Run census

Momentum runs: 29 (~29 per session)
Run length distribution: {2: 2, 3: 3, 4: 2, 5: 2, 6: 1, 7: 1, 8: 1, 9: 1, 10: 1, 11: 3, 13: 1, 14: 1, 17: 4, 19: 2, 20: 4}
Runs lasting <= 2 candles: 2 (6.9%)
Direction at start: {'DOWN': 15, 'UP': 14}
Boundary structure at start: {'none': 29}

Self-consistency

Self-consistency (labeled direction vs the run's own realized drift):
  mismatched runs: 13/29 (44.8%) | by run length: {'>2': 13}

Predictiveness

Predictiveness of run-start direction (signed forward return):
  +5 candles: hit rate 55.2% over 29 runs, average signed return +0.00112
  +10 candles: hit rate 51.9% over 27 runs, average signed return -0.00021
  by starting state (+10 candles):
    STARTING_DOWN      n=   9 hit= 44.4% avg=-0.00041
    STARTING_UP        n=   7 hit= 71.4% avg=+0.00213
    DECELERATING_UP    n=   3 hit= 66.7% avg=+0.00182
    ACCELERATING_DOWN  n=   3 hit= 33.3% avg=-0.00166
    DECELERATING_DOWN  n=   2 hit=  0.0% avg=-0.00983
    CONSOLIDATING      n=   2 hit= 50.0% avg=-0.00091
    ACCELERATING_UP    n=   1 hit=100.0% avg=+0.00404
  by boundary trigger (+10 candles):
    non-structural     n=  27 hit= 51.9% avg=-0.00021

Boundary churn

Boundary churn:
  direction reversals between consecutive runs: 13
  flip-flops (reversal after a run of <= 2 candles): 1
  flip-flops by the reversing run's boundary trigger: {'non-structural': 1}

Change rate

Absolute change_rate at run start (vs MOMENTUM_MIN_CHANGE_RATE=0.0001):
  DOWN           n=  15 median=0.000567 below threshold: 13.3%
  UP             n=  14 median=0.000825 below threshold: 7.1%

Anchor angle

No anchor-angle observations to measure.

Velocity quality

Kinematic velocity quality (recency-weighted velocity / noise, #642), signed +10-candle forward return:
  |q| 0.0-0.1    n=   83 hit= 57.8% avg=+0.00103
  |q| 0.1-0.2    n=   76 hit= 42.1% avg=-0.00023
  |q| 0.2-0.3    n=   76 hit= 48.7% avg=-0.00028
  |q| 0.3-0.4    n=   47 hit= 68.1% avg=+0.00039
  |q| >= 0.4     n=   97 hit= 62.9% avg=+0.00079

Occurrence census

Trend occurrences: 111 (~111.0 per session)
Occurrence length distribution (candles): {1: 99, 2: 6, 10: 1, 16: 1, 47: 1, 54: 1, 60: 2}
Occurrences lasting <= 5 candles: 105 (94.6%)
Direction at start: {'UP': 77, 'DOWN': 34}
Starting state: {'BULLISH_EMERGING': 77, 'BEARISH_EMERGING': 34}

Self-consistency

Self-consistency (labeled direction vs the occurrence's own realized drift):
  mismatched occurrences: 8/111 (7.2%)

Predictiveness

Predictiveness of occurrence-start direction (signed forward return):
  +10 candles: hit rate 40.0% over 110 occurrences, average signed return -0.00189
  +30 candles: hit rate 43.2% over 95 occurrences, average signed return -0.00176
  by starting state (+30 candles):
    BULLISH_EMERGING     n=  61 hit= 59.0% avg=+0.00149
    BEARISH_EMERGING     n=  34 hit= 14.7% avg=-0.00760

Occurrence churn

Occurrence churn:
  direction reversals between consecutive occurrences: 11
  flip-flops (reversal after an occurrence of <= 5 candles): 6

Momentum-trend alignment

Momentum-trend alignment (momentum run starts, signed +5-candle forward return):
  with-trend     n=  17 hit= 58.8% avg=+0.00111
  counter-trend  n=  11 hit= 45.5% avg=+0.00007
  no-trend-yet   n=   1 hit=100.0% avg=+0.01282

Trades: 12 | wins 5 (41.7%) | net +1513.23
Positions (round trips): 12 | wins 5 (41.7%)

Session and cohort net PnL

  tsla_2026-07-06    trades=  12 wins=  5 net=  +1513.23  [training]
  training   ( 1 sessions): trades=  12 wins=  5 net=  +1513.23
  all        ( 1 sessions): trades=  12 wins=  5 net=  +1513.23

Net PnL by session regime

  TREND_UP       ( 1 sessions): trades=  12 wins=  5 net=  +1513.23

By exit path

  STOP_MARKET    n=   7 wins=  0.0% pnl=   -934.77
  TAKE_PROFIT    n=   5 wins=100.0% pnl=  +2448.00

By direction

  DOWN           n=   5 wins= 20.0% pnl=   -321.59
  UP             n=   7 wins= 57.1% pnl=  +1834.82

Hold duration in candles (all closes): {0: 4, 2: 2, 5: 1, 11: 1, 15: 1, 17: 1, 26: 1, 30: 1}
Hold duration in candles (stop-outs):  {0: 4, 5: 1, 11: 1, 17: 1}
Longest hold: 32 wall-clock minutes | cross-session holds: 0 (must stay 0 with the session flush enabled)

Round-trip friction / 1R: median=0.03 p90=0.07 (#612 target: <= 0.20)
Stop distance as fraction of entry price: p10=0.00052 median=0.00150 p90=0.00324
Entry risk/reward ratio: min=2.00 median=2.72 max=5.85
Maximum favorable excursion before stop-out (in R): median=0.00 p75=1.49 | >=0.5R 42.9%, >=1R 28.6% of stopped trades

MAE/MFE in R (closed-candle excursions vs initial risk)

  winners n=   5 | MAE median=0.18R p75=0.86R p90=0.99R | MFE median=2.00R p75=3.24R
  losers  n=   7 | MAE median=1.21R p75=1.26R p90=1.28R | MFE median=0.00R p75=1.88R

Win rate by entry-confidence decile

  0.4-0.5: n=   3 wins= 33.3%
  0.5-0.6: n=   7 wins= 28.6%
  0.6-0.7: n=   2 wins=100.0%

Entry expectancy by 30-minute session bucket (minutes since open)

     0-30   n=  3 wins=  67% net= +1117.41 meanR=+0.75
    30-60   n=  1 wins=   0% net=  -118.20 meanR=-1.03
    60-90   n=  1 wins=   0% net=  -327.83 meanR=-1.01
    90-120  n=  1 wins= 100% net=  +502.67 meanR=+1.66
   120-150  n=  2 wins=  50% net=  +163.12 meanR=+1.58
   240-270  n=  2 wins=  50% net=  +292.96 meanR=+0.79
   270-300  n=  1 wins=   0% net=   -46.88 meanR=-1.08
   330-360  n=  1 wins=   0% net=   -70.03 meanR=-1.05

Churn checks (both should stay at 0 after the 2026-07-08 fixes)

  REDUCE fills on in-profit positions: 0 (of 0 REDUCE fills)
  Same-direction re-entries within 3 candles of a stop-out: 0 (of 7 stop-outs)

Bracket vs realized outcome

Filled candidates: 12
Win-direction agreement rate: 75.0%
Realized-R mean absolute error: 1.146

Regime

Regime: TREND_UP
Signed efficiency: 0.1503
Mean absolute return: 0.0008
Flip density: 0.5389