C0: paired D3 N7M6 minus D4 s5 contrast from retained per-game records
On the retained drop7-lifetime-cohort-v1 per-game records for D3 N7M6 (runs/RUN-A525-reveal/d3-n7-m6.json) and D4 s5 (runs/RUN-A51D-s7confirm/fresh-s5.json), both covering seeds 0xa51d1000-0xa51d103f, the paired score delta has a one-sided 95% bootstrap lower bound and a Student-t lower bound both above zero.
On this page
- Created
- Updated
No explanation has been written for this record yet.
Technical recordThe registered protocol
- Hypothesis
- On the retained drop7-lifetime-cohort-v1 per-game records for D3 N7M6 (runs/RUN-A525-reveal/d3-n7-m6.json) and D4 s5 (runs/RUN-A51D-s7confirm/fresh-s5.json), both covering seeds 0xa51d1000-0xa51d103f, the paired score delta has a one-sided 95% bootstrap lower bound and a Student-t lower bound both above zero.
- Arms
Arm Name Entry point Manifest Candidate D3 N7M6: factored-chance-fair-search, depth 3, discSamples 7, revealSamples 6, terminalUtility -1000000, maximumWork 51084852, 2000-move cap, corrected 17,000-point Hardcore scoring (finding-16 arm, RUN-20260821T035407Z-00483c6c) runs/RUN-20260823T191900Z-b9f8f80d/c0/c0_compare.py– Comparator D4 s5: parameterized-fair-search, depth 4, chanceSamples 5 (N5 M1), terminalUtility -1000000, maximumWork 3200000, 2000-move cap, corrected 17,000-point Hardcore scoring (finding-05 fresh-s5 arm, runs/RUN-A51D-s7confirm/fresh-s5.json; byte-identical games to runs/RUN-A52-LEAF/eval/s5-w000.json) approaches/fair-expectimax/reference/fair-only-depth4.cppresearch/benchmarks/baselines-v1.json- Classification
- diagnostic
- Information boundary
- seed-free
- Benchmark tier
- CHECK
- Lifecycle
- completed
- Theories tested
- Primary metric
- paired mean whole-game score delta, D3 N7M6 minus D4 s5, 64 games, seeds 0xa51d1000-0xa51d103f
- Secondary metrics
- paired mean moves delta
- first-half / second-half deltas in seed order (seeds 0xa51d1000-0xa51d101f and 0xa51d1020-0xa51d103f)
- Q25 delta (marginal), median, min, max per arm
- W-T-L
- detection floor 1.645*sd/sqrt(64)
- paired numbered clears and cover reveals deltas
- work-per-move ratio
- the same contrast D3 N7M6 minus D4 s7 (fresh-s7.json), reported as a diagnostic, not gated
- Statistical unit
- whole-game
- Uncertainty method
- one-sided 95% percentile bootstrap over whole games (20,000 resamples, numpy PCG64 seed 0xB0071EAF, via the paired() function of approaches/lifetime-objective/chain-reveal-leaf/compare.py imported unchanged by the c0_compare.py wrapper) and Student-t one-sided 95% lower bound; detection floor reported
- Data role
- previously-evaluated-development
- Seed leases
- none recorded
- Dataset references
runs/RUN-A525-reveal/d3-n7-m6.json (sha256 36aa1d0f0ec29b0ec70bd540e7d06758e37147a521f6891c0fde930943aa83fc)runs/RUN-A51D-s7confirm/fresh-s5.json (sha256 d6c1dbd403dbf5b477f1b0905b4aa440e443c2dfba8eef29085b7f5bee024d9a)runs/RUN-A51D-s7confirm/fresh-s7.json (sha256 0c95442620a78ef5a928a4b8cdab9f57b3bdb8a3d7df5fcd1a6d818785bd4209; secondary D4 s7 contrast only)
- Whole-origin split
- yes
- Reuse disclosure
- Reanalysis only: no game is played, no seed is opened, no lease is consumed. Both arms were played under SEEDLEASE-A51D on the already-read shared development cohort 0xa51d1000-0xa51d103f (exploratory-development-diagnostic). The arm MEANS (376,442 and 297,327) and their difference (+79,115) were published in finding-16 and finding-05 and were known to the author before this gate was written, as was the finding-16 contrast D3 N7M6 minus D4 N7 M1 (-22,056). What was never computed or printed is the paired per-game D3 N7M6 minus D4 s5 delta and its bounds. This is therefore a bound computation on already-read data whose sign is already known; it is diagnostic tier only and its pass cannot be promoted above public-development on this cohort. The bootstrap seed, resample count and analysis code are pinned in expectedArtifacts before the wrapper is run.
