Debate Quality
Debate Quality measures reasoning — a debater can lose while arguing well.
Debate ID
DB-000127
Locked DebateOpenAt RiskPreview mode — participation requires approval or invitation.
The Earth is flat.
The Earth is flat.
Context
Objective claim opposing overwhelming scientific consensus — burden and evidence are central.
Debate Completion (v1)
ConcededTheDebator rewards finishing debates, honest concessions, and intellectual accountability.
Reliability
50/100
Completion Rate
0%
- • Honest concession — positively affects reliability.
Conceded · Reliability 50/100
Debate Outcome (v1)
ConcededTheDebator is not designed to create champions. It is designed to create stronger thinkers. Victory is secondary to intellectual honesty, reasoning quality, burden fulfillment, question accountability, and steelmanning.
Debate Completion
Debate Completion measures follow-through — separate from who won.
Debate Outcome
Conceded — intellectual honesty recorded separately from reputation.
- • Explicit concession detected — outcome is Conceded, not a quality penalty.
TheDebator rewards intellectual rigor more than victory. Winning is an outcome. Reputation is a behavior. The two should never be merged.
AI Judge (preview)
ConcededHigh confidenceAI-assisted mock judgment.AI Judge preview — final judgment available after resolution.
AI Judge evaluates debate performance, not personal worth.
TheDebator should reward users for staying focused on the actual debate, not for pivoting, dodging, or changing the subject.
Topic adherence summary
@taylor_flat maintained the tightest focus on “The Earth is flat.”. Both sides remained reasonably anchored to the central disagreement. Flags: Significant Topic Drift.
Debate topic adherence score: 22/100
Significant Topic Drift
AI Judge Scorecard
Composite 33/100
Topic Adherence
22/ 100Weight 25%
Argument Quality
22/ 100Weight 18%
Evidence Integrity
51/ 100Weight 18%
Question Accountability
0/ 100Weight 14%
Burden of Proof
50/ 100Weight 14%
Steelmanning
50/ 100Weight 6%
Civility / Good Faith
80/ 100Weight 5%
Weighted: Topic 25% · Argument 18% · Evidence Integrity 18% · Questions 14.000000000000002% · Burden 14.000000000000002% · Steelman 6% · Civility 5%
Evidence Analysis (v2)
Evidence should support the specific claim under debate — classification, relevance, and source credibility are scored separately from argument style.
Debate evidence integrity: 52/100
Evidence Review
0/100No evidence artifacts were classified — claims rely on assertion until sources are submitted.
Source Review
Sources are bucketed by type (primary, academic, news, anecdotal, etc.) to separate provenance from rhetorical quality.
Credibility Review
0/100Average source credibility 0/100 — weighted by publisher tier, citation signals, community votes, and recency.
- • Publisher reputation inferred from classification tier (mock heuristic).
- • Citation quality boosted when labels include references or DOI-style markers.
- • No strong misuse patterns detected in submitted labels.
Evidence Gap Review
Gaps highlight missing premise-linked sources, weak credibility, and unsupported strong claims.
- • No classified evidence submissions linked to this debater.
Per-debater evidence
@taylor_flat
Integrity 52/100Evidence gaps
- • No classified evidence submissions linked to this debater.
Per-debater topic adherence
@taylor_flat
Topic 22/100Significant Topic Drift
No additional topic notes.
Performance analysis (not a simple win call)
- • @taylor_flat topic adherence scored 22/100 against the premise.
- • Burden partially met — outstanding support requested.
- • Evidence gaps: No classified evidence submissions linked to this debater.
- • Debate-wide composite 33/100 — topic adherence weighted at 25%.
- • TheDebator should reward users for staying focused on the actual debate, not for pivoting, dodging, or changing the subject.
- • No evidence artifacts were classified — claims rely on assertion until sources are submitted.
- • Average source credibility 0/100 — weighted by publisher tier, citation signals, community votes, and recency.
AI-assisted mock judgment. @taylor_flat maintained the tightest focus on “The Earth is flat.”. Both sides remained reasonably anchored to the central disagreement. Flags: Significant Topic Drift. @taylor_flat topic adherence scored 22/100 against the premise. Outcome label: Conceded — see narrative below, not a declaration of absolute truth.
Strongest arguments
- • @taylor_flat: Satellite imagery is fabricated by governments and space agencies.
Weakest arguments
- • @taylor_flat: Satellite imagery is fabricated by governments and space agencies.
