The draft trade value chart needs a rewrite
The Jimmy Johnson chart is the most influential single document in the modern history of NFL roster building. It is a one-page table that assigns a numerical value to every pick in the draft — pick one is worth 3,000 points, pick thirty-two is worth 590, pick one-hundred is worth 100, and so on down through the seventh round. Two general managers exchanging picks have spent three decades opening that table, adding up both sides, and considering the trade "fair" if the totals match within a small margin. The chart has governed billions of dollars of roster decisions.
It is also wrong. It has been wrong since it was written in 1991, and the economics literature published a careful explanation of how wrong in 2005, and the chart has gone on being the basis of trades anyway for another twenty years. The story of why is the most useful case study in sports analytics of how a flawed model survives long contact with better information.
What the chart actually claims
The shape of the Johnson chart is roughly exponential. The first pick is worth about five times as much as the last pick of the first round, about fifteen times as much as the last pick of the third round, and about thirty times as much as a typical fourth-round pick. Every step toward the top of the draft is worth disproportionately more than the step before it.
The implication is that a team trading down a few spots in the first round is giving up enormous value. A move from pick five to pick ten costs roughly 800 points on the chart — equivalent to a high second-round pick. A move from pick one to pick five costs roughly 1,300 points — roughly equivalent to a top-ten pick of its own. The chart's math assumes the difference between elite picks and merely good picks is enormous, and that anyone trading down should demand an enormous return.
What the data actually shows
Cade Massey and Richard Thaler's 2005 paper — built from decades of actual draft outcomes — showed that the curve of real player value across draft slots is much flatter than the Johnson chart suggests. The top of the draft is better than the bottom, but not nearly by the multiples the chart implied. Once you adjusted for the salary cap cost of high picks under the rookie wage scale that the league later adopted in 2011, the top-of-draft premium compressed further. A first overall pick is more likely to become a star than a tenth overall pick, but it is also meaningfully more expensive, and it carries the single-largest bust risk on the board.
Updated work using career approximate value, snap counts, Pro Bowl rates, and second-contract earnings has only sharpened the conclusion. The expected on-field contribution of a draft pick declines roughly linearly across the first two rounds, not exponentially. By the time you reach the third round, the marginal value of one slot over another is essentially indistinguishable from zero. The flatness of the real curve is striking enough that a team trading down within the first round and collecting a second-round pick in exchange is almost always making a positive-expected-value trade, often by a wide margin.
Why the chart survived
Three reasons. The first is institutional inertia. Johnson's chart was published, fit on one page, and produced answers that felt right to people whose mental model of the draft was the same as the chart's. When a new GM took over a front office, the chart was sitting in the binder. Updating it required generating an alternative the building would actually use, and front offices have not historically been the most receptive environments for replacing a tool that has produced an unbroken record of defensible-looking trades.
The second reason is that the chart is self-confirming. If every team uses the same chart to evaluate trades, the prices in the trade market converge on the chart's numbers. A team that tried to act on a flatter curve would find that nobody would trade with them at those prices, because the rest of the league was still anchored to the Johnson values. Markets driven by a shared reference point can persist in mispricing for a long time, because the only way to capture the mispricing is to find a counterparty willing to transact at the correct price.
The third reason is psychological. The general manager who trades up and lands a star looks brilliant. The general manager who trades up and lands a bust looks irresponsible but recoverable. The general manager who trades down and the player they could have had becomes a star looks catastrophic. The asymmetry of the public reaction rewards aggression on draft day and punishes patience, even when patience is the correct move on the math. The chart's exponential curve gives cover to the aggressive decision. A flatter chart would not.
The teams that have noticed
A handful of teams have visibly broken from the chart over the last decade. The Patriots under Bill Belichick traded down constantly, accumulating picks in the second through fourth rounds and treating first-round slots as currency to be spent rather than positions to be defended. The Eagles under Howie Roseman have been similarly comfortable moving back and stockpiling middle-round capital, especially when their roster needed depth rather than a single high-leverage star.
The pattern in the results is consistent with what the flatter-curve analyses would predict. Teams that accumulate volume in the second and third rounds produce more starters per draft class than teams that trade up to consolidate value at the top. The marginal pick at the end of the second round is not worth much less than the marginal pick at the top of the second, and it is worth very much more than zero, which is what teams that have traded their second-round picks away end up with.
The rookie wage scale changed the math again
The 2011 CBA capped rookie contracts on a slot-based scale, which collapsed the cost differential between early and late first-round picks. Before the cap, the top of the draft carried enormous downside — Sam Bradford-type contracts that crippled a team's cap for years if the pick missed. After the cap, the top picks became more valuable on an expected-cost-adjusted basis, because the contracts attached to them became affordable even when the player turned out merely solid.
This actually moved the optimal trade-value curve a bit toward the Johnson shape, not away from it. The top of the draft is genuinely worth more under the rookie wage scale than it was under the old contract regime, because you're now buying a four-year option on a high-upside player at a controlled price. But the corrected curve still doesn't match the Johnson numbers. The top picks are valuable; they are not as exponentially valuable as the chart claims. A version of the chart that incorporates modern compensation economics would look meaningfully flatter through the first round than the version the league still uses.
What a replacement chart would look like
Public research has converged roughly on this shape: the first pick is worth about twice the last pick of the first round, not five times; the first round is worth roughly twice the second round in total, not four times; the third through fifth rounds are essentially interchangeable within a few slots; and the sixth and seventh rounds are mostly lottery tickets whose value comes from the rare hit rather than the median pick. Teams that internally use a curve closer to this shape will find themselves frequently in a position to trade down profitably against teams still using Johnson.
The chart is the rare front-office tool that has survived public, repeated, methodologically sound refutation without being meaningfully updated by the people who use it. The next decade of draft-day trades will probably look more like the Patriots' approach than the consolidate-at-the-top approach the chart implicitly recommends. The chart itself will probably go on living in the binders anyway.