The headline number is the pool. The real budget is three layers, and the edge is the gap between price and value.
College money is not one pool, it is three that stack, so the revenue-share cap badly understates what a top roster actually costs. On that budget the engine runs the same doctrine as the pro cap, price is not value, and surplus, a player's on-field worth minus his cost, is the whole game. But the college market has no public price data, so overpaying is rampant and the surplus read is intelligence the market cannot see for itself. The engine prices a player into a program at a wide, honest band, finds the cross-level value the rankings miss, and builds the roster in four buckets. All dollar figures are v0, current-as-of the 2025-26 cycle.
The pool is the small, visible one. The budget is the sum.
A composite top program. The revenue-share pool is the number everyone quotes, and it is the small, visible one. The real budget is the sum of three layers that stack far above it.
The budget is allocated by positional value (the cap is the budget), forked by conference, and the compliance surface is porous: the clearinghouse sees only a fraction of the real money and underreporting is large, which widens the band on every cost read. Model all three layers or the number is wrong, and hold the band wide.
Illustrative on the real three-layer money system (the pool, the collective layer, the NIL and clearinghouse layer, the stacked reality, positional-value allocation forked by conference, the porous compliance surface). Composite program. All dollar figures v0, current-as-of the 2025-26 cycle.
Cost is what he is paid. Value is what he wins. Live in between.
A composite roster of signings. The engine keeps two numbers apart on every one of them: what he costs, and what he is worth to winning. The space between is where it lives.
The same player is a different value to different programs, so the engine prices him into a context and reports value-to-program alongside, never merged with, the market price. The band is wider than the pro side, because the inputs are noisier and the market data is poor, and a value without a band is a false precision. Two numbers, never merged, and the surplus is the space between them.
Illustrative on the real surplus doctrine (price is not value, the surplus equation, the cross-level value find, the transparency gap, value-to-program, the wider college band). Composite players. All dollar figures v0, current-as-of the 2025-26 cycle.
Everyone can spend three of them. The fourth is where you win.
Maximizing Team KR subject to the multi-layer budget is maximizing accumulated surplus, and the build runs across four buckets. Three of them any funded program can fill. The fourth is where the roster is actually won.
The build is a rolling, eligibility-driven cycle, sequenced against the eligibility clock and the portal windows, timing the retention decisions before the portal resets the price. A program's ability to execute is capped by its institutional commitment. The plan is only as real as the institutional commitment funding it.
Illustrative on the real roster-construction framework (the four-bucket allocation, the sunk-cost and development-timing discount, the moneyball bucket as the edge, the multi-year eligibility-driven build, the institutional-commitment ceiling and its three archetypes). Composite program.
Every roster decision in college is the same one the pro game makes, a bet that a player is worth more to winning than he costs, so the engine keeps cost and value as two separate objects and lives in the surplus between them. But the college market is different in kind: the budget is not a hard cap but three porous layers that stack far above the headline pool, a signing is a sunk-cost bet with no rookie scale to control it, the player chooses the program, the deals are near-non-binding, and the portal is the market. And above all, there is no public price data, so nobody can see what the market actually is, and overpaying is rampant. That opacity is the whole opportunity. The engine prices a player into a specific program at a wide, honest band, it finds the cross-level value the rankings and the market miss, the four-star who grades like a five and the lower-level starter whose production translates, and it builds the roster in four buckets, three of which any funded program can fill and a fourth that only evaluation can. Spending is table stakes in this era, and seeing value where everyone else sees only price is the edge. Model every layer, price into the context, hold the band wide, and win the bucket that money alone cannot buy.
Model every layer, and find the value the market cannot price. That is the whole edge.
NIL and Revenue Share models the three-layer budget, prices a player into a program at a wide, honest band, surfaces the cross-level surplus the transparency gap hides, and builds the roster across four buckets, protecting the one that money alone cannot fill.