PHASE 3.08 SECOND INTERVENTION RESULTS

PRIMARY
Average ending bankroll Textbook/Selective: 99.956 / 90.594
Maximum bankroll observed Textbook/Selective: 126.0 at hand 29 (session 1777) / 130.0 at hand 30 (session 1543)
Positive sessions Textbook/Selective: 48.14% / 16.14%
Paired winner Textbook/Selective/tie: 80.26% / 18.42% / 1.32%
Doubles Textbook/Selective: 15716 / 115659; splits Textbook/Selective: 3808 / 21673
Mean cards consumed Textbook/Selective: 165.674 / 174.385
Ending bankroll >=2x (200) Textbook/Selective: 0.00% / 0.00%
Ending bankroll >=3x (300) Textbook/Selective: 0.00% / 0.00%
Ending bankroll >=5x (500) Textbook/Selective: 0.00% / 0.00%
Ending bankroll >=10x (1000) Textbook/Selective: 0.00% / 0.00%

HOLDOUT
Average ending bankroll Textbook/Selective: 99.972 / 90.478
Maximum bankroll observed Textbook/Selective: 122.5 at hand 24 (session 1273) / 132.0 at hand 30 (session 3855)
Positive sessions Textbook/Selective: 47.42% / 16.14%
Paired winner Textbook/Selective/tie: 79.90% / 18.42% / 1.68%
Doubles Textbook/Selective: 15412 / 115491; splits Textbook/Selective: 3747 / 21686
Mean cards consumed Textbook/Selective: 165.647 / 174.221
Ending bankroll >=2x (200) Textbook/Selective: 0.00% / 0.00%
Ending bankroll >=3x (300) Textbook/Selective: 0.00% / 0.00%
Ending bankroll >=5x (500) Textbook/Selective: 0.00% / 0.00%
Ending bankroll >=10x (1000) Textbook/Selective: 0.00% / 0.00%


PHASE 3.08 — SELECTIVE-AGGRESSION INSTRUMENTATION SUMMARY

PRIMARY resolved selective-aggression hands: 171673; mean net/hand -0.2739; mean committed stake 1.6737; bust rate 17.93%; third-card bust rate when a third card was taken 26.58%.
Textbook would have chosen on those same visible first-decision states: HIT 69943, STAND 68255, DOUBLE 15787.

HOLDOUT resolved selective-aggression hands: 171686; mean net/hand -0.2773; mean committed stake 1.6727; bust rate 18.01%; third-card bust rate when a third card was taken 26.73%.
Textbook would have chosen on those same visible first-decision states: HIT 70301, STAND 67744, DOUBLE 15798.

INTERPRETATION
Phase 3.08 carries forward split-child HARD 20 STAND, HARD 19 STAND, and split-child HARD 18 STAND. It adds one new family only: HARD 17 now STANDS instead of automatically DOUBLING. All other Maximum Aggression behaviour remains frozen.

