Best Ball Strategy: Roster Construction for the Format With No Waivers

The Whole Season Happens at the Draft

Best ball strips fantasy football to its draft. You select a roster — typically 18 players on Underdog-style formats — and then never touch it again: no waivers, no trades, no start/sit. Each week, the platform automatically counts your highest-scoring players at each position, per RotoWire's best ball guide. That one rule change inverts everything. In managed leagues, the draft builds a floor and the waiver wire builds your ceiling. In best ball there is no wire — every injury replacement, bye-week filler, and lottery ticket has to be on the roster before Week 1. Draft-day mistakes aren't recoverable; they just quietly score zeros for four months. The format's popularity exploded for exactly this reason — it distills fantasy to the part most players love best, runs all offseason, and never punishes you for going on vacation in October. But the missing safety net changes what a good draft looks like, and managers who import their managed-league habits usually build rosters that leak points all season.

Roster Construction Is the Strategy

Because there are no in-season moves, best ball theory obsesses over one question: how many players at each position? The consensus bands are remarkably stable. FantasyPros' roster construction guide puts the standard 18-round build at 2–3 quarterbacks, 4–6 running backs, 6–8 wide receivers, and 2–3 tight ends, and The Fantasy Footballers' Underdog construction piece lands in the same ranges. The logic is positional math. Receivers get the most slots because lineups demand three plus a flex, and weekly receiver scoring is volatile enough that you want extra lottery tickets spinning. Running backs break — Fantasy Life's construction guide notes recent tournament winners have leaned on six-back builds for exactly that reason. And the onesie positions get two to three apiece: enough to cover byes and a bust, never so many that they crowd out the positions that win weeks.

Draft for Spike Weeks, Not Floors

The auto-lineup rule quietly changes what a player is worth. In managed leagues, consistency has real value because you have to choose who starts; a boom-bust receiver who alternates 3-point and 25-point games is a weekly headache. In best ball he's a dream — the platform only counts his booms, and his busts land on your bench automatically. Underdog's own drafting guide builds its whole framework around this: you're drafting distributions of weekly outcomes, and the right tail is what cashes. The same logic powers stacking — pairing a quarterback with his own receivers, the format's signature move. When that offense erupts, your quarterback's big week and your receiver's big week arrive together, concentrating the spike instead of diluting it. Correlation costs nothing to draft and compounds exactly when it matters.

What Transfers Back to Managed Leagues

Even if you never enter a tournament, best ball is the best draft laboratory in fantasy. It runs all spring and summer, it's cheap or free, and it forces reps on the exact skills managed drafts reward: reading ADP as a price, drafting tiers instead of names, and building rosters whose bye weeks and injury risk don't stack on top of each other. Best ball's one warning label runs the other way — don't import its roster shapes wholesale into a managed league, where the waiver wire exists precisely so you don't have to draft your December roster in August. The formats reward different builds; the drafting muscles are the same.

• •Anchor to the consensus bands: 2–3 QB, 4–6 RB, 6–8 WR, 2–3 TE across 18 rounds
• •Never leave a draft with fewer than two quarterbacks or tight ends — a zero at a onesie position is unrecoverable
• •Prefer volatile scorers over steady ones at receiver; the format only counts the booms
• •Stack at least one quarterback with one of his pass-catchers
• •Check your byes at the onesie positions — two tight ends sharing a bye is a guaranteed zero

Correlation: Why Stacking Works Here and Not Elsewhere

In a managed league, pairing a quarterback with his own receiver is a mild preference. In best ball it is a structural advantage, and the reason is the format's scoring mechanism rather than anything about the players. Because your lineup is chosen after the fact from your best performers, what you want is not a roster that scores steadily but a roster whose good weeks arrive together. A quarterback and his top receiver have the same good afternoon by construction: the throws that make one of them useful are the catches that make the other one useful.

That shared upside is worth more than the smoothness you give up. Two players who boom in the same week put two big scores into a lineup that only counts your best, whereas two players who alternate good weeks leave one of them on the bench every time. The same logic extends to the opposing side of a game: in a matchup that turns into a shootout, players on both teams benefit, which is why some best-ball drafters deliberately take a receiver from each side of a game they expect to be high-scoring.

The limit is concentration. A stack that fails takes several roster spots down with it, so the technique argues for stacking a couple of games rather than building the entire roster around one. Correlation is a way to shape your upside, not a substitute for having any.

Field Size Changes the Right Answer

Best-ball advice often reads as though there is one correct strategy, and there is not — it depends on how many teams you have to beat and how the prizes are shaped. In a small league where a top-three finish pays, being reliably good is enough, and a roster of dependable producers with a solid weekly floor will finish in the money more often than a boom-or-bust one.

In a large tournament where the prize is concentrated at the very top, reliably good is a losing strategy, because reliably good finishes in the top quarter and the top quarter pays nothing. There the aim is to build a roster that can plausibly produce an extreme outcome, which means accepting a much higher chance of finishing nowhere in exchange for a real chance of finishing first. Concretely: more stacking, more players whose range of outcomes is wide, and much less concern about the weeks where the whole thing goes quiet.

The mistake is importing one setting's habits into the other. A tournament roster in a small-field league loses to steadiness; a steady roster in a large-field tournament is a well-constructed way to finish two hundredth. Decide which game you are playing before you decide who to draft.