- Pass criteria
- Both arms are drop7-lifetime-cohort-v1 artifacts covering exactly the same 64 seeds 0xa51d1000-0xa51d103f in order, 0 censored games, 2000-move cap, terminalUtility -1000000, seedLease SEEDLEASE-A51D.
- D3 N7M6 minus D4 s5 paired score delta: percentile bootstrap 95% lower bound > 0 AND Student-t 95% lower bound > 0.
- On pass
- Record valid + pass at public-development tier on already-read data; no tier above CHECK/diagnostic is claimed because the point estimate preceded the gate; any promotion requires a fresh-development cohort under a new protocol.
- On fail
- Record valid + fail; the theory is not-supported-as-tested at public-development tier; the +79,115 difference of means may not be quoted as a head start in the K2 program without the bound attached; open no cohort.
- Gate fixed before controlled data
- yes
- Resources
Wall seconds 60 CPU threads 1 Max host bytes 1073741824 Max GPU bytes – GPU devices – - Stop conditions
- Stop on any rules, information-boundary, legality, determinism, or parity failure.
- Stop and record nothing beyond c0.md if the seed lists of the two arms differ or either artifact is not drop7-lifetime-cohort-v1.
- Expected artifacts
runs/RUN-20260823T191900Z-b9f8f80d/c0/inputs-manifest.json (sha256 of every file read)runs/RUN-20260823T191900Z-b9f8f80d/c0/compare-d3n7m6-vs-d4s5.jsonruns/RUN-20260823T191900Z-b9f8f80d/c0/compare-d3n7m6-vs-d4s7.jsonruns/RUN-20260823T191900Z-b9f8f80d/c0/c0.mdsha256 runs/RUN-20260823T191900Z-b9f8f80d/c0/c0_compare.py f25581560c2ffe1e5dcab10d2fdeab06dc97c1e879cbf18a447bdb3d5b5636e5sha256 approaches/lifetime-objective/chain-reveal-leaf/compare.py 2994995e12687a512173758a9ff11a43797ef20378c0ecdde1fedbd3be23bb18sha256 approaches/lifetime-objective/leaf-evolution/compare.py d40d28b4cd78c40b6cb22b289203d633e44c3d8a98a2b70e65a1623d7b8a670b
- Amendments
Timestamp Before controlled data Reason 2026-08-23T19:38:45Z yes c0_compare.py as pinned (sha256 f25581560c2ffe1e5dcab10d2fdeab06dc97c1e879cbf18a447bdb3d5b5636e5) failed to import at all: chain-reveal-leaf/compare.py performs `import compare as le` for the leaf-evolution resampler, so importing it under the module name `compare` resolved that statement to the half-initialised module. The wrapper now loads compare.py via importlib under the distinct module name chain_reveal_leaf_compare; no statistic, seed, or resample count changed. Amended sha256 b26715431715a3f848f227357c163dbc2a9bde73874cc8b835cdafbb94e94ac0. No compare output existed before this amendment (both earlier invocations died at import, see c0.md). 2026-08-23T19:38:51Z no lifecycle advanced preregistered -> completed after the wrapper ran and the run and result records were written; protocol content otherwise unchanged. protocolSha256 recomputed over the amended, completed record (repository practice: the run record retains the pre-data freeze hash b276634a2195f78e89c999e5f8b79b3e1fa18d58d3634d6fcea127b4e3de935e).
