Best Ball vs Redraft: How the Format Changes What You Should Draft

Best ball scores your lineup after the games are played. The engine reads your roster, finds the highest-scoring legal combination for that week, and counts it. No lineup lock, no Sunday morning guess, no points stranded on your bench.

Redraft asks for that same lineup in advance, from a manager holding a projection and a weather report. That single difference is the whole comparison. Every downstream change in how you draft traces back to whether the lineup gets set before the information arrives or after it.

The Decision The Format Deletes

Our board publishes a median and a band for every player, and the methodology behind those projection bands covers how the P10 and P90 edges get built. Redraft managers spend most of their attention on the median, which is correct for a format where you commit to a player and live with the week he gives you. Best-ball managers should be spending a large share of theirs on the distance between the edges.

Band width is the price of admission in one format and the product in the other. A player whose weekly outcomes run from useless to the best line at his position forces a redraft manager into a coin flip every Sunday. That same player hands a best-ball roster a set of ceiling weeks and quietly deletes the floor weeks, provided somebody else on the roster covers the slot.

Why Volatility Gets Cheaper To Own

A high-variance player costs a redraft manager twice. The bad weeks land in the starting lineup, and the good weeks frequently land on the bench, because the manager benched him after the last bad one. Best ball waives the second cost entirely and softens the first: a low week is counted only when nobody else on the roster beat it.

The pricing consequence follows directly. Two players sharing a median are close to interchangeable on a redraft roster where you set the lineup blind. In best ball the wider band is worth more, because the format keeps the right tail and discards the left.

Contest structure amplifies the same effect. Most large best-ball fields pay through an advance structure, where finishing eighth of twelve and finishing twelfth settle in the same place, so the roster you want is the one with a live path to first. A build engineered to land in the middle is a build engineered for nothing.

Where the room has already priced that premium, the edge is gone. Our read of current drafts — a market judgment with no backtest behind it, the same standing as the five-point rule below — is that the top of the board taxes upside in full, while middle-round bands stay wide against prices that have thinned. Verify both halves of that read against the room you actually draft in before spending on either. The full case for and against high-variance players gets its own page.

A Small Piece Of Board, Worked Through

Take an invented stretch of receiver board. The numbers below are made up to show the shape, and none of them are model output. Four receivers, each projected at a season median of 11.0 points per game, with P10-to-P90 bands running 9 to 13, 8 to 15, 6 to 18, and 4 to 22.

Redraft treats those four as near-equals, and it is roughly right to. You start one at the same slot every week and collect something close to 11 per game whichever you picked, with the wide one delivering that 11 as a lumpier ride and a worse night of sleep.

The best-ball version pairs them. Roster the narrow 9-to-13 receiver alongside the wide 4-to-22 receiver, and let one flex slot take whichever of them scores more. Say the wide one clears 11 in five weeks of a fourteen-week regular season, averaging 19 across those five, and averages 6.6 in the other nine, which is what a season-long mean near 11 requires after the big weeks are set aside.

Compute the slot. Five weeks at 19 from the wide receiver plus nine weeks at 11 from the narrow one is 194 points over 14 weeks, or 13.9 per game from a lineup slot a redraft manager fills at 11.0. That 2.9 per week is the format paying you for volatility you never had to time correctly.

The catch sits in the same arithmetic. Two roster spots produced one lineup slot, and the narrow receiver spent nine of those weeks contributing a floor and nothing else. Pairs built like that are most of what a best-ball roster is, which is the real reason the roster is bigger.

Depth, Byes, And The Absent Wire

Best-ball rosters run several spots deeper than redraft rosters, and the common formats have no waiver wire at all. Every bye, every hamstring, every backfield that turns into a committee gets covered by somebody you drafted in August or it goes uncovered. A redraft manager solves the same problems in-season for the price of a claim.

That reshapes the last four rounds far more than the first four. Redraft's late picks are disposable, cut in week two for whatever live role the wire turns up; best ball's late picks are structural, because those players will be scored in some week whether the week goes well or badly. Drafting a fourth tight end you would never voluntarily start is sound construction in one format and a burned pick in the other.

Correlation is worth more here than the sum of two medians suggests. A quarterback and one of his own pass-catchers spike in the same weeks, and a format that harvests weekly maximums banks both spikes in a single lineup, while their quiet weeks fall out of scoring together. The same pairing in redraft doubles your exposure to one bad afternoon you already committed to.

Positional timing moves with that logic. The argument for waiting on quarterback and tight end holds in both formats, and best ball sharpens it, since two late quarterbacks with opposing byes produce a better weekly maximum than one mid-round quarterback produces alone. That build is argued in full on its own page; take it as given here.

Scoring settings work underneath all of it. Full PPR puts a reception floor under bad games for pass-catchers, which narrows their bands and changes which players even qualify as volatile. The scoring-format math is settled elsewhere, and this piece assumes you have already applied it to your league.

How To Draft Each One

Redraft rewards role certainty. You are buying a player you will have to start or sit on incomplete information week after week, so a narrow band and a locked usage share are worth real draft capital, and the weekly start/sit tool exists because that decision stays hard all season.

Best ball rewards band width at an equal median, plus depth that covers its own byes. A working rule: when two players sit within about a point of each other on median and one carries a P90 five or more points higher, the wide one is the best-ball pick and the narrow one is the redraft pick. That five-point line is a judgment call; no backtest set it, and it should move with how much of the board you have left.

One limit belongs on the record before you act on any of this. Band edges are harder to grade than medians, because far fewer observed weeks land in the tails, which makes P90 the softer of the two numbers we publish. The accuracy ledger holds the hits and the misses in the same table, including the calls that missed badly.

This piece is one slice of a draft plan, and the plan itself lives on the draft strategy hub. The model reruns on demand, so the boards will have moved under this prose by the time you sit down with them. Read the record, then decide how hard to lean on any of it, including this piece.