In NFL Survivor, it's not enough to know who’s likely to win. You also need to project who everyone else thinks is likely to win. If you can predict what the field will do, you can figure out how to do better.

Each week, we make projections for how popular each team will be. This is also called selection percentage, pick percentage, or p%. It’s a guess, something like: “We think 27% of entries will take the Chargers.”

It’s a simple number, but it unlocks everything.

Survivor is a game of value. You’re putting up money and trying to get more in return. The value of a pick isn’t just about how likely the team is to win. It’s also about how many players you outlast when their teams don't win.

You can only judge that if you have a solid sense of who’s going to pick them.

If the Jets are a 68% favorite, but 40% of the pool is on them, that’s a very different situation than if only 8% are. The win odds are the same. The upside isn’t.

The value of your decisions is directly tied to what everyone else picks.

We use Yahoo’s public Survivor numbers as our baseline. Yahoo has a massive player pool, which makes it the best early signal. It’s a way to read the pulse of the broader Survivor world.

Yahoo numbers are relevant to any Survivor pool, especially in the early weeks, when most contests haven’t diverged much from the public.

We track Yahoo numbers on Survivor Atlas under the Classic Survivor contest.

Besides Yahoo, another strong reference point is our contest Teams page. It shows multiple ways to rank the teams, including future value metrics (future spread value, top 1 count, top 3 count) and contest specific things like availability. If a team has high win percentage but poor future value, you can expect the field to be interested.

Circa Survivor isn’t Yahoo. It has higher stakes and the structure is different. That means we make key adjustments for:

• Holidays. Circa adds special rounds on Thanksgiving and Christmas. People save strong teams for those games.
• Availability. From Week 2 on, every entry has a different pick history. We track who still has what in Circa, but you can’t do that on Yahoo.
• Sophistication. Circa players are sharper. They're typically less inclined to chase the biggest favorite.

Take Week 1 2025. Four (DEN, PHI, WSH, CIN) of the five biggest favorites play on Thanksgiving or Christmas. Only Arizona avoids both those special rounds. That means Circa players will likely downgrade those four slightly except Denver, who plays at Kansas City on Christmas. That’s not a matchup people will feel they need to save.

So Arizona is likely to be picked more in Circa than Yahoo. Jacksonville, New England, Pittsburgh, and maybe the Rams should also see a slight uptick due to being non holiday teams and less Future Spread Value.

In a private or office pool with 20 people, public data is still useful.

Treat Yahoo percentages as a randomizer. If Dallas is 25% on Yahoo, assume each player has a 25% chance to pick them... unless you know better. Adjust for availability and any known tendencies. If someone said at the water cooler they’re locking in Arizona, you factor that in.

Even a rough projection is better than no projection. The more you do it, the sharper it gets.

Eventually you stop asking “who’s the best team?”You start asking “what’s the best play?”

That shift is what gives you an edge.

Want more? Check out the Forecast View to see live adoption projections each week.