This site publishes no mock drafts. It never has and it will not start; the reasons are set out on the about page. What follows is about the format itself — what it is trying to do, how hard that actually is, and how to tell the useful ones from the rest.
The draft calendar covers where mocks sit in the process year and gives the short version of how to read one. This page is the long version.
Two hard problems, multiplied
A mock draft looks like one prediction. It is two, stacked.
The first is player evaluation: how good are these people, and in what order would a competent room stack them? That problem is difficult and it is at least the same problem the clubs are trying to solve, using better information.
The second is behavioural: what will thirty-two specific organisations, each with its own board, its own scheme, its own contract situation and its own internal politics, actually do when it is their turn? That problem is harder, and it is not a football question. It is a question about institutions.
A mock has to get both right, at once, for every pick. And the errors do not add — they multiply, because each pick removes a player from the pool and changes the calculation for every club behind it. Get pick four wrong and you have not made one mistake; you have invalidated the premise of picks five through thirty-two.
What that does to accuracy
The arithmetic of a compounding forecast is unforgiving, and you can see the shape of it without needing anyone's published hit rate.
Two caveats on the model, both of which make real life harder rather than easier. Picks are not independent: an error early propagates, so the true joint probability of a long run is worse than the multiplication suggests. And trades are not in the model at all. Every draft has them in round one, and a single unforecast trade voids the ordering of everything after it — see the trade value chart for why those calls happen so quickly and so late.
Why mocks look like each other
Read five mocks in April and you will notice they broadly agree. It is tempting to read that agreement as corroboration. It is not.
The sourcing is shared. A relatively small number of people have genuine information — front office staff, agents, the reporters they talk to — and it circulates through the same channels. When six mocks move a player up in the same week, that is usually one piece of information reaching six writers, not six independent observations.
The tape is shared. Public evaluators watch overwhelmingly the same widely available games. Their errors are correlated because their inputs are identical.
Being wrong alone is expensive. A mock that matches the field and misses costs its author nothing. A mock that breaks from the field and misses looks foolish. That asymmetry produces herding, and herding produces convergence that has no evidentiary content whatsoever.
Consensus boards feed back. Aggregations of public rankings are published, mock drafters read them, and the next round of mocks is built partly on the aggregate. The crowd is reading itself.
This is why a consensus board is not a wisdom-of-crowds estimate. Wisdom of crowds requires independent errors that cancel. These errors do not cancel; they reinforce. What a consensus board measures accurately is the state of the public conversation, which is a real and useful thing to know — just not the thing it is usually presented as.
What a mock is genuinely for
Given all that, the format still does three things well, and none of them involve the specific pairings.
It teaches team needs. Working through why a club is linked to a position forces you to learn its contract situation, who left in March, what the coordinator actually runs, and where the roster is thin. A reader who follows mocks for a month knows the league's rosters better than one who does not, even if every individual projection is wrong.
It maps the shape of a position group. Where the cluster of tackles sits, where the drop-off comes, whether the class is deep at corner and shallow at safety — that structural picture emerges from the aggregate of many mocks and is far more durable than any single pick within them.
It gives you a range for a player. No mock tells you where a player will go. Twenty mocks tell you the band of clubs he is being linked to, and the band is real information about how the league is talking, even when the midpoint is wrong.
Read the range, not the pairings. That is the whole technique.
Judging the drafter rather than the mock
You cannot evaluate a forecast from its output alone, especially one that resolves once a year. Judge the method.
Do they separate what they were told from what they inferred? The best public drafters mark this explicitly. A pick sourced from a conversation with someone in a building and a pick derived from need plus board are both legitimate, and they have completely different reliability. Blending them silently is the single clearest tell.
Does the mock respond to March? Free agency rewrites needs overnight. A board that looks the same in April as it did in February is not being maintained; it is being reposted.
Do they handle trades honestly? Either mock the trades and own them, or state plainly that the projection assumes no movement. What is not defensible is a straight thirty-two-pick list from someone who knows perfectly well that four of those slots will change hands.
Do they explain the constraint? "Best player available" is not a reason. A real justification names the contract that expires, the scheme requirement, the tier boundary. If the reasoning would work equally well for any of six clubs, it is not reasoning.
Do they revisit their own record? Very few do it, and the ones who do — publishing what they got wrong and, more usefully, why the model failed — are worth more than the ones with the loudest hit-rate claims.
Do they know what a mock is? A drafter who confuses a mock with a big board will produce lists that are internally incoherent: a ranking of talent presented as a prediction of behaviour. The two documents answer different questions, as covered in how a scouting profile is built, and mixing them is the most common structural error in the format.
The final word is simple enough. Nobody gets a first round right. Everybody knows that, including the people writing them. A mock read as a prediction is entertainment; a mock read as a map of what the league currently believes is a decent piece of research — and it costs nothing to read it the second way.