Survivor is the simplest pool format to explain and the hardest one to finish. Pick one team each week to win its game. Win and you move on. Lose and you're out. The catch that turns it into a season-long puzzle: each team can only be used once all season. This post covers the decisions that actually end runs — the opening week, the crowd, your strongest teams, and the bye stretch — with the reasoning laid out so you can disagree with it intelligently. It's an approach, not a system. Nobody has a system.
The rules that decide your season
Worth stating precisely, because every one of these ends somebody's run each year:
- Straight up, never against the spread. Survivor asks one question: did your team win. A three-point win counts exactly as much as a thirty-point win.
- A tie is not a win. By default a tie eliminates you, same as a loss. Commissioners can set ties to survive instead — check your pool's rules before Week 1, not after a tied game. The NFL produces a tie roughly once every season or two, so most pools never think about this until the Sunday it happens.
- One team, once, all 18 weeks. There are 32 teams, so you can't run out — but you can absolutely run out of teams you want.
- A missed pick is an elimination. Picks lock game by game: a team playing Thursday night locks at Thursday kickoff, and everything still open locks together when the Sunday afternoon slate starts. If the slate starts and you have no pick in, you're out. We send reminders; set your own anyway.
- A postponed game voids the pick. If your team's game gets postponed or cancelled, the pick neither wins nor loses — it's voided and the team goes back on your available list. The schedule failing you is not the same as losing.
Week 1 ends more runs than any other week
This isn't a statistic; it's arithmetic. Week 1 is the only week when every entry in every pool is still alive, so a single upset can never take out more people than it does in Week 1. The same upset in December only hits whoever is left standing.
Week 1 is also the week you know the least. Nobody has seen a real game. Every judgment you're making is built on last season, preseason snaps, and off-season headlines — and rosters, coordinators, and offensive lines have all changed since the games those opinions came from. The opener kicks off Wednesday, September 9 at 8:20 PM ET, and until then, everyone is guessing.
So treat Week 1 as a week to avoid losing rather than a week to win cleverly. You get no bonus for surviving with a bold pick. A sensible approach: take a solid favorite at home whose opponent you genuinely think is bad, and skip the team everyone is talking themselves into off a big off-season.
The public-percentage trap
In every survivor pool, picks cluster. Each week has one or two obvious teams, and most of the pool takes them. That creates the central strategic question of the format: do you ride with the crowd or fade it?
Here's the honest version of both sides.
Riding with the crowd means you can never fall behind your pool alone. If the popular pick loses, a big slice of the pool goes out with you — and if everyone still alive loses in the same week, our endgame rule crowns them co-champions rather than eliminating them all. Losing with the crowd is genuinely safer than losing alone.
Fading the crowd is how you win outright. If the popular team loses and you're on something else, the pool thins dramatically and you're suddenly one of a few entries holding all the leverage. The cost is obvious: most of the time the popular team wins, and you've spent a week of risk differentiating for nothing.
The size of your pool should shape this. In a small pool, a few surviving entries can see the end of the season, so surviving each week is nearly the whole game — ride the crowd. In a big pool, you eventually have to separate from the pack to win, and the cheapest separation is early, when a contrarian pick that fails still leaves you tied with everyone else who lost that week.
Save strong teams — but not forever
The one-team-once rule gives every pick a price. Burn a top team in September on a game a dozen teams could have won for you, and you won't have it in December when the slate is thin and your list is short.
That's the case for saving. Now the case against overdoing it: a saved team only has value if you're alive to use it. Surviving this week at slightly worse odds to protect a team for a hypothetical future week is often a bad trade, because the future week only exists if you get there. Plenty of survivor runs die in October protecting teams for a January that never came.
A workable middle: sketch the whole season now. Write down, week by week, which team you'd ideally use — not as a commitment, but to see the shape of the problem. You'll notice certain weeks look thin and certain teams are the only good answer to a specific week. Protect those teams. Spend everything else freely on the best available game.
Plan around the byes
Byes hit 9 of the 18 weeks in 2026 — every week from Week 5 through Week 14 except one — and at the heaviest point, Week 11, 6 teams sit out at once. A team on bye isn't pickable, so mid-season your real choice set is 32 minus your used teams minus that week's byes. That's when a September plan pays off: the sketch you made above should already show which of your remaining teams are actually available in the thin weeks.
The endgame
Survivor pools end one of two ways: one entry outlasts everybody, or the last survivors all fall in the same week and share the crown as co-champions. Both are legitimate finishes — agree with your pool up front how the pot splits for co-champions, because it happens more often than people expect. Deep in the season, think about the field as much as the football: when only a handful of entries remain, picking the same team as the others can't win you the pool, and picking differently is the whole game.
Survive the first month, respect the byes, and make your bold moves on your own schedule — not the crowd's.