How much volume does LB6 trade, and how often does it point one way?
FRAME
- Subject: LB6 Config B, 6 BTC engines (3 long-only, 3 short-only); simulated tradelist.
- Why it matters: pak Yon's grid request judges grids by traded volume (rebates) and wants LB6 to set the grid's side. This measures LB6's own volume and how often it has a net direction at all.
- Scope: trades entered before 2026-02-01 (201 trades); holdout untouched (
docs/data-policy.md). Descriptive, no claim. Method locked indocs/bets/2026-11-18-lb6-volume-baseline.md, commit4bcda3c, written before computing. - Terms: volume per 1,000 USDT position = 2,000 USDT per trade (open + close). Net direction per minute = long engines holding minus short engines holding; long > 0, short < 0, flat = 0.
MEASURES
| measure | value |
|---|---|
| window (days) | 283.6 (2025-04-23 to 2026-01-31) |
| trades | 201 |
| trades per 30 days | 21.3 |
| volume per 30 days per 1,000 USDT position | 42,521 USDT |
| share of time net long | 27.3% |
| share of time net short | 30.5% |
| share of time flat | 42.3% |
| median net-long stretch | 12.0 h |
| median net-short stretch | 23.0 h |
| direction changes per 30 days (any) | 30.3 |
| long-to-short or short-to-long flips per 30 days | 0.4 |
Per engine:
| engine_id | side | trades | trades_per_30d |
|---|---|---|---|
| FOCUSED_TOP_SHARPE | short | 22 | 2.3 |
| deepdw_iv_M04 | long | 25 | 2.6 |
| deepput_strike_C14_G09 | long | 34 | 3.6 |
| dw_tail_C04 | short | 26 | 2.8 |
| iv_deepcall_0804_SL8 | long | 34 | 3.6 |
| netdw_skew_C01 | short | 60 | 6.3 |
UPDATE
- Status: MEASURED.
- What it means: an LB6-directed grid would run 58% of the time and sit flat 42%; it would restart about 30 times per 30 days.
- Next bets:
grid-strike-lb6(Is a BTC grid whose side follows LB6 more efficient than the better plain grid on the holdout?)