How Waterfalls works.

Every fund on Waterfalls traces back to a PPM we've indexed. Extraction reads the offering document, pulls out key economics (pref rate, compounding rule, catch-up, carry split), every fee schedule (acquisition, asset management, disposition, property management, loan servicing), equity class definitions, and per-class waterfall tiers.

Every numeric value is recorded with a page reference and a short source quote so the provenance is auditable. When a PPM is ambiguous, we tag it as Medium confidence or worse and flag specific clauses for review.

The engine is a pure TypeScript port of the Python model we used to prototype this. Given equity, a hold period, and a gross cash multiple at exit, it walks every tier of the fund's actual waterfall — RoC → pref → catch-up → split, plus any per-class overlays — and produces the dollar flow to LP, GP, and fees.

Five fund archetypes are supported today: simple single-class, debt funds, preferred equity funds, the Alabama Class A/B structure we first indexed, and the Georgia storage multi-class structure. A sixth category (side letters) is coming as part of the Institutional tier.

The Composite score blends five independent dimensions, each on a 0–100 scale:

• →Absolute LP Take. The fraction of gross proceeds that actually lands in LP pockets at a base-case 1.75× exit. Bigger is better.
• →GP Timing Risk. How much of the GP's economics land as fees vs. back-ended promote. Fee-heavy funds lose points because the GP gets paid whether the deal works or not.
• →Pref Quality. Rate × compounding × compounding cap. An 8% compounded pref beats a 6% simple pref at the same headline rate — the scoring reflects that.
• →Fee Drag. Total fees as a share of gross proceeds in the base case. Lower is better.
• →Alignment. GP commit %, clawback provisions, and whether carry is computed on a deal-by-deal or fund-as-a-whole basis.

The default Composite weights these 30/25/15/15/15. /leaderboard lets you re-weight them live — an allocator who cares more about downside protection shouldn't be forced to use our default blend.

This is an educational analysis tool. It isn't an offer, solicitation, recommendation, or endorsement of any fund. We model the documents as written; real distributions depend on deal execution, amendments, side letters, and factors that can't be read out of a PPM.

Scores are relative, not predictive. A high composite means the documents are LP-favorable as drafted — it says nothing about whether the sponsor will hit projections, fill the raise, or navigate a downturn. Always consult qualified legal, tax, and financial advisors before investing.

Every fund page cites the PPM version and extraction date it's derived from. Corrections welcome: sponsors and service providers can claim their profile to mark data reviewed, and anyone can upload a PPM to add or correct a fund. Three validated uploads earns a free year of Analyst.