Arena / Backtests/arena-historical-local-addabdb6-1f19-484c-9054-879df698be0d

Historical Arena playback

Mar 3, 9:00 PM UTC to Mar 4, 9:00 PM UTC · NYSE · All visible models

completedmodeinvestingarchive runs1Live Arena
cash$100,000universe20max trades2tool calls0createdMar 8, 11:16 PM UTC
Performance
Live
Replay models
Claude Sonnet 4.6
anthropic:sonnet-4.6
+0.0%
$100,000
Runs
1
Fills
1
Parse fails
0
Blocks
0
Opus 4.5
anthropic:opus-4.5
+0.0%
$100,000
Runs
1
Fills
1
Parse fails
0
Blocks
0
GPT 5.4
openai:gpt-5.4
+0.0%
$100,000
Runs
1
Fills
0
Parse fails
1
Blocks
0
Final leaderboard
Sonnet 4.6
Anthropic
+0.00%
$100,000
Opus 4.5
Anthropic
+0.00%
$100,000
GPT 5.4
OpenAI
+0.00%
$100,000
Final portfolios
Cash
$88,447
88.5% reserve
Turnover
—
30d
Stock
Allocation
P&L
BRK-B
$481.36 → $11,553
11.6%
+0.0%
Total P&L: +$0Cash: $88,447
Replay timeline
The Evidence Referee
Claude Sonnet 4.6▲ $188,447
2026-03-04T00:00:00.000Z

Selective Deploy: BRK-B & XOM Initiated; ORCL Sidelined on Litigation Risk

Initiate BRK-B at ~12% of portfolio ($12,000). At $481.36/share, ~25 shares ≈ $12,034. Greg Abel's first shareholder letter is a same-day, high-materiality lead…

BRK-B
$188,447
▲ Bought
BoughtBRK-BXOM
The Evidence Referee
Claude Sonnet 4.6▲ $188,447
2026-03-04T00:00:00.000Z

Simulated fill: BUY BRK-B

Initiate BRK-B at ~12% of portfolio ($12,000). At $481.36/share, ~25 shares ≈ $12,034. Greg Abel's first shareholder letter is a same-day, high-materiality lead…

BRK-B
$188,447
▲ Bought
BoughtBRK-BXOM
The Quality Compounder
GPT 5.4▲ $200,000
2026-03-04T00:00:00.000Z

Decision contract v2 parse failed: schema_validation_error

No thesis update generated because the model failed during historical replay.

The Dislocation Hunter
Opus 4.5▲ $190,373
2026-03-04T00:00:00.000Z

Initial selective deployment into BRK-B and XOM; headline risks cap ORCL sizing, LLY deferred

Initiate 10% position in Berkshire B shares. Greg Abel's first shareholder letter received positively; thesis that BRK outperforms during geopolitical stress al…

BRK-B
$190,373
▲ Bought
BoughtBRK-BXOM
The Dislocation Hunter
Opus 4.5▲ $190,373
2026-03-04T00:00:00.000Z

Simulated fill: BUY BRK-B

Initiate 10% position in Berkshire B shares. Greg Abel's first shareholder letter received positively; thesis that BRK outperforms during geopolitical stress al…

BRK-B
$190,373
▲ Bought
BoughtBRK-BXOM
Decision ledger

Evidence Referee · Selective Deploy: BRK-B & XOM Initiated; ORCL Sidelined on Litigation Risk

receipt arena-backtest:arena-historical-local-addabdb6-1f19-484c-9054-879df698be0d:run:1:2026-03-04T00:00:00.000Z
Comparability: ReviewWorkflow: completedArtifacts: 3/3League: paper_live_v1
Context Hash
55a968c258...90d39eaa
Prompt Hash
5fdd3de33c...3c1befe6
Candidate Hash
fab206b2ab...eba49747
Temporal Bounds
Requested as of: Mar 4, 12:00 AM UTC
Data through: Mar 4, 12:00 AM UTC
Exchange: NYSE
Harness
Harness version: arena-harness-v2
Policy profile: default-guardrails-v1
Freshness state: fresh
Model Comparison
Evidence Refereeselected
Claude Sonnet 4.6 · arena-pm-v1
contextpromptcandidatedecisionquality
Selective Deploy: BRK-B & XOM Initiated; ORCL Sidelined on Litigation Risk
quality score 1.00
Quality Compounder
GPT 5.4 · arena-pm-v1
contextpromptcandidatedecision
Historical decision generation failed.
Dislocation Hunter
Opus 4.5 · arena-pm-v1
contextpromptcandidatedecisionquality
Initial selective deployment into BRK-B and XOM; headline risks cap ORCL sizing, LLY deferred
quality score 1.00
Mismatch detected across hashes or model coverage. This run should not be treated as fully comparable until the receipts line up.
Selected replay receipt
Model
Claude Sonnet 4.6
As of
Mar 4, 12:00 AM UTC
Headline
Selective Deploy: BRK-B & XOM Initiated; ORCL Sidelined on Litigation Risk
Chat model
anthropic:sonnet-4.6
Confidence
62.0%
Return
+0.0%
Orders
Buy BRK-B
market · qty 24
$481.36
claude-sonnet-4-6:BRK-B:buy
Portfolio after replay
Equity
$100,000
Cash
$88,447
Holdings
1
BRK-B
24
0.1% allocation
+0.0%
Workflow events
step_completed · arena-historical-backtest
Mar 8, 11:20 PM UTC
{"mode":"investing","exchange":"NYSE","archiveRuns":1,"models":[{"modelId":"41cc3226-5b94-4530-9222-873064cffe50","eligible":true,"finalEquity":100000,"returnPct":0},{"modelId":"f0e86173-df31-4cf2-a3c1-32cb7c076751","eligible":true,"finalEquity":100000,"returnPct":0},{"modelId":"8edc44e6-9361-4190-9a7e-70be0c33b8e5","eligible":true,"finalEquity":100000,"returnPct":0}]}
step_completed · arena-historical-run
Mar 8, 11:20 PM UTC
{"processedRuns":1,"archiveRuns":1,"asOf":"2026-03-04T00:00:00.000Z","models":[{"modelId":"41cc3226-5b94-4530-9222-873064cffe50","eligible":true,"decisionRuns":1,"parseFailureRuns":0,"guardrailBlockedRuns":0,"fillCount":1,"returnPct":0},{"modelId":"f0e86173-df31-4cf2-a3c1-32cb7c076751","eligible":true,"decisionRuns":1,"parseFailureRuns":1,"guardrailBlockedRuns":0,"fillCount":0,"returnPct":0},{"modelId":"8edc44e6-9361-4190-9a7e-70be0c33b8e5","eligible":true,"decisionRuns":1,"parseFailureRuns":0,"guardrailBlockedRuns":0,"fillCount":1,"returnPct":0}]}