Value versus need in a live draft: when to take the best player and when to fill the hole

The false choice

Every draft guide eventually tells you to take the best player available, and every league mate eventually tells you that you have too many receivers. Both are half right, and the reason they argue is that they are using two words, value and need, without numbers attached.

Put numbers on them and the argument mostly disappears. Value is how much better a player is than the one you would otherwise end up starting. Need is how far your roster still is from the shape it has to finish in. Neither one is a tiebreaker for the other. They multiply.

What "best player" means here

Raw board rank is the wrong measure of value, and it is wrong in a specific way. The tenth receiver on the board is usually a better player than the tenth running back, but a league does not start ten receivers. What matters is how much a player is worth over the one you could have for free.

That is value over replacement. For each position, work out the rank of the last player who would be started in your league: dedicated starters times teams, plus that position's share of the flex slots, plus a small buffer for positions that are easy to stream. In a 12 team PPR league starting two backs, two receivers and one flex, the replacement level for running back sits around the 29th back and for receiver around the 30th receiver. A player's value is his projected points minus the projected points at that replacement rank.

This is why the assistant keeps taking running backs after it has two. The seventh receiver on your bench is worth roughly nothing over replacement. The fourth running back is still worth real points, because backs get hurt and cannot be streamed off the wire the way a receiver can.

What "need" means here

Need is measured against where your roster has to end, not against the starting slots. The old mistake was treating "two running backs rostered" as "running back need met." In a 12 team PPR league the finishing shape is about five backs and six receivers, so after two backs you are still short.

The assistant scores need as a multiplier on value. While you hold less than three quarters of your target at a position, value counts in full. As you approach the target it counts at 72 percent. At the target, 45 percent. Once you are past it by a fifth, 25 percent. It is a slope, not a wall, so a genuinely large value gap still wins even at a position you have filled.

The other half of need is how replaceable a position is off the wire. The weights are blunt: running back 1.0, receiver 0.6, tight end 0.45, quarterback 0.4, defense 0.12, kicker 0.08. That is the "stream your kicker, never your starting back" rule turned into arithmetic.

When value wins

Take the best player, over the position you need, in three situations.

First, when a shelf is about to end. If the last player of a tier is on the board and the next shelf is three picks below him, the value of taking him now over the first player after the shelf is the whole gap. On the September 3 board, Omarion Hampton at an average pick of 22.2 is followed by a 3.4 pick gap to Zay Flowers at 25.6. If Hampton is there at 24, he is the pick regardless of what you already have.

Second, when a player has fallen a full round past his price. A player with an ADP of 30 who is still there at 42 is a round of free value. The grader will call him a steal of 12 picks, and it is right.

Third, when the player is elite at a position that cannot be streamed. The assistant does not discount a genuine top three quarterback the way it discounts the twelfth, because you cannot find Josh Allen's weekly ceiling on the wire. The exemption is earned by separation from the streamer, not by a rank cutoff.

When need wins

Fill the hole, over the better name, when the value gap is inside one tier. The compare page says it directly for Puka Nacua and Ja'Marr Chase: both sit in tier 1, one spot and 1.1 picks of ADP apart, so the gap is inside one tier and slot need decides. That sentence is the rule. If two candidates are on the same shelf, the board has no preference between them and your roster does.

Need also wins late. In the last four rounds the assistant adds a small nudge to any running back while you are still under your back target, because a scarce position dart is where leagues are won in November and a seventh receiver never is.

The opportunity cost check

One more filter before any pick. Ask whether the player you are considering would still be there at your next turn. From slot 1 in a 12 team league your next pick is 23 picks away; from slot 6 it is 13 picks away in the odd rounds and 11 in the even ones. If the market takes a player before your next turn, waiting is not an option and taking him now costs nothing. If the market takes him after your next turn, you can take the need pick now and still get him. The player page shows the market's latest pick on him, which is the honest version of "will he be there."

What to do with this

1. Before your draft, write your finishing shape for your league's roster, not its starting lineup. In a 12 team PPR league with a flex, that is about five backs, six receivers, two quarterbacks, two tight ends, and exactly one kicker and defense.
2. On each pick, check the shelf first. If a tier ends within two picks, take the last player in it. If not, take the position you are furthest below target, choosing the highest player on the board at that position.
3. When you are torn between two names on the same shelf, stop debating them. The board thinks they are the same price. Take the one that fills the hole, and if the other one falls back to you next round, that is the value you were looking for.