PEER BEHAVIOR ANALYTICS

Compare how managers changed their reported holdings.

Trophik turns a named set of public 13F filers into a structured evidence report: what changed, whether a pattern persists, and whether a model beats the last filing. Filings are delayed and incomplete, and those limits travel with every result.

Discuss a research projectSee report structure

WHAT IT IS

A research report about the peers you follow

Trophik turns a named set of public 13F filers into a structured evidence report. It describes the disclosed portfolio, tests questions stated before analysis, and measures whether historical models add anything beyond reading the previous filing.

Evidence before narrative

Filings, coverage, sample sizes, comparisons and uncertainty travel with the finding. Construction and portfolio tilts remain separate.

Bounded claims

A 13F is delayed and incomplete. The report describes the filing record; it does not claim to reveal a manager's reasoning or current portfolio.

How it works

FILINGSQUESTIONSCOMPARISONS

01

Define the peer set

Start with named public filers and the decision the research should inform. The selection and filing coverage are recorded before a finding is presented.

02

Build the filing record

Reported positions are reconciled across periods, amendments and available security data. Missing identities and characteristics are counted, not hidden.

03

Read the evidence together

The report shows descriptive profiles, declared hypothesis tests, historical model comparisons, and the limits of each. Private beta delivery includes a readout.

REPORT STRUCTURE

A report your team can review together

Begin with three agreed public filers and a recurring research decision. Each delivery pairs the evidence report with a live readout and a record of its sources, coverage, method, comparisons, and limits.

The outline below describes a proposed report. It is not a computed result.

Scope, filers, and terms are agreed per project; there is no published rate card.

Discuss a research project

FOUNDER-OPERATEDACCESS-CONTROLLED DELIVERY

PROFILE       what the selected public filings report
HYPOTHESES    questions declared and tested explicitly
TOURNAMENT    historical model vs persistence and matched nulls

WHY IT MATTERS

Public filings are easy to find. Patterns are harder to defend.

Holdings tables show what was reported. Research teams still have to decide what changed, whether a pattern persists, and whether a model improves on the last filing. Trophik makes those comparisons reviewable.

A model is compared with persistence, not only with chance.
Coverage and disclosure delay stay visible in the readout.

PROVENANCE & RE-EXECUTION

The finding carries its evidence.

Each private-beta run records its method, parameters, frozen task frame and result digests. The verifier is tested in a pinned local environment; the public record shown here remains illustrative until a cleared sample ships.

  1. 01Named, versioned methods, so there is no silent drift between runs
  2. 02Frozen task frames and full hashes for the inputs and scored rows
  3. 03Persistence and matched-null comparisons, with uncertainty and sample sizes
  4. 04Explicit current-identity and disclosure-delay limits where point-in-time support is unavailable
Inspect the illustrative run record
RUN RECORD
run_id8f21c4d90
methodpeer.construction_tournament
method_version0.1.0
data_snapshotillustrative:form13f:quarterly
snapshot_hashsha256:illustrative
inputs_hashsha256:illustrative
reproducehttps://trophik.ai/r/8f21c4d90

Illustrative record. The shape of a published run, not a computed result.

Discuss a research project

We are preparing a founder-operated peer behavior analytics pilot. Tell us which public filers you follow and what recurring decision the research would support.

We keep your email and this note to reply. Nothing else.
See the privacy policy.