The Usage-Based Fantasy Process: A Weekly Routine

A usable fantasy football weekly process is five short sessions, not one long one, and each session answers a different question. Tuesday asks what changed. Wednesday asks who to add. Thursday and Friday ask who is healthy. Saturday asks who to start. Sunday morning asks what broke in the last hour.

Everything below is built on one idea: usage is a coaching decision and production is a result. Coaching decisions repeat week to week. Results do not. So the input you should be reading every Tuesday is snaps, routes, and touches — not the fantasy points those things happened to produce.

Here is the routine, in order, with what to actually do.

The weekly checklist

Print it, pin it, run it in the same order every week. The order matters more than the total time. Adding a player on Wednesday because his snaps jumped is a different decision from adding him on Saturday because you panicked about a lineup hole.

Tuesday: review usage, not results

Tuesday is when the previous week’s participation data is out and the takes are not. Open snap counts before you open a single recap.

For every player you roster and every player on your watch list, pull three numbers:

1. Snap share — his snaps divided by his team’s offensive plays.
2. Route participation — routes run divided by team dropbacks.
3. Touch or target share — carries plus targets as a share of the team’s opportunities. For pass-catchers this is target share; if any of these terms are unfamiliar, the NFL usage stats glossary defines each in a line.

Then compute the usage delta: this week’s figure minus last week’s, in percentage points. Flag anything that moved 10 points or more in either direction. That threshold is arbitrary but it is a good filter — smaller moves are usually rotation noise, and larger ones almost always have a cause you can name.

Ask why before you ask what next

A flagged delta is a question, not an answer. Work through the causes in this order:

• Injury. Somebody above or below him on the depth chart got hurt or came back. Durable while the injury lasts.
• Blowout. The game was decided by the third quarter. Not durable, not a signal.
• Personnel grouping. The offense played more three-wide or more two-tight-end sets than usual because of the matchup. Semi-durable, matchup-dependent.
• Earned promotion or demotion. No external cause, share moved anyway. The most valuable finding on the list and the rarest.

Game script is the reason most Tuesday flags turn out to be nothing. A team that trails by three scores throws 48 times and inflates every pass-catcher’s snap and route numbers; a team that leads by three scores does the opposite. This is the single biggest source of false signals in snap share data, which is why you read percentages and raw snap counts side by side. Ninety percent of a 50-play game is fewer snaps than 60% of a 75-play game.

Update the watch list

Tuesday’s real output is not a roster move. It is an updated list of maybe 15 players you are tracking who you do not roster: backups whose share is creeping up, receivers whose route participation jumped without targets following, rookies climbing 5 points a week. Most will never matter. The three that do will matter enormously, and you will already know their names on the Wednesday they become available.

Wednesday: waivers on usage, not points

Wednesday is bid day, and the discipline is simple to state and hard to follow: bid on role changes, not on box scores.

Your leaguemates are shopping the scoring leaderboard. That is where the competition is, and that is where the prices are inflated. The player who scored 19 points on four touches will cost real budget. The player who scored 4 points but just played 71% of snaps for the first time will not.

The full signal hierarchy — and the reasoning behind spending aggressively on some adds and not others — is laid out in the guide to waiver wire usage signals. The short version for Wednesday morning:

• Snap-share jumps outrank everything.
• Route-participation jumps outrank target-share jumps, because routes are the opportunity and targets are partly the quarterback’s choice.
• Red-zone usage appearing where it did not exist is the highest-leverage small signal in fantasy.
• Points scored is the last thing to look at, and mostly it tells you what the add will cost.

Do your bidding before the Wednesday afternoon news cycle if your league’s waivers allow it. Practice reports start moving prices on Wednesday evening.

Thursday and Friday: read practice reports correctly

Practice participation is reported in three buckets — did not participate, limited participation, full participation — across the practice days leading into a game, with an official game-status designation attached at the end of the week. That is the mechanism. What it means in any specific case is genuinely uncertain, and anyone who tells you otherwise is guessing.

