crypto-service · delta exchange india · perpetual futures

Crypto Book

Built 2026-09-29 16:21 IST · positions priced at the venue's mark

Positions

The live book, priced at the venue's mark.

2 open. Every position has a stop recorded. R is measured against the stop the position opened with, not the trailed one; 'locked' is where the trail has pushed the stop, in the same units — positive means the position can no longer lose.

Setups

29 proposal(s) across 5 strategy, 3 already placed. Each table is one rule's own pass over the universe, read back from the journal it wrote — the bracket shown is the one the scanner derived, not a second derivation made here.

volume_breakout [placed automatically] — 1690 evaluated, 3 with a full bracket — refused by no_trend x1567, no_data x80, range_too_wide x17, already_held x14, too_illiquid x9

volume_breakout: the same proposals with their own evidence.

daily_tis [placed automatically] — 1690 evaluated, 0 with a full bracket — refused by no_trend x1266, range_too_wide x330, no_data x80, already_held x14

Nothing qualified. The funnel above says what refused them.

manual_tis_automatic_8_5r [placed on a trigger] — 169 evaluated, 0 with a full bracket — refused by no_trend x161, no_data x8

no eligible automatic plan; no setups from 169 symbols: {'no_trend': 161, 'no_data': 8}

manual_tis_m20 [recorded only] — 507 evaluated, 17 with a full bracket — refused by no_trend x179, too_illiquid x147, too_extended x128, no_data x24, range_too_wide x1

manual_tis_v2 [recorded only] — 507 evaluated, 9 with a full bracket — refused by no_trend x222, too_illiquid x147, too_extended x94, no_data x24, range_too_wide x1

When each scan ran. A missing slot is a scan that did not fire.

Placed automatically — `take-setups` runs at 00:20, 04:20, 08:20, 12:20, 16:20 and 20:20 UTC, sizes each setup at the risk budget and submits it with the stop and target attached. Nothing here waits for a person — these rows are orders the service will place, or has placed, subject to the position, risk and notional caps.

Placed on a trigger — `plan-automatic-tis` ranks one plan at 00:45 UTC and holds it 48 hours; `execute-automatic-tis` checks every hour, arms on a high through the trigger, and then submits an immediate-or-cancel limit. An unfilled order never rests.

Recorded only — `shadow-tis` runs these at 00:40 UTC to measure them. They are not in the registry, so nothing here is placed however good it looks — these rows exist to earn a promotion, not to be traded.

Recent exits

How the last positions actually left.

The last 12 closes. Every one priced from the venue's own fill.

Performance

Realised P&L, by window.

Counted from the rows' own closed_at and realised_pnl.

Funding is not included. `settle-funding` sees the venue's settlements and records none of them — it reads meta_data['symbol'] where Delta sends product_symbol — so every figure above is overstated by whatever each position accrued while it was held. The error grows with holding period.

Market

The regime the entry gates read, and the universe conditions behind it.

BTC daily close 80871.0 is above its 20-day EMA 77227.04980543

What ran

Every scheduled job, and the difference between running on time and working.

12 registered jobs. All reporting ok on their last run.