Ask why a trade was fair and someone will produce a points total. The points come from a chart that assigns a value to every pick in the draft, and the chart in general circulation is the one built for the Dallas Cowboys in the early 1990s, usually credited to minority owner Mike McCoy and associated ever since with Jimmy Johnson.
It is worth being clear about what it is, because almost every argument about it comes from misunderstanding its origin.
It was descriptive, not prescriptive
The chart was not derived from how much production a pick generates. It was derived from what teams had historically been willing to trade for picks. Somebody looked at a pile of completed deals and fitted a curve to them.
That makes it a very good record of the market's collective behaviour in 1990 and a very poor model of what a pick is actually worth. If the market was systematically wrong, the chart encodes the error faithfully and then hands it back to you as an answer.
The shape of the curve
The essential feature is how violently steep it is at the top. A few landmark values give the shape:
| Pick | Chart value |
|---|---|
| 1 | 3,000 |
| 16 | 1,000 |
| 32 | 590 |
| 64 | 270 |
| 100 | 100 |
The first pick is worth three times the sixteenth, which is barely half a round later. Roughly half of round one's entire point total sits in the top ten selections. By the hundredth pick you are down to a thirtieth of pick one.
Two consequences follow directly. Moving up costs a fortune, because you are climbing a cliff. And moving down looks like a bad deal on the chart even when the extra picks are worth more in aggregate, because the chart says the picks you receive are individually worth very little.
Note also how small the drop is at the round boundary — 590 to 580 between the last pick of round one and the first of round two. The chart treats that gap as trivial. As the rookie wage scale page explains, it is not trivial at all, because only round one carries a fifth-year option. The chart predates that rule and never learned about it.
The economists' objection
The best-known challenge is Cade Massey and Richard Thaler's work on the draft, published in Management Science as "The Loser's Curse" after circulating for years as a working paper.
Their argument: clubs are far too confident in their ability to distinguish between prospects. The differences in expected performance between the first pick and, say, the tenth are real but modest, while the price differential between them is enormous. When they compared what a pick produced on the field against what it cost in salary, surplus value did not peak at the top of the draft. It peaked well down the board — into the second round.
The blunt version of the finding: trading down was systematically undervalued, and clubs that hoarded early picks were paying a premium for the feeling of certainty.
The caveat that keeps getting dropped
That research used contract data from before 2011. At the time, the first overall pick cost more than most established stars — enormous guarantees paid to a player who had never taken an NFL snap. A large part of "high picks are overpriced" was simply "high picks are absurdly expensive to sign."
The rookie wage scale removed that. The cost of a top pick collapsed, and the surplus value of the top of round one rose accordingly. The direction of the original finding survives; the magnitude at the very top does not, and anyone citing "surplus peaks in round two" without mentioning the wage scale is citing a paper they have not read carefully.
The alternatives
Several public charts try to fix the curve rather than the concept.
Production-based charts start from a measure of career value — Pro Football Reference's Approximate Value is the usual raw material — and ask what a pick at each slot has historically produced. Chase Stuart's chart is the best-known of these. The result is dramatically flatter than the Johnson chart: the top picks are worth more than everything else, but nothing like three times pick sixteen.
Market charts rebuilt on modern trades take the Johnson approach and redo it with deals from the wage-scale era, on the reasonable basis that the market has moved since 1990.
Club charts exist and are not public. It is safe to assume they are not identical to each other and that at least some of them have absorbed the research.
So how should you read a trade?
Three questions get you most of the way.
Which chart is the argument using? A flat chart and a steep chart give opposite verdicts on the same trade-down. Neither party is lying; they are using different priors about how well teams can scout.
Is a quarterback involved? Charts price picks, not positions. A club moving up for a quarterback is not paying chart value for a slot, it is paying for the option on a starting quarterback on a rookie contract, and that option is worth more than any generic pick model can express.
What was the alternative use of the picks? Chart value is only realisable if you would actually have used the picks well. A club with a full roster and no board conviction converting three mid-round picks into one good player has not lost points; it has recognised that its own drafting has a low expected return.
The chart's real function is diplomatic. It gives two front offices a shared vocabulary so a phone call takes four minutes instead of forty. That is genuinely useful, and it is not the same thing as being right.