Drafting from Picks 9–12: The Turn Playbook

The Turn Runs On Two Clocks

Draft slot sets one number that governs everything downstream: how many players leave the board between your selections. Picks nine through twelve in a twelve-team room swing that number between almost none and almost two full rounds. The tenth slot selects at 10 and 15, then sits until 34. The twelfth slot selects at 12 and 13 with nobody in between, then watches twenty-two players go.

A drafter in the middle of the round waits ten or twelve picks every time, and that regularity is what most draft advice quietly assumes. The turn breaks the assumption in both directions at once. Your first wait is short enough that the board barely moves. Your second is long enough to empty whole tiers.

Same total wait, different distribution. Variance is the entire story at the turn, and every adjustment below is an adjustment to variance.

Two picks separated by four players, or by none at all, amount to one selection of a combination. The room rarely gets to break that combination up on you, which is a real edge and the only structural one the turn hands out. A middle-round drafter takes a player and then lets ten or twelve unknown picks happen before acting again.

Count The Survivors Against The Gap

The operation that decides every turn pick is a survival count. Take the tier you want, count the players sitting above the next break, and compare that count against how many of that position your gap will consume. Tier construction gets its own treatment on the tier-break page; assume the breaks here and work with the counting.

Consumption rate is the piece most drafters skip. Through the second and third rounds of a standard room, running back and receiver together absorb the large majority of picks, and a working figure of forty percent of picks going to a single position is a judgment dial you tune to the room in front of you. A room drafting three receivers for every back is telling you its rate out loud.

Take an invented stretch of receiver board — numbers made up to show the shape. Six receivers with projected medians of 214, 211, 207, 190, 187, and 184. The steps inside the top three run three or four points; the step from 207 down to 190 runs seventeen. That seventeen-point step is the tier line, and it leaves three bodies above it.

Run both gaps against those three bodies. From pick 10 to pick 15, four players come off the board; at a forty percent take rate, four picks eat roughly two receivers, so one of the three survives to 15 more often than it does not. From pick 15 to pick 34, eighteen players come off, which eats about seven receivers — more than double the tier. The three-man tier cannot survive the long gap, and the 190-187-184 tier behind it cannot either.

That arithmetic converts into a rule you can execute at draft speed. Any tier smaller than what your gap consumes gets bought inside the pair; any tier deeper than that can wait.

The short gap is where you can afford to take the best player on the board. The long gap is where scarcity gets paid for. Replacement level is the machinery underneath the whole calculation, and value-based drafting makes that case in full.

Reaches Cost Less At The Front Of The Pair

A reach is normally scored against ADP: you took a player well before the market would have. The denominator that matters at the turn is your own next selection. The real cost of a reach is the distance between the player you took and the best player at that position still available when you act again.

Across an eighteen-player gap, that distance is often smaller than the ADP delta implies, because the picks between yours are being spent by other people on other positions. Across a four-player gap, the distance runs close to zero, which makes reaching at the front of a pair nearly free. The ADP-versus-value page carries the market side of this; ADP updates daily and the gap you are reaching across never moves.

The inversion is where turn drafters actually lose points. Reaching at the back of a pair spends your one efficient pick right before a wait of up to twenty-two players, and the board does not refill during that wait. Front of the pair is for the greedy pick. Back of the pair is for the pick that has to hold up for two rounds.

Runs, And What Waiting Costs From The Turn

Positional runs are the turn's specific hazard. A run that begins two picks after your pair completes will finish its entire course during your long wait — the whole run goes, and you act on what is left of the position. The middle of the round gets two chances to interrupt a run; the turn gets one, at the front of the pair.

The turn also starts more runs than any other slot. Two picks at the same position taken back to back are the loudest signal the room receives all draft, and rooms react to signal. Doubling receiver at 12 and 13 hands the next four drafters a reason to take running backs, which is occasionally the outcome you were hoping for.

Structural builds are the organized version of that move. Taking two backs or two receivers inside a pair is how the turn executes a robust-RB or zero-RB build cleanly, since the pair strips most of the risk that the room splits your combination in half. The build pages own that argument; the turn contributes the timing.

"Wait one more round" means something different from the ninth through twelfth slots. A middle drafter who waits a round pays ten or twelve players. A turn drafter who waits a round from the back of a pair pays sixteen to twenty-two, worst at the twelfth seat, and thin positions empty inside that stretch. Quarterback and tight end feel this hardest — the late-round quarterback and tight end case argues the strategy in full, and executing it from the turn means taking the position one pick earlier than the plan reads, at the front of a pair.

One limit belongs on the page. The model projects players and it does not project your room, so nothing in a projection knows that four drafters ahead of you are about to run running backs. Every survival count above rests on a room-behavior assumption the model has never graded, and the record on the accuracy page covers projection error; it says nothing about heuristics like these.

Build The Pair Menu Before The Draft

Preparation for the turn is two lists, written before anyone picks. The first is a pair menu: the three or four two-player combinations you would accept with your opening pair, ranked. The second is a survival count for every tier you care about, run against the long gap in your room.

During the draft you do exactly one thing with those lists. Your pair arrives, you check which combinations are still intact, and you take the highest-ranked survivor. Combination-level thinking removes the most common turn mistake, which is treating the second pick of a pair as a fresh decision when it is the back half of one.

Numbers move under all of this. Current medians, bands, and gaps live on the half-PPR draft board and they change whenever the model reruns; the arithmetic here stays fixed while the inputs underneath it do not.

This page is the slot-specific slice of a draft plan, and the tier-break page supplies the breaks you spend the whole draft counting. Read the record on the accuracy page, then decide how hard to lean on any of it, including this piece.