PRIMARY — WORST SELECTIVE-AGGRESSION VISIBLE STATES BY MEAN NET/HAND
Rank | Visible state | Hands | Mean net | Loss% | Bust% | 3rd-card bust% of 3rd-card draws | Mean stake
 1 | HARD 5 vs 11 [split-child] | textbook=HIT     |     15 |  -2.000 | 100.00% |   0.00% |   0.00% | 2.000
 2 | HARD 4 vs 9 [split-child] | textbook=HIT      |     12 |  -1.667 |  91.67% |   0.00% |   0.00% | 2.000
 3 | HARD 4 vs 8 [split-child] | textbook=HIT      |      9 |  -1.556 |  88.89% |   0.00% |   0.00% | 2.000
 4 | HARD 4 vs 11 [split-child] | textbook=HIT     |      9 |  -1.556 |  88.89% |   0.00% |   0.00% | 2.000
 5 | HARD 5 vs 11 | textbook=HIT                   |     92 |  -1.478 |  86.96% |   0.00% |   0.00% | 2.000
 6 | HARD 6 vs 9 [split-child] | textbook=HIT      |     23 |  -1.478 |  86.96% |   0.00% |   0.00% | 2.000
 7 | SOFT 15 vs 11 [split-child] | textbook=HIT    |     10 |  -1.400 |  80.00% |   0.00% |   0.00% | 2.000
 8 | HARD 6 vs 11 | textbook=HIT                   |    102 |  -1.353 |  83.33% |   0.00% |   0.00% | 2.000
 9 | HARD 5 vs 9 | textbook=HIT                    |    127 |  -1.339 |  83.46% |   0.00% |   0.00% | 2.000
10 | SOFT 16 vs 3 [split-child] | textbook=HIT     |     18 |  -1.333 |  77.78% |   0.00% |   0.00% | 2.000
11 | HARD 5 vs 10 [split-child] | textbook=HIT     |     89 |  -1.326 |  83.15% |   0.00% |   0.00% | 2.000
12 | HARD 6 vs 11 [split-child] | textbook=HIT     |     29 |  -1.310 |  82.76% |   0.00% |   0.00% | 2.000
13 | HARD 5 vs 9 [split-child] | textbook=HIT      |     21 |  -1.238 |  80.95% |   0.00% |   0.00% | 2.000
14 | HARD 7 vs 11 [split-child] | textbook=HIT     |     33 |  -1.212 |  78.79% |   0.00% |   0.00% | 2.000
15 | HARD 7 vs 11 | textbook=HIT                   |    198 |  -1.202 |  76.77% |   0.00% |   0.00% | 2.000
16 | HARD 4 vs 7 [split-child] | textbook=HIT      |     10 |  -1.200 |  80.00% |   0.00% |   0.00% | 2.000
17 | HARD 7 vs 9 [split-child] | textbook=HIT      |     32 |  -1.188 |  75.00% |   0.00% |   0.00% | 2.000
18 | HARD 7 vs 8 [split-child] | textbook=HIT      |     38 |  -1.158 |  73.68% |   0.00% |   0.00% | 2.000
19 | HARD 5 vs 10 | textbook=HIT                   |    486 |  -1.128 |  78.19% |   0.00% |   0.00% | 2.000
20 | HARD 16 vs 10 [split-child] | textbook=HIT    |    839 |  -1.106 |  74.26% |  61.50% |  61.50% | 2.000
21 | HARD 6 vs 10 [split-child] | textbook=HIT     |    118 |  -1.102 |  77.12% |   0.00% |   0.00% | 2.000
22 | HARD 15 vs 9 [split-child] | textbook=HIT     |    275 |  -1.098 |  74.18% |  56.00% |  56.00% | 2.000
23 | HARD 14 vs 11 [split-child] | textbook=HIT    |    177 |  -1.096 |  75.14% |  49.15% |  49.15% | 2.000
24 | HARD 16 vs 9 | textbook=HIT                   |    650 |  -1.086 |  74.31% |  65.38% |  65.38% | 2.000
25 | HARD 16 vs 11 | textbook=HIT                  |    464 |  -1.078 |  73.71% |  62.50% |  62.50% | 2.000

