0.00138 | 90%: +0.0290 | 95%: +0.0463
| 99%: +0.0826
• Distribution is positively skewed: a small fraction of trades (top 1–5%) drives most profits.
• The profile relies on capturing tails while keeping downside controlled.
8
ASSUMPTIONS & PLANNED IMPROVEMENTS
• Assumptions:
• Transaction costs & slippage assumed to be zero, strategy is evaluated on price behavior, not execution
optimization.
• Trades are assumed at bar-close prices, no modeling of: Bid/Ask spread, Queue position, Partial Fills.
• Trades assume sufficient intraday liquidity at Index level.
• These return are upper limits and would reduce once cost and other factors are incorporated.
• Improvements:
• Introduce vol-aware slippage model, incorporate fixed+ variable transaction cost, stress test strategy.
• Measure expectancy degradation under: low cost, medium cost & high cost regimes. Identify breakeven cost
threshold .
• Infrastructure Update: Move to faster research execution stack( Python/R), Enable tick level simulation for
realistic fills, improve logging for execution diagnostics.
• Cost were intentionally excluded to validate signal robustness first; execution realism is the next step forward.
9
THANK YOU