Before buying any stock, the bot checks multiple technical signals: RSI, volume spike, price momentum, candle pattern, and more. MIN_SIGNAL_SCORE is how many of these must say "BUY" before the bot actually places a trade.

What it does: Checks whether the NIFTY index (India) or S&P 500 (US) is down by more than the threshold percentage. If the index is down more than that, the bot stops buying new positions for the rest of the day. Existing open positions still close normally via target or SL.

Turn ON when: You want to avoid buying into a falling market. Conservative approach — avoids chasing stocks on overall bad days.

Turn OFF when: You want to trade regardless of the index direction. Individual stocks can go up even when the index is falling. The original $300/4hr profitable session had this OFF.

Threshold: -0.5 = very strict (blocks if index down 0.5%). -1.5 = standard. -3.0 = very loose (only blocks on crash days). Only applies when Mood Filter is ON.

What it does: Limits how many stocks from the same industry sector you can hold open at the same time. With MAX_SECTOR_POSITIONS=2, the bot won't buy a 3rd IT stock if you already hold 2 IT stocks open.

Turn ON when: You want to spread risk across sectors. Protects you from being overexposed if one sector (e.g. Banks) suddenly drops.

Turn OFF when: You want maximum trades. If 5 IT stocks all have strong buy signals, the bot takes all 5. The original profitable session had this OFF. Sector concentration can work in your favor when a sector is on a run.

This is the total number of stocks the bot can hold open simultaneously across all markets. Once this limit is hit, it won't open new positions until existing ones close via target or stop-loss.

5 (default): Conservative. Only 5 open trades total at any time.

10–15: Active. Good for ₹10,000 capital — each position is ~₹500–₹700.

20+: High-volume. More positions means each individual SL hit hurts less. Recommended when you want maximum activity.

Rule of thumb: Keep (Max Positions × average position size) below 80% of your capital. With ₹10,000 and ₹600 avg position, 20 positions = ₹12,000 which is too much. Set target to 15 for ₹600 avg positions.

Sometimes a stock goes flat after you buy — neither hitting the profit target nor the stop-loss. It just sits there, tying up a position slot that could be used for a better opportunity.

Time Exit ON: After MAX_HOLD_MINUTES, close the position at whatever the current price is — profit, loss, or break-even. Frees the slot for new opportunities. Recommended for high-volume trading.

Time Exit OFF: Wait for target or stop-loss naturally. Better for high-confidence setups that just need more time.

MAX_HOLD_MINUTES: India market is 6 hours (9:15 AM – 3:30 PM IST). Setting 30 minutes means any stuck position is closed within 30 min and a new one can open. 90 minutes (default) means a stuck stock ties up a slot for 1.5 hours.

SCALP_TARGET_PCT (default 1.5%): Exit and take profit when the stock gains this much. On a ₹600 position that's ₹9 profit per win.

SCALP_SL_PCT (default 0.6%): Exit and cut the loss when the stock drops this much. On a ₹600 position that's -₹3.60 loss per loss.

Risk/Reward = 2.5:1. You win ₹9 when right, lose ₹3.60 when wrong. You only need to be right about 30% of the time to break even. Being right 50% of the time means every ₹10,000 in daily volume generates profit.

To get more fills faster: Tighten the target (e.g. 1.0%) so positions close quicker. But don't tighten the SL too much — below 0.4% you'll get false exits on normal price noise.

This was broken. It compared cash balance (which drops when you deploy capital) against the starting balance, making the bot think you'd lost 40–50% when you'd actually just bought stocks. This is now fixed — the check only fires when you've actually lost money through closed losing trades.

All position slots are occupied. Bot waits for existing positions to close. Raise MAX_CONCURRENT_POSITIONS to allow more simultaneous trades.

Total realized losses for the day exceeded this %. Default is 2% = ₹200 on ₹10,000 capital. Bot stops buying for the rest of the day.

The bot has already completed N trades (wins + losses combined) today. Default is 30.

NIFTY/S&P 500 dropped more than MOOD_FILTER_THRESHOLD%. Set USE_MOOD_FILTER to OFF to trade regardless of index direction.

Paper mode means the bot simulates trades without spending real money. It tracks virtual positions and P&L exactly as if live — but no real orders are ever placed on Angel One, Alpaca, or Binance.

PAPER_MODE = India • ALPACA_PAPER = US • BINANCE_PAPER = Crypto

The portal intentionally blocks turning these off. To go live, edit the .env file directly on the server and restart. This prevents accidentally going live from the web UI while still testing.

Position size = 5% of current balance per trade. On ₹10,000 balance: 5% = ₹500 per trade. For a stock at ₹250, that buys 2 shares. For a stock at ₹800, it buys 0 shares (can't afford a full lot) — the bot then tries 1 share if it's within 10% of balance.

As your balance grows from winning trades, position sizes grow automatically. As balance drops from losses, position sizes shrink — a built-in risk scaling mechanism.

MAX_DEPLOYED_PCT (default 70%) is the hard cap — bot will never have more than 70% of balance in open positions at any moment. The other 30% stays as cash buffer.