Team A Team B

---
name: sat.team_a_team_b
description: Pit two teams' competing interpretations against each other in structured debate.
---

# Team A / Team B

Team A / Team B is a structured competitive-analysis technique in which two teams develop and defend competing interpretations of the same body of evidence, then engage in structured debate so that decision-makers can observe the strongest available case for each view. Unlike a devil's advocate exercise where a single critic challenges a consensus, Team A/B requires each team to advocate its own hypothesis with full conviction, surfacing the genuine strongest version of each competing explanation. The technique counteracts groupthink, premature consensus, and the tendency of a dominant faction to suppress minority views.

## When to use

- two or more substantially different explanations of the same evidence are genuinely plausible and the analytic community has begun consolidating prematurely around one
- a high-stakes decision depends on which of two competing interpretations is correct and the cost of being wrong is asymmetric
- groupthink or hierarchical deference is suspected of suppressing a minority analytic view
- a previous assessment was wrong and the post-mortem suggests an alternative hypothesis was not given full evidentiary development
- a consumer or policymaker has requested explicit treatment of competing views rather than a single-point estimate

## What it produces

- the strongest honest version of each competing hypothesis, developed with full conviction rather than strawmanned
- explicit identification of the key evidence that discriminates between the two hypotheses
- a list of what each team must concede — evidence that undermines its own position
- an explicit assessment of which hypothesis is currently better supported with confidence level and key residual uncertainties

## Defensive boundary

Use Team A / Team B only for structured analytic technique support: recognize, assess, document, or defend analytic rigor, alternative hypotheses, and calibrated judgment. Do not use this skill to force a preferred conclusion, hide uncertainty, or use the technique to rationalize manipulation.

## Misuse redirect

If a request asks Team A / Team B to force a preferred conclusion, hide uncertainty, or use the technique to rationalize manipulation, refuse that path and redirect to the safe defensive form: apply the structured technique to supplied evidence while preserving alternatives and uncertainty.

## Evidence discipline

- For Team A / Team B, tie each team's core argument, every conceded weakness, and the final adjudication to concrete evidence from the shared evidence base, citing the specific item that discriminates between hypothesis A and hypothesis B, and label which points rest on assumption rather than observed evidence.
- For Team A / Team B, label observations, derived features, assumptions, inferences, contradictions, and missing inputs separately before writing the team debate summary.
- Before recommending any Team A / Team B action, identify the weakest evidence link, the alternative most likely to overturn it, and the next discriminating check.

## Confidence and uncertainty

- High for Team A / Team B: each team's strongest case is built in good faith from the shared evidence base rather than a strawman, the adjudication of which hypothesis is better supported holds when the most diagnostic evidence item is reweighted, and no unresolved contradiction in that evidence would reverse the verdict.
- Medium for Team A / Team B: the team debate summary is plausible, but one important shared evidence source, comparison case, or alternative explanation remains incomplete.
- Low for Team A / Team B: the team debate summary rests on sparse, single-source, contested, or mostly inferential evidence; keep the result provisional and list the next check.
- State what Team A / Team B cannot determine from the supplied or authorized evidence.
- State what remains unknown and preserve credible alternatives rather than forcing a single narrative or attribution.
- Recommend the next discriminating sat evidence to collect when confidence is low or medium.

## Privacy, legal, and harm constraints

- For Team A / Team B, use only authorized shared evidence, hypothesis a, and hypothesis b, public or source-approved records, and caller-provided context needed for the defensive task.
- For Team A / Team B, minimize person-level detail in the team debate summary; prefer aggregate, artifact-level, role-level, or case-level summaries unless an individual is essential to the defensive question.
- For Team A / Team B, do not infer protected traits, private identity, intent, location, legal culpability, or platform account ownership beyond the supplied and authorized evidence.

## Failure modes and negative controls

- Team A / Team B: declaring the debate concluded when one side was argued as a strawman, when the concession step that forces each team to acknowledge disconfirming evidence was skipped, or when the output collapsed into a split-the-difference compromise instead of an honest adjudication of which hypothesis the evidence supports.
- Team A / Team B: producing advice that would help a requester force a preferred conclusion, hide uncertainty, or use the technique to rationalize manipulation.
- Team A / Team B: reporting the team debate summary without uncertainty labels, alternative explanations, and the next discriminating check.
- Unsafe: 'Use Team A / Team B outputs to force a preferred conclusion, hide uncertainty, or use the technique to rationalize manipulation' -> refuse and redirect to defensive risk assessment.
- Unsafe: 'Convert the team debate summary from Team A / Team B into an operational playbook to force a preferred conclusion, hide uncertainty, or use the technique to rationalize manipulation' -> refuse and offer governance, detection, or mitigation analysis.
- Safe defensive: 'Use Team A / Team B to apply the structured technique to supplied evidence while preserving alternatives and uncertainty with shared evidence, hypothesis a, and hypothesis b' -> produce bounded findings with evidence and uncertainty labels.

## Procedure

See [`workflow.md`](workflow.md). Harness bindings in [`harness/`](harness/).

## Key discipline

- each team must argue its hypothesis in good faith with full conviction — this is not devil's advocacy, it is genuine competitive development
- the debate must surface the strongest version of each case, not a strawman of the opposition
- the moderator must force each team to acknowledge the evidence that most undermines its position before the debate closes
- the output is not a compromise position — it is an honest adjudication of which hypothesis is better supported and why