Evidence gaps
- • No classified evidence submissions linked to this debater.
Burden of proof
Burden partially met — outstanding support requested.
AI Judge may be wrong. Users should inspect the reasoning.
31 Intellectual Outcomes12 supports · 312 counterpoints · 45 evidence
Share Debate
DB-000127
Locked DebateStalledConcededThe Earth is flat.
89 participants · Direct URL below
/debate/DB-000127
Initial Stance
Taylor F.
Satellite imagery is fabricated by governments and space agencies.
🔒 Participation locked
You can view this debate preview, save, and share — but you cannot post support, counterpoints, evidence, questions, clarifications, or takeaways until access is approved.
Locked debates create higher-control discussions without hiding the public intellectual value of the debate. The initiator controls participation, not visibility.
Main Actions
Respond with structured reasoning — Support, Counterpoint, Question, or Takeaway.
Debate Outcomes
Intellectual outcomes logged across this debate — learning, understanding, common ground, and open questions.
Total Takeaways: 31
Top Takeaways
Learned New Relevant Information: 0Gained Understanding: 0Found Common Ground: 0Changed My Mind: 3Need More Evidence: 28
Analytics Preview (Mock)
Total Takeaways
31
Learning Events
0
Understanding Events
0
Common Ground Events
0
Mind Changes
3
Evidence Requests
28
Question Accountability
Answer rate
0%
Questions asked
0
Avoided
0
🔒 Participation locked
You can view this debate preview, save, and share — but you cannot post support, counterpoints, evidence, questions, clarifications, or takeaways until access is approved.
Locked debates create higher-control discussions without hiding the public intellectual value of the debate. The initiator controls participation, not visibility.
Response Composer
Draft responses are evaluated for relevance, criteria definition, and evidence alignment. Press Ctrl/Cmd+Enter to post.
Select Support or Counterpoint to unlock structured response options.
Debate Tree
1 response — live from Firestore.
TF
Taylor F.
TD: 75
AQ: 70
Initial Stance
Satellite imagery is fabricated by governments and space agencies.
Evidence Type: No Evidence
Timestamp: 1d ago
0 Supports
0 Counterpoints
0 Evidence
0 OutcomesClaim Resolution Framework v2 — Argument Evaluation
Argument Quality and Claim Truth Status are evaluated independently.
Argument Quality ≠ Claim Truth
TheDebator rewards Reasoning quality. It does not automatic agreement with ai.
- • Good arguments can defend unpopular positions.
- • Bad arguments can defend popular positions.
- • A debater can score high on argument quality even when the claim is false.
Argument Quality (Independent)
72/100
Measures how well the debater argues — not whether the claim is true or popular.
Topic Discipline75
Reasoning Strength70
Counterpoint Engagement65
Evidence Relevance72
Steelmanning70
Structure71
Civility84
Claim Truth Status (Independent)
Subjective ClaimPending claim truth resolution.
Subjective Claim — No Winner Declared
Subjective claim — AI does not pick a winner.
Placeholder.
Standards
50
Reasoning
50
Evidence
50
Consistency
50
Burden of Proof (v1)
Burden assigned: Moderate. Burden met (81/100).
Burden Assigned: ModerateMet
Burden Met
81/100
Flags
Burden Avoided: No
Burden Shifted: No
- • Burden met for the assigned level of proof.
Steelman Framework (v1)
Steelman 70/100 · ModerateSteelman Score: 70/100 — Partial steelman; room to engage strongest view.
70/100
- • Continue separating understanding from critique.
Detailed Argument Analysis
Reasoning Strength
Moderate70/100
Claim → Reasoning → Evidence → Conclusion· Chain incomplete
- • Pending full text analysis
Placeholder reasoning evaluation.
AI Summary
Mock evaluation for Initial Stance — assesses reasoning quality, not whether the opinion is correct.
Topic Discipline
75/100
Legacy Quality Score
70/100
Topic Discipline
75/100
Reasoning Quality
70/100
Evidence Quality
68/100
Evidence Relevance
72/100
Counterpoint Engagement
65/100
Nuance & Context
68/100
Scope Accuracy
72/100
Placeholder — Argument Evaluation Framework will score this when connected.
Suggested Improvement
Define your standard, support with relevant evidence, and engage the strongest counterpoint.
Branch Reading Mode
Isolate one branch to read only that reasoning path.
Focused Path
Initial Stance → Mixed Paths