What is reliable is the shape of the week rather than any single day:

Two honest caveats. First, teams differ enormously in how they use these buckets — some list nearly every veteran as limited on Wednesday as routine rest, others do not. Second, on short weeks the reports are estimated rather than observed, which makes them weaker evidence. Learn the tendencies of the specific teams your players are on rather than applying a universal rule.

The practical Thursday–Friday job is not predicting who plays. It is identifying which of your starters has a real chance of missing so that you can look at his backup’s usage profile on Friday instead of at 11:40 a.m. Sunday.

Saturday: set lineups on role certainty vs upside

By Saturday the injury designations are out and the thinking is calm. Frame every start-sit decision on two axes.

Floor is stable usage. A player with a locked role — high snap share, consistent route participation, a target share that has not moved in a month — has a predictable range of outcomes. He rarely wins you a week and rarely loses you one.

Ceiling is concentrated usage. A player whose value comes from carries inside the 10, or from a small number of deep targets, has a wide range. The distribution is lumpy because touchdowns are lumpy. That concentration is exactly what red zone usage in fantasy measures, and it is the difference between a player who scores 11 every week and one who alternates 4 and 24.

Then apply the situation:

• You are a heavy favorite. Start floor. You do not need variance; you need your projected edge to survive.
• You are a heavy underdog. Start ceiling. Two concentrated-usage players give you more paths to an upset than four stable ones.
• The matchup is close. Start the higher projected total and stop optimizing. Most close calls are genuinely close, and the confidence you feel about a coin flip is not information.

One rule that survives every situation: do not bench a high-usage player for a low-usage player because of last week’s points. That is the single most expensive habit in fantasy football, and it is what the whole Tuesday routine exists to prevent.

Sunday morning: inactives discipline

Inactive lists are released roughly 90 minutes before each kickoff. This is the only part of the week where speed genuinely beats analysis, and the way to be fast is to have done the analysis on Saturday.

Build a short contingency sheet before you go to bed Saturday night. For every questionable player in your lineup, write one line: if he is out, this is who I start instead, and here is my second option if that player’s game has already kicked off. Ten minutes of writing on Saturday buys you the ability to act in 30 seconds on Sunday.

The backup-elevation logic

When a starter is scratched, resist the phrase “the backup.” Vacated work is rarely absorbed by one player. It splits along functional lines:

• Early-down carries typically go to the most similar body type on the roster.
• Passing-down snaps go to whoever the coaching staff already trusts in protection — often a different player entirely, and one whose usual snap share is low.
• Red-zone and goal-line work can go to a third player, sometimes a short-yardage specialist who barely plays otherwise.

So know the handcuff chain in advance: for each of your key running backs, you should be able to name who takes the early downs, who takes third downs, and who takes the goal line if he misses. You establish this in August and September by watching who is on the field in each situation, not by reading a depth chart. This is also why route participation is worth tracking for backup running backs — a backup who already runs routes is a much better emergency start than one who only takes handoffs.

If you have no confident answer, the right move on a Sunday scratch is usually the safest available floor, not the most exciting name. A 9-point certainty beats a 3-or-22 lottery ticket unless you specifically need the lottery ticket.

Making the routine stick

Three habits determine whether this survives past Week 4.

Write down your reasoning, not just your move. One line per transaction: “Added him, snap share went 41% to 66% with the starter on IR.” In Week 9 you can check whether your reads were right. Nobody improves at this without a record, because memory selectively keeps the wins.

Do not skip Tuesday. Wednesday’s decisions are only as good as Tuesday’s review. When the week gets busy, cut the Thursday session, not the Tuesday one.

Separate process from outcome. You will make correct usage-based decisions that lose, because a 22% target share can still catch two passes for 9 yards. The question at the end of a season is whether the roles you bought were real, not whether every week broke your way.

The one thing worth adding as the season goes on: track how often your flagged usage deltas actually persisted into the following week. If you find that most of your flags were game script in disguise, raise your threshold from 10 points to 15 and stop bidding on two-week leans. Calibrating your own filter is the part of the process nobody does, and it is the part that compounds.