HOLDOUT — WORST SELECTIVE-AGGRESSION VISIBLE STATES BY MEAN NET/HAND
Rank | Visible state | Hands | Mean net | Loss% | Bust% | 3rd-card bust% of 3rd-card draws | Mean stake
 1 | HARD 4 vs 2 [split-child] | textbook=HIT      |      7 |  -2.000 | 100.00% |   0.00% |   0.00% | 2.000
 2 | HARD 4 vs 5 [split-child] | textbook=HIT      |      3 |  -2.000 | 100.00% |   0.00% |   0.00% | 2.000
 3 | SOFT 15 vs 11 [split-child] | textbook=HIT    |     15 |  -1.600 |  86.67% |   0.00% |   0.00% | 2.000
 4 | SOFT 13 vs 2 [split-child] | textbook=HIT     |     14 |  -1.571 |  85.71% |   0.00% |   0.00% | 2.000
 5 | HARD 7 vs 11 [split-child] | textbook=HIT     |     26 |  -1.538 |  88.46% |   0.00% |   0.00% | 2.000
 6 | HARD 5 vs 8 [split-child] | textbook=HIT      |     22 |  -1.455 |  86.36% |   0.00% |   0.00% | 2.000
 7 | HARD 4 vs 10 [split-child] | textbook=HIT     |     35 |  -1.429 |  85.71% |   0.00% |   0.00% | 2.000
 8 | HARD 5 vs 11 [split-child] | textbook=HIT     |      7 |  -1.429 |  85.71% |   0.00% |   0.00% | 2.000
 9 | SOFT 17 vs 11 [split-child] | textbook=HIT    |     17 |  -1.412 |  82.35% |   0.00% |   0.00% | 2.000
10 | HARD 4 vs 9 [split-child] | textbook=HIT      |     13 |  -1.385 |  84.62% |   0.00% |   0.00% | 2.000
11 | HARD 6 vs 9 [split-child] | textbook=HIT      |     41 |  -1.366 |  82.93% |   0.00% |   0.00% | 2.000
12 | HARD 7 vs 11 | textbook=HIT                   |    204 |  -1.304 |  80.88% |   0.00% |   0.00% | 2.000
13 | HARD 6 vs 11 [split-child] | textbook=HIT     |     16 |  -1.250 |  81.25% |   0.00% |   0.00% | 2.000
14 | HARD 6 vs 11 | textbook=HIT                   |    100 |  -1.220 |  80.00% |   0.00% |   0.00% | 2.000
15 | HARD 5 vs 6 [split-child] | textbook=HIT      |     20 |  -1.200 |  80.00% |   0.00% |   0.00% | 2.000
16 | HARD 6 vs 8 [split-child] | textbook=HIT      |     38 |  -1.158 |  78.95% |   0.00% |   0.00% | 2.000
17 | HARD 5 vs 9 [split-child] | textbook=HIT      |     19 |  -1.158 |  78.95% |   0.00% |   0.00% | 2.000
18 | HARD 5 vs 10 | textbook=HIT                   |    486 |  -1.144 |  78.60% |   0.00% |   0.00% | 2.000
19 | HARD 5 vs 11 | textbook=HIT                   |     93 |  -1.140 |  78.49% |   0.00% |   0.00% | 2.000
20 | HARD 16 vs 4 [split-child] | textbook=STAND   |    218 |  -1.119 |  76.15% |  64.68% |  64.68% | 2.000
21 | HARD 4 vs 8 [split-child] | textbook=HIT      |      9 |  -1.111 |  77.78% |   0.00% |   0.00% | 2.000
22 | HARD 16 vs 10 [split-child] | textbook=HIT    |    843 |  -1.106 |  74.38% |  61.92% |  61.92% | 2.000
23 | HARD 16 vs 9 [split-child] | textbook=HIT     |    236 |  -1.093 |  74.58% |  59.32% |  59.32% | 2.000
24 | HARD 16 vs 9 | textbook=HIT                   |    669 |  -1.088 |  74.89% |  60.39% |  60.39% | 2.000
25 | HARD 16 vs 10 | textbook=HIT                  |   2440 |  -1.080 |  74.14% |  62.70% |  62.70% | 2.000


PHASE 3.08 — REMAINING REPLICATED DAMAGE STATES AFTER HARD 17 FAMILY

Filter: same visible state appears >=100 times in BOTH primary and holdout, and Selective Aggression has negative mean net/hand in BOTH cohorts. This avoids choosing later interventions from tiny-sample extremes.

