What is ADP in fantasy football? Average Draft Position

Most fantasy football guides treat ADP as a ranking. It is not. ADP is the market price of draft-room attention, and it tells you when a player usually disappears in your draft, not whether that player makes your roster easier to run for the 17 weeks after the draft ends.

That distinction is the whole post. The rest is how to use it.

What ADP actually means

ADP stands for Average Draft Position. It is a number that aggregates where a given player tends to get selected across many real and mock drafts on a given platform. If a quarterback is drafted 47th overall in one draft, 52nd in another, and 50th in a third, his ADP across those three drafts is 49.7.

That sounds boring, and the boring part is the important part. ADP is not a player ranking. It is a behavioral record of how the average drafter behaved in the average draft on the average platform. The player ranking lives in a separate document, often produced by the same site, and the two documents disagree more than most readers notice.

Most explainers stop here. They define the term, sketch the calculation, and move on. The harder questions, the ones that decide whether you read ADP correctly or get punished by it, almost never get answered. This post answers them.

Why ADP varies between platforms, and what that variance actually proves

Pick a player and look up their ADP on five platforms. You will not get five matching numbers. In Andrew Erickson’s July 2024 FantasyPros piece on platform ADP variance, Kyle Pitts had a Yahoo ADP of Round 7 and an ESPN ADP of Round 10. Jonathon Brooks had an ESPN ADP of Round 6 and a Yahoo ADP of Round 10. Same player, same preseason, four-round swing depending on which draft room you walked into.

Most readers see numbers like that and decide ADP is unreliable. The deeper read is the opposite: ADP is exactly as reliable as the draft room it came from. The Pitts gap is not a measurement error. It is evidence that ESPN’s drafter pool and Yahoo’s drafter pool disagree about how much tight-end risk to absorb in Round 7. The variance is the signal.

This matters for one practical reason. If you are drafting on Sleeper, the Sleeper ADP is the right input for your draft. If you are drafting on ESPN, the ESPN ADP is the right input. Pulling a composite ADP from a third-party aggregator and using it as gospel in a Yahoo draft will routinely cost you players you wanted because Yahoo drafters reach for them earlier than the composite predicts.

The player coming back after a multi-year break is the reader who gets surprised by this the hardest. ADP looks like a single objective number on a single results page. It is not. It is a market price, and every market has its own house.

ADP is not the same thing as rankings, and the gap is where the read happens

Rankings come from projection. ADP comes from behavior. The two are different inputs, and the gap between them is the most useful signal in your draft prep.

If a player is ranked 23rd and has an ADP of 38, the market is afraid of something the ranker is not. Maybe it is injury history. Maybe it is a coaching change that worries the room more than it worries the analyst. Maybe it is a contract holdout that just got resolved last week but has not yet shown up in the consensus. The reader’s job is not to pick a side. The reader’s job is to figure out what the market is pricing in and decide whether they agree.

If a player is ranked 38th and has an ADP of 23, the market is bidding the player up past where the projection says they should go. Now the reader’s job reverses. What does the room know that the ranker does not?

The discipline is the same in both directions: the gap is a question, not an answer. Reading ADP as a ranking collapses the question. Reading ADP as market data preserves it.

Why “ADP value” can be a trap

Every fantasy guide loves the word “value.” Value picks are players available later in a draft than their projection suggests they should be. The standard advice is: when you see value, take it.

That advice is not wrong, but it is incomplete. A player available 20 picks below their projected ceiling is often cheap because the market has priced in something the reader has not. Injury risk, role uncertainty, a coach who has historically benched second-half running backs, a weekly committee that splits touches in ways that destroy your lineup confidence by Week 4. The discount is real. The reason for the discount is also real.

Most abandoned teams did not collapse because the manager drafted poorly in Round 1. They collapsed because the manager drafted a Round 7 “value pick” who turned out to be a 50/50 split-time back, then spent four weeks chasing waiver replacements that did not exist, and gave up.

The frame most explainers skip: an ADP discount is a market signal that you should investigate, not a free lunch. The data-first player who treats fantasy as a stats exercise reads a 20-pick discount the same way they read a stock priced 30% below book value: the discount is real and the discount has a reason, and the work is figuring out which one is bigger. The reader who treats every ADP discount as a steal is bringing optimism to a market that has already done the math.

This is the value-trap heuristic most guides do not write down: when the discount is large enough to feel like a steal, the reason for the discount is large enough that someone has already worked it out. Find out who, and find out what they found.

ADP has an auction cousin: AAV

Most explainers treat ADP as the single metric for every draft format. It is not. ADP is positional, full stop. The auction equivalent is a separate metric with a separate name: AAV.