Technical recordResults recorded against this protocol
C0 reanalysis (K2 program section 8 row 0): the paired per-game bound behind the +79,115 head start is now printed, and it clears zero. On the 64 shared seeds 0xa51d1000-0xa51d103f, depth 3 with seven disc strata and six reveal samples (fair leaf, finding-16) scores 376,442 against fair depth 4 at five strata's 297,327 (finding-05): paired +79,115, one-sided 95% percentile bootstrap lower bound +29,033 (upper +129,722; 20,000 resamples, seed 0xb0071eaf), Student-t lower bound +27,548, W-T-L 35-0-29, halves +88,138 / +70,092, Q25 +34,872 (227,224 vs 192,352), median 322,859 vs 260,415, median paired delta +32,926, moves +22.30 (LB +8.67), paired sd 247,113, detection floor 50,813, at 3.27x the logical work per move (4,244,020 vs 1,296,034). Both gate bounds are positive, so the experiment passes and the theory is supported-as-tested at public-development tier. READ THE CAVEATS: the means and their difference were published before this gate was written, so this is a bound attached to a known sign on already-read development data, not a discovery; it is not promotable above this tier on this cohort. The delta is heavy-tailed: the five largest paired gains (seeds 0xa51d1012 +747,416, 0xa51d1020 +684,129, 0xa51d1008 +643,799, 0xa51d1001 +643,344, 0xa51d1033 +558,163) carry 64.7% of the summed delta, the sixteen largest carry 134% (the remaining 48 games net negative), and the worst loss is -332,950 (0xa51d103c); the minimum leave-one-out mean is still +68,507, so no single game is load-bearing, but against a detection floor of 50,813 the mean sits 1.56 floors above zero and the bootstrap lower bound only 0.57 floors above it. The flow statistics move with the score: numbered clears per move 2.0447 vs 1.9489, cover reveals per move 1.1423 vs 1.0697 (paired clears +53.9 per game, LB +21.1; reveals +31.8, LB +12.6), mean occupancy 23.49 vs 24.29. SECONDARY, not gated: against fair depth 4 at seven strata (398,498) the same arm is -22,056 with bounds (-92,005, +45,490), t lower bound -92,330, W-T-L 30-0-34, halves -27,121 / -16,991, Q25 +14,360, moves -5.20, at 0.86x the work; this reproduces finding-16's -22,056 point estimate exactly and its bootstrap lower bound to within 2,138 (finding-16 printed -89,867 from its own resampler; this run's pinned seed gives -92,005), and remains a wash. So the head start the K2 fallback route (c) stands on is real against D4 s5 but the same arm does not separate from D4 s7, which is the stronger comparator and the one the program's end state must beat on the strength-cost frontier.
- ✓Both arms are drop7-lifetime-cohort-v1 artifacts covering exactly the same 64 seeds 0xa51d1000-0xa51d103f in order, 0 censored games, 2000-move cap, terminalUtility -1000000, seedLease SEEDLEASE-A51D — observed: seed lists identical and in order for d3-n7-m6.json, fresh-s5.json, fresh-s7.json and s5-w000.json; censoredGames 0 in all; maximumMoves 2000; terminalUtility -1000000; seedLease SEEDLEASE-A51D; dataRole exploratory-development-diagnostic. Work bounds differ by design (51,084,852 vs 3,200,000 vs 16,000,000): they are the arms' own budgets, not a scoring setting. fresh-s5.json and s5-w000.json carry identical scores and moves on every game.
- ✓D3 N7M6 minus D4 s5: percentile bootstrap 95% lower bound > 0 AND Student-t 95% lower bound > 0 — observed: bootstrap LB +29,033, Student-t LB +27,548 (t quantile 1.6694, df 63); mean +79,115; W-T-L 35-0-29; halves +88,138 / +70,092
Technical recordRecorded metrics
- artifact
- runs/RUN-A525-reveal/d3-n7-m6.json
- sha256
- 36aa1d0f0ec29b0ec70bd540e7d06758e37147a521f6891c0fde930943aa83fc
- meanScore
- 376442.1406
- median
- 322,859
- q25
- 227224.5000
- min
- 103,015
- max
- 1,002,557
- sd
- 223365.7773
- meanMoves
- 109.4531
- workPerMove
- 4244020.0805
- maximumWork
- 51,084,852
- censoredGames
- 0