Rank | State | Primary n / net | Holdout n / net | P bust% / H bust% | Textbook first action
 1 | HARD 6 vs 11                       |   102 / -1.353 |   100 / -1.220 |   0.00% /   0.00% | HIT
 2 | HARD 7 vs 11                       |   198 / -1.202 |   204 / -1.304 |   0.00% /   0.00% | HIT
 3 | HARD 5 vs 9                        |   127 / -1.339 |   146 / -1.014 |   0.00% /   0.00% | HIT
 4 | HARD 5 vs 10                       |   486 / -1.128 |   486 / -1.144 |   0.00% /   0.00% | HIT
 5 | HARD 16 vs 10 [split-child]        |   839 / -1.106 |   843 / -1.106 |  61.50% /  61.92% | HIT
 6 | HARD 16 vs 9                       |   650 / -1.086 |   669 / -1.088 |  65.38% /  60.39% | HIT
 7 | HARD 16 vs 4 [split-child]         |   241 / -1.021 |   218 / -1.119 |  64.73% /  64.68% | STAND
 8 | HARD 16 vs 11                      |   464 / -1.078 |   496 / -1.040 |  62.50% /  62.50% | HIT
 9 | HARD 6 vs 10                       |   483 / -1.077 |   490 / -1.024 |   0.00% /   0.00% | HIT
10 | HARD 16 vs 10                      |  2596 / -1.020 |  2440 / -1.080 |  60.67% /  62.70% | HIT
11 | HARD 16 vs 5 [split-child]         |   254 / -1.063 |   223 / -1.031 |  62.20% /  63.68% | STAND
12 | HARD 14 vs 11 [split-child]        |   177 / -1.096 |   169 / -0.982 |  49.15% /  44.97% | HIT
13 | HARD 15 vs 9 [split-child]         |   275 / -1.098 |   236 / -0.949 |  56.00% /  55.08% | HIT
14 | HARD 14 vs 10 [split-child]        |   951 / -1.020 |   975 / -1.017 |  45.95% /  46.26% | HIT
15 | HARD 6 vs 9                        |   154 / -0.961 |   140 / -1.071 |   0.00% /   0.00% | HIT
16 | HARD 15 vs 10                      |  3041 / -1.065 |  3035 / -0.964 |  54.62% /  53.61% | HIT
17 | HARD 5 vs 8                        |   162 / -1.012 |   121 / -0.975 |   0.00% /   0.00% | HIT
18 | HARD 16 vs 3                       |   774 / -0.990 |   692 / -0.994 |  63.82% /  62.14% | STAND
19 | HARD 15 vs 9                       |   829 / -0.992 |   849 / -0.989 |  52.11% /  57.13% | HIT
20 | HARD 13 vs 9 [split-child]         |   247 / -0.907 |   245 / -1.053 |  39.68% /  46.12% | HIT

DESIGN USE
Phase 3.08 carries forward the validated split-child HARD 20, HARD 19 and split-child HARD 18 protections and replaces automatic DOUBLE in one coherent family: HARD 17 -> STAND. Remaining replicated damage states are retained for later one-rule-at-a-time interventions.

PHASE 3.08 — RUNNING TXT AUDIT RECORD

OBJECTIVE
Make the fourth controlled stabilisation intervention while preserving the frozen architecture. Carry forward the three validated protections, then add one new policy family: HARD 17 STANDS rather than taking Maximum Aggression's automatic DOUBLE.

EVIDENCE HIERARCHY
1. Primary design authority: finite six-deck Phase 2.xx evidence and new Phase 3.xx data.
2. Earlier non-depleting/infinite-generator behaviour-profile data: historical corroboration and hypothesis generation only.
3. Old behaviour-profile evidence must not directly define a new policy unless finite-shoe evidence independently supports the same conclusion.

FROZEN VARIABLES
Six decks, deterministic session shuffle, 30 hands, 100-unit start, 1-unit base wager, dealer rules, blackjack payout, split limit, no in-session reshuffle, textbook control and Maximum Aggression policy are unchanged from Phase 2.02.

POLICY INFORMATION BOUNDARY
The future optimiser may use player hand composition/total, pair status, dealer upcard and legal actions. Phase 3.08 deliberately uses no previous-card-history, running count, remaining-rank inventory, hidden card or future-card feature. The accumulated policy uses only current visible states: split-child HARD 20, HARD 19, HARD 18, split-child HARD 16 vs Ace, and HARD 17. No card-history or composition feature is introduced.

PHASE 3.08 DECISION RULE
Carry forward split-child HARD 20 STAND, HARD 19 STAND and split-child HARD 18 STAND. ONE additional Phase 3.08 intervention: if an ordinary first-decision hand is hard (not soft) and totals 18, Selective Aggression STANDS. Otherwise the frozen Maximum Aggression policy is unchanged. Textbook remains the control.

PHASE 3.08 REPLAY FINGERPRINT: e01a9f9dbcf55332a525cf42db7e78ae8b353ae5a9f8d73f11200dd5338d328f