In a snake draft, ADP tells you when a player typically comes off the board. A 4th-round ADP means the player usually disappears between picks 37 and 48 in a 12-team league. Snake ADP is mostly an information-asymmetry tool: if your turn is pick 38 and the player has an ADP of 41, you have a real decision about whether to reach by three picks or hope they slide back to your next selection.

In an auction draft, the analogous market signal is AAV, average auction value, also called salary-cap value or simply auction value. AAV is monetary. It tells you what auction rooms typically pay for that player. An AAV of $25 in a $200-budget auction means the room usually pays roughly 12.5% of the budget to land that player. AAV is mostly a budget-allocation tool: if you have $80 left and the player has an AAV of $25, you know you can afford the nomination but you also need to plan what the other $55 buys you.

ADP and AAV are not the same metric. ADP answers a timing question, AAV answers a budget question. A reader who only studies ADP and walks into an auction draft will misprice nominations because they have brought the wrong instrument. A reader who only studies AAV and walks into a snake draft will draft late at every position because they keep waiting for the auction-style “discount” that the snake format does not produce. The positional snake-draft format and the auction format trade in different currencies, and the reader has to know which currency they are spending.

Most guides treat this distinction as a footnote. It is not a footnote. It is the difference between using the right market signal for your league and using a metric that does not match the room you are sitting in.

How to use ADP when fantasy is your fifth priority

Everything above is knowledge-gap content. This section is the operational overlay, and it is the shortest one in the post for a reason: the framework matters more than the optimization.

The manager with one or two leagues has a job, a family, and 30 minutes a week for fantasy maintenance. The reason ADP matters to that reader is not “find sleepers.” The reason is: ADP tells you which of your draft picks creates more weekly work after the draft, and which ones do not.

A Round 3 ADP wide receiver who anchors a stable target share creates roughly one start/sit decision per week and one waiver scan per bye. A Round 3 ADP running back in a committee creates roughly four start/sit decisions per week, three waiver scans per bye, and a real risk of a Week 7 injury that punches a hole in the lineup. The two players might have the same projection. They have wildly different operational profiles.

When two players grade equal by ADP and ranking, the manager juggling work and family should prefer the one who creates less in-season work. That is the execution-first heuristic from the snake-draft guide showing up again here: every draft pick buys a 17-week workload, and the reader who does not have spare hours every week should be biased toward picks that keep the workload low. ADP is the input. The operational overlay is the filter.

The reader who internalizes that filter draft-night will have a less fragile roster by Week 4. The reader who chases every “ADP value” pick will be in waiver scrambles by Week 6. Sleeper’s autopilot tools cannot save a roster that was over-allocated to fragile picks on draft night, which is why the filter matters at the draft, not after.

Common ADP mistakes, and how to avoid them

Treating composite ADP as platform ADP. Composite ADP is an average of platform ADPs. Use the platform’s own ADP for the draft you are actually in.

Treating ADP as projection. ADP is behavior. Projection is the analyst’s opinion. The two disagree by design, and the disagreement is where the work happens.

Assuming every discount is a steal. Cheap players are often cheap for a reason the market has already priced in. The discount might be larger than the reason, or it might not. Find out before you draft.

Confusing ADP with AAV in auction drafts. ADP is positional and tells you when a player typically comes off the board. AAV (average auction value, sometimes called salary-cap value) is the auction-dollar equivalent and tells you what auction rooms typically pay. They are different metrics for different formats. Use the one that matches your league.

Ignoring ADP entirely because “I trust my rankings.” Rankings tell you who you like. ADP tells you when you can get them. Both are inputs. Skipping either one is leaving information on the table.

What ADP does not tell you

ADP does not tell you whether a player will produce in October. It does not tell you whether the FAAB waiver bid you place in Week 3 will land. Sleeper Support can explain how blind bidding works on one platform; ADP still cannot tell you whether your league’s FAAB system is shallow enough that you will burn 40% of your budget by Week 5 trying to replace a busted Round 4 pick. ADP is a draft-night input. The 17 weeks after the draft run on different data.

This is the reason “ADP value” thinking gets so many readers in trouble. The “value” only exists if the rest of the season cooperates. When the season does not cooperate, the cheap pick becomes the expensive lineup hole, and the cheap pick’s discount was always priced for that risk.

Closing

ADP is one of the most-misused metrics in fantasy football. It is misused in three predictable ways: as a ranking when it is not one, as a free lunch when discounts are priced for reasons, and as a single number when it is actually one number per platform.

The post you just read fixes all three. ADP is market timing, not evaluation. ADP discounts are signals, not steals. ADP belongs to the draft room it came from, not to the composite that lives on an aggregator page.

The harder problem ADP does not solve is the 17 weeks after the draft ends. That is where the in-season execution layer actually lives, and that is the next step for any reader who wants their fantasy football to be a game they enjoy instead of a part-time job.

Request access