BSD Testbench

todo: new large vs old large

new vs base

Duplicate Cancelled

Wins

Losses

Draws

Score

0.488

Elo Estimate

-8.3 ( -117.6, 99.4 )

Configuration

{
  "name": "todo: new large vs old large",
  "description": null,
  "system_a": {
    "name": "new",
    "repo_url": "https://github.com/XInTheDark/brawlstarsdraft_0827.git",
    "commit_or_ref": "website",
    "subdir": null,
    "config_path": "webapp/config.yaml",
    "model": "Large",
    "models_dir": "models",
    "download_link": null,
    "fm_device": null,
    "env_overrides": {},
    "request_timeout_seconds": null,
    "port_hint": null
  },
  "system_b": {
    "name": "base",
    "repo_url": "https://github.com/XInTheDark/brawlstarsdraft_0827.git",
    "commit_or_ref": "20251020_base",
    "subdir": null,
    "config_path": "webapp/config.yaml",
    "model": "Large",
    "models_dir": "models",
    "download_link": null,
    "fm_device": null,
    "env_overrides": {},
    "request_timeout_seconds": null,
    "port_hint": null
  },
  "judges": [
    {
      "name": "old-judge",
      "weight": 1.0,
      "epsilon": null,
      "evaluate_both_sides": true,
      "system": {
        "name": "new-policy",
        "repo_url": "https://github.com/XInTheDark/brawlstarsdraft_0827.git",
        "commit_or_ref": "20251020_base",
        "subdir": null,
        "config_path": "webapp/config.yaml",
        "model": "Large",
        "models_dir": null,
        "download_link": null,
        "fm_device": null,
        "env_overrides": {},
        "request_timeout_seconds": null,
        "port_hint": null
      }
    },
    {
      "name": "website-judge",
      "weight": 1.0,
      "epsilon": null,
      "evaluate_both_sides": true,
      "system": {
        "name": "new",
        "repo_url": "https://github.com/XInTheDark/brawlstarsdraft_0827.git",
        "commit_or_ref": "website",
        "subdir": null,
        "config_path": "webapp/config.yaml",
        "model": "Large",
        "models_dir": "models",
        "download_link": null,
        "fm_device": null,
        "env_overrides": {},
        "request_timeout_seconds": null,
        "port_hint": null
      }
    },
    {
      "name": "fast-judge",
      "weight": 0.5,
      "epsilon": null,
      "evaluate_both_sides": true,
      "system": {
        "name": "website-judge",
        "repo_url": "https://github.com/XInTheDark/brawlstarsdraft_0827.git",
        "commit_or_ref": "website",
        "subdir": null,
        "config_path": "webapp/config.yaml",
        "model": "Fast",
        "models_dir": null,
        "download_link": null,
        "fm_device": null,
        "env_overrides": {},
        "request_timeout_seconds": null,
        "port_hint": null
      }
    }
  ],
  "judge_policy": "weighted",
  "time_control": {
    "seconds": 5.0,
    "rollout": "pad",
    "policy_temperature": 0.8,
    "seed_mode": "deterministic",
    "extra_flags": null,
    "normalize_sides": true
  },
  "draft": {
    "bans_per_side": 3,
    "epsilon_draw": 0.03,
    "book_path": "books/ranked.yaml",
    "map_mode_weights": {}
  },
  "sprt": null,
  "fixed_n": {
    "games": 100
  },
  "parallelism_hint": 4,
  "max_trials": null,
  "id": "97539161-ccd3-4572-ac63-e0a3226d6879",
  "status": "cancelled",
  "created_at": "2025-10-20T16:38:01"
}

Recent Games

Export JSON Export CSV Export NDJSON

Pair	Game	Result	Map	p(A)
2028	0	B	knockout / Goldarm Gulch	0.633
2028	1	D	knockout / Goldarm Gulch	0.751
2027	0	D	brawlBall / Pinball Dreams	0.582
2027	1	A	brawlBall / Pinball Dreams	0.586
2026	0	A	hotZone / Ring of Fire	0.595
2026	1	B	hotZone / Ring of Fire	0.550
2024	0	A	brawlBall / Backyard Bowl	0.660
2024	1	B	brawlBall / Backyard Bowl	0.439
2025	0	B	knockout / Goldarm Gulch	0.635
2025	1	B	knockout / Goldarm Gulch	0.580
2023	0	B	heist / Kaboom Canyon	0.520
2023	1	A	heist / Kaboom Canyon	0.534
2020	0	B	brawlBall / Pinball Dreams	0.597
2020	1	A	brawlBall / Pinball Dreams	0.642
2021	0	A	knockout / New Horizons	0.771
2021	1	B	knockout / New Horizons	0.621
2019	0	D	knockout / Belle's Rock	0.620
2019	1	D	knockout / Belle's Rock	0.591
2018	0	B	bounty / Hideout	0.545
2018	1	A	bounty / Hideout	0.768
2017	0	B	knockout / Goldarm Gulch	0.503
2017	1	A	knockout / Goldarm Gulch	0.813
2016	0	B	knockout / Flaring Phoenix	0.341
2016	1	A	knockout / Flaring Phoenix	0.809
2015	0	A	heist / Safe Zone	0.669
2015	1	B	heist / Safe Zone	0.515
2014	0	B	hotZone / Ring of Fire	0.519
2014	1	A	hotZone / Ring of Fire	0.683
2013	0	A	bounty / No Excuses	0.580
2013	1	B	bounty / No Excuses	0.270
2012	0	A	brawlBall / Center Stage	0.724
2012	1	B	brawlBall / Center Stage	0.569
2011	0	D	bounty / No Excuses	0.458
2011	1	A	bounty / No Excuses	0.597
2010	0	A	gemGrab / Hard Rock Mine	0.645
2010	1	B	gemGrab / Hard Rock Mine	0.442
2007	0	B	brawlBall / Triple Dribble	0.459
2007	1	A	brawlBall / Triple Dribble	0.670
2005	0	B	hotZone / Open Business	0.486
2005	1	A	hotZone / Open Business	0.704
2004	0	B	brawlBall / Grass Knot	0.518
2004	1	A	brawlBall / Grass Knot	0.582

todo: new large vs old large

Configuration

Recent Games

Draft Details