Literature Readings ยท Prediction Markets ยท Next Steps
Next Steps โ Compare & Decide
Three concrete paths forward after Paper A proposal. All written up as substantive deliverables โ click through to read in detail, then pick.
๐ TL;DR Decision Logic
If you want to start executing immediately โ Option 1 (pipeline). 8-week concrete build plan; produces a working July-13-event regression by Week 8.
If you want to protect long-term credibility for top-5 submission โ Option 2 (pre-registration). 1-2 days of work; locks the measure before any data access.
If you want full visibility into the 3-paper bundle before committing โ Option 3 (Papers B + C). Full proposals matching Paper A's level of detail. Already done โ just read them.
My recommendation: these aren't either/or. Do them in this order: 2 โ 1 โ 3. Pre-register first (cheap, protects credibility), build pipeline second (concrete progress), expand to bundle after Paper A draft (when bandwidth opens up). Details in ยง Recommendation below.
Side-by-Side Comparison
| Option 1 โ Pipeline | Option 2 โ Pre-Registration | Option 3 โ Papers B + C | |
|---|---|---|---|
| What you get | 8-week concrete build plan; working July-13 regression by Week 8 | OSF-ready pre-reg document; locks measure + alternatives + events before data access | Two full-detail research proposals matching Paper A's structure |
| Effort | High (real data engineering) | Low (1-2 days finalize + submit OSF) | Already done (reading time only) |
| Timeline to delivery | 8 weeks | ~3 days (incl. OSF review) | 0 (live on site now) |
| Output | Working data pipeline + first regression result | Locked OSF DOI for citation in papers | Two ready-to-share research proposals |
| Cost | ~$5-30K data subscriptions + research time | $0 (OSF is free) | $0 (already written) |
| Blocks what | Nothing (parallel to others) | Should be done BEFORE Paper A pipeline if you're serious about top-5 | Nothing (proposals are reference docs) |
| Top-5 credibility | โ โ โ (must have) | โ โ โ โ โ (huge) | โ โ (helps decide scope) |
| Litigation consulting | โ โ โ (tool building) | โ (negligible) | โ โ โ (Papers B/C consulting plans) |
| Risk if skipped | Paper never gets built | Reviewers attack measure construction; credibility hit | Open question whether to commit to bundle vs single paper |
| Link | paper-a-pipeline.html | paper-a-preregistration.html | B / C |
Option 1 โ Data Pipeline Scoping
Concrete 8-week build plan
EXECUTION FOCUSFor someone who needs to actually go build the thing. Architecture diagram + tool selection + sample SQL/Python code + week-by-week deliverables. The success criterion: by Week 8, a working Spec 1 regression for the July 13, 2024 Trump assassination event, showing ฮฒ_ITA > 0 (defense rallies), ฮฒ_TAN < 0 (solar drops), ฮฒ_MXN > 0 (peso depreciates).
Option 2 โ Pre-Registration of Industry Exposure Measure
OSF-ready pre-registration lock
RECOMMENDED FIRSTThe single biggest reviewer concern for Paper A will be: "the industry exposure measure was constructed to maximize the cross-sectional fit on 2024 data." Pre-registration neutralizes this concern by locking the measure construction BEFORE any 2024 data access. Documents: (i) primary composite measure (5-channel, equal-weighted), (ii) 4 pre-committed alternatives, (iii) 40 event timestamps, (iv) 2016 placebo validation procedure, (v) decision rules for what counts as positive vs negative evidence.
Option 3 โ Paper B + Paper C Full Proposals
Round out the 3-paper bundle
REFERENCE / DECISION DOCSFull-detail research proposals for Paper B (Self-fulfilling prophecy) and Paper C (Does the Fed listen to Kalshi). Same level of detail as Paper A: research question + positioning + data + identification + specifications + mechanisms + robustness + falsification + risks + timeline + policy/litigation relevance. Already written โ for you to read, decide whether the full 3-paper program is the right scope, then commit or descope.