- gamesAtOrAboveMillion
- 1
- artifact
- runs/RUN-A51D-s7confirm/fresh-s5.json
- sha256
- d6c1dbd403dbf5b477f1b0905b4aa440e443c2dfba8eef29085b7f5bee024d9a
- meanScore
- 297327.3906
- median
- 260,415
- q25
- 192,352
- min
- 86,935
- max
- 836,427
- sd
- 150549.5574
- meanMoves
- 87.1563
- workPerMove
- 1296033.6877
- maximumWork
- 3,200,000
- censoredGames
- 0
- gamesAtOrAboveMillion
- 0
- artifact
- runs/RUN-A51D-s7confirm/fresh-s7.json
- sha256
- 0c95442620a78ef5a928a4b8cdab9f57b3bdb8a3d7df5fcd1a6d818785bd4209
- meanScore
- 398498.2344
- median
- 344630.5000
- q25
- 212,864
- meanMoves
- 114.6563
- workPerMove
- 4956614.2652
- maximumWork
- 16,000,000
- censoredGames
- 0
- n
- 64
- meanScoreDelta
- 79114.7500
- pairedSd
- 247112.6017
- bootstrapLower95
- 29032.5102
- bootstrapUpper95
- 129721.6797
- studentTLower95
- 27548.4592
- studentTQuantile
- 1.6694
- detectionFloor
- 50812.5287
- winTieLoss
- 35-0-29
- firstHalfMeanDelta
- 88137.8438
- secondHalfMeanDelta
- 70091.6563
- q25Delta
- 34872.5000
- medianPairedDelta
- 32925.5000
- minLeaveOneOutMeanDelta
- 68506.7900
- top5Share
- 0.6470
- top16Share
- 1.3400
- meanMovesDelta
- 22.2969
- movesBootstrapLower95
- 8.6719
- movesStudentTLower95
- 8.2664
- numberedClearedDelta
- 53.9375
- numberedClearedLower95
- 21.1406
- coversRevealedDelta
- 31.7969
- coversRevealedLower95
- 12.6094
- workRatio
- 3.2746
- bootstrapSeedHex
- 0xb0071eaf
- resamples
- 20,000
- n
- 64
- meanScoreDelta
- -22056.0938
- pairedSd
- 336763.8340
- bootstrapLower95
- -92004.8930
- bootstrapUpper95
- 45490.1641
- studentTLower95
- -92330.3803
- detectionFloor
- 69247.0634
- winTieLoss
- 30-0-34
- firstHalfMeanDelta
- -27121.1250
- secondHalfMeanDelta
- -16991.0625
- q25Delta
- 14360.5000
- meanMovesDelta
- -5.2031
- movesBootstrapLower95
- -24.0313
- workRatio
- 0.8562
- finding16PointEstimate
- -22,056
- finding16BootstrapLower95
- -89,867
- Bound computation on already-read development data whose means and sign were published (finding-16, finding-05) before the gate was written; diagnostic, not promotable above public-development on this cohort. A fresh-development replication under a new protocol is required before the +79k is used as anything but a planning prior.
- The two arms were played in different runs (RUN-20260821T035407Z-00483c6c for D3 N7M6; the finding-05 fresh-s5 arm) by different binaries (factored-chance-fair-search vs parameterized-fair-search); the seed lists, move cap, terminal utility and scoring are identical but the pairing is across builds, not within one runner invocation. Engine parity between these families was established separately (finding-09/16 checks; finding-15 engine control), not re-run here.
- Heavy tail: five games carry 64.7% of the summed delta; the bootstrap lower bound sits 0.57 detection floors above zero. 64 games cannot resolve anything below about 50,800.
- The arm does not separate from D4 s7 (-22,056, bounds -92,005 to +45,490), so 'beats D4 s5' does not transfer to 'beats the current best fair comparator'.
- The wrapper pinned at freeze failed to import (module-name collision) and was amended before any output existed; the amendment changed no statistic. The first pinned wrapper hash is retained in expectedArtifacts and the amended hash in amendments.
Agent contextHow to extend this record
To add a reader-facing explanation, write web/content/research/EX-20260823-d3n7m6-vs-d4s5-paired-reanalysis-ea66f4ec.mdx; it renders above this record on the next request. The registered protocol itself is in the technical record above.
Record file: research/experiments/EX-20260823-d3n7m6-vs-d4s5-paired-reanalysis-ea66f4ec.json, validated against research/schemas/experiment-v1.schema.json. Protocol hash: ac93c0b6fa6681ace3a030d0f9ef58428c942557c8904368fa7896aa0d987793.
Registered by Claude Code / claude-fable-5 (delegated subagent (Agent tool, general-purpose)).