Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› restud-referee-strategy
restud-referee-strategy
GitHub针对REStud投稿的审稿策略工具。用于预判双盲审稿人及其异议，指导选择建议/反对审稿人，决定稳健性检验的前置或保留，并在正文中主动回应潜在方法论缺陷，以应对严苛审稿并争取R&R机会。
Trigger Scenarios
投稿前预判审稿人异议
填写推荐/反对审稿人名单时
决定稳健性检验的呈现方式
模型存在已知弱点需主动应对
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill restud-referee-strategy -g -y
SKILL.md
Frontmatter
{
"name": "restud-referee-strategy",
"description": "Use when anticipating the referee pool and their likely objections for a The Review of Economic Studies (REStud) manuscript, and when pre-empting those objections in the paper before submission. Maps reviewers and hardens the paper against them; does not write the rebuttal (that is restud-rebuttal)."
}
REStud Referee Strategy (restud-referee-strategy)
When to trigger
- Before submission, to pre-empt the objections a demanding referee will raise
- When choosing suggested / opposed referees in the submission form
- When deciding which robustness checks to include up front vs. hold in reserve
- When the design or model has a known soft spot you want to confront proactively
How REStud refereeing works
REStud uses double-anonymous (double-blind) refereeing — referees do not see author identities and authors do not see referee identities, so anonymize accordingly (see restud-submission). Papers are handled by editors drawn from the active frontier; the Joint Managing Editors (2026: Jan De Loecker, KU Leuven, IO/applied econometrics; Antonio Penta, UPF, micro theory / mechanism design; Jakub Steiner, Zurich/CERGE-EI, behavioral/information economics) span both theory and applied work, reflecting the journal's balanced identity — so anticipate either a theory-leaning or an applied-leaning reader depending on who handles you. REStud is also known for demanding referees and for constructively developing strong papers across rounds. Two implications:
- The first report is rarely an accept; a strong paper typically earns an R&R with substantive requests. Plan the submission so the first report's requests are ones you can satisfy, not ones that kill the paper.
- Referees expect state-of-the-art design (empirics) or complete, correct proofs (theory). The fastest path to rejection is a soft spot the referee can name in one sentence that you did not address.
Mapping the referee pool
- Identify the 3–5 most likely referees. They are the inner-ring authors from
restud-literature-positioning— the people whose work you are closest to. Assume the referee is the author of the nearest paper. - For each, predict the objection. "This author will ask why we don't use their estimator / their data / their setting." Pre-empt it in the text.
- Predict the methodological referee. Beyond the topic expert, REStud often uses a methods-focused referee. For empirics, that means the modern-estimator and inference objections; for theory, the assumption-minimality and proof-completeness objections.
Pre-empting, not hiding
- Address the single most plausible alternative explanation in the main text, not buried in an appendix.
- For the design's soft spot, include the diagnostic that confronts it (the placebo, the sensitivity bound, the alternative estimator) rather than hoping it goes unnoticed.
- Cite the likely referee's relevant work where genuinely relevant — not as flattery, but because their result is part of the positioning.
Suggested / opposed referees
- Suggested: scholars who know the subfield, have relevant publications, and are not collaborators or advisors (conflicts disqualify them). Aim for genuine experts, not friends.
- Opposed: only with a real conflict (direct competitor with a working paper on the same question, a personal conflict). Over-listing opposed referees signals defensiveness.
- Verify the current submission system's referee-suggestion fields and limits on the journal's official page.
Sequencing checks for the first round
Because REStud develops papers across rounds, structure the submission so the first report is survivable:
- Front-load the design's hardest diagnostic. Put the placebo / sensitivity bound / alternative estimator in the main text, so the referee sees you already did it.
- Hold a reserve. Keep one or two additional robustness exercises ready but not in the paper, so an R&R request can be answered with new evidence rather than a promise.
- Anticipate the "scope" request. Demanding referees often ask to extend the result (another setting, another mechanism, a generalization of the model). Decide in advance which extensions are feasible and which would turn it into a different paper.
- Map theory and empirics consistency. If the paper has both, the referee will check that the model's assumptions match the estimation sample — confirm this before they do.
Checklist
- 3–5 likely referees identified from the inner-ring literature
- Predicted objection written for each, and pre-empted in the text
- Methods-referee objections (estimator / inference / proof-completeness) addressed
- Most plausible alternative explanation confronted in the main text
- Design soft spot has a diagnostic that addresses it
- Suggested referees are real experts, not collaborators
- Opposed referees limited to genuine conflicts
Anti-patterns
- Submitting without asking "who referees this, and what will they want?"
- Hiding the design's soft spot and hoping the referee misses it (a demanding REStud referee will not)
- Suggesting friends as referees — editors discount obviously friendly suggestions
- Listing many opposed referees, which reads as defensiveness
- Treating the first report as a verdict rather than the start of a development process
Output format
【LIKELY REFEREES】[author — predicted objection — pre-empted? where]
【METHODS-REFEREE RISKS】[estimator / inference / proof gaps]
【SOFT SPOT】<the one-sentence objection> — confronted in: <section>
【SUGGESTED REFEREES】[experts, non-conflicted]
【OPPOSED REFEREES】[genuine conflicts only, or "none"]
【NEXT SKILL】restud-submission
Version History
- 1839142 Current 2026-07-05 14:21