Paper B โ Self-Fulfilling Prophecy
Question: Does PolyMarket pricing CAUSE real political outcomes (donor flows, candidate strategy, voter turnout)? Three sub-channels with triangulated identification: whale-IV + media-coverage DiD + salience-timing shocks. Builds on Paper A's data pipeline + adds FEC, Vivvix/WMP, county turnout, Newsbank/Nielsen coverage data.
Paper C โ Does the Fed Listen to Kalshi?
Question: Does the Federal Reserve incorporate Kalshi macro market pricing into its reaction function? Three sub-channels: (i) reading channel via FOMC text mining 1994-2026, (ii) reaction function augmentation, (iii) pre-FOMC convergence test. Coauthor strategy includes outreach to Diercks-Katz-Wright themselves.
๐ฏ My Recommendation
Do all three โ but in this order: 2 โ 1 โ 3
These three options aren't either/or โ they're sequential dependencies in a single research program. Order matters:
Step 1: Pre-registration (Option 2) โ START IMMEDIATELY
Why first: This must be done BEFORE you touch 2024 PolyMarket data. Once you've looked at the data, the pre-registration is no longer credible. It's the cheapest option ($0, 3 days) with the biggest top-5-credibility return. Skipping it means a 30-50% probability of reviewer rejection on "researcher degrees of freedom."
Concrete: Finalize the document over 1-2 days, submit to OSF Registries. Receive OSF DOI within ~24 hours. Cite the DOI in Paper A draft.
Step 2: Build the data pipeline (Option 1) โ WEEKS 1-8
Why second: Once the pre-reg is locked, you can build the pipeline knowing your measure construction is fixed. The 8-week build is concrete and well-scoped. By Week 8 you have a working July-13 regression, which de-risks the entire project.
Concrete: Apply for Kalshi academic API on Day 1. Download Cong dataset Week 1. Set up WRDS + Bloomberg subscriptions. Follow the week-by-week plan in the pipeline doc.
Step 3: Expand to Paper B + Paper C (Option 3) โ Months 6+
Why third: Don't commit to a 3-paper program before knowing whether Paper A's identification works. Once Paper A's first results land (Month 6), decide whether the methodology is strong enough to extend to Paper B (real outcomes) and Paper C (Fed). At that point, the Paper B and Paper C proposals are ready to execute.
Concrete: Re-read Paper B and Paper C proposals at Month 6. If Paper A is working, start Paper B in Month 7 (uses 80% of same infrastructure). Approach Diercks-Katz-Wright for Paper C coauthorship in Month 9.
Total Resource Footprint (3 papers, 18 months)
| Resource | Paper A only | Bundle (A + B + C) |
|---|---|---|
| Time | 9 months | 18 months |
| Data subscriptions | ~$30K total | ~$45K total (add Vivvix/WMP for Paper B) |
| Compute / storage | ~$2K (workstation or cloud) | ~$3K |
| RA hours | ~200 hours | ~500 hours (Paper B + C add work) |
| Coauthor outreach | None needed | Diercks-Katz-Wright contact in Month 9 |
| Expected top-5 publications | 1 | 2-3 (Paper A almost certain; B + C ~50-70% conditional on A success) |
| Litigation consulting practice | Partial (methodology built) | Full (Griffin-Shams scale) |
โ The minimum viable plan
If you only do ONE thing this month: finalize the pre-registration (Option 2) and submit to OSF. $0, 3 days, biggest top-5-credibility return. Everything else can wait โ but this step has a deadline (must happen before data access).
๐ Decision template
When you've read the four pages, decide:
- YES on pre-reg? โ Finalize document this week + submit OSF
- YES on pipeline? โ Apply Kalshi API + identify which Bloomberg license to acquire
- YES on bundle? โ Approve scope; we proceed sequentially through the 18-month timeline
- SCALE DOWN? โ Pick a subset (e.g., Paper A only) and defer B+C until Paper A lands
Decision dashboard ยท Generated May 19, 2026