The model’s straight-up win probability for every Week 8 game, ranked safest to riskiest, plus which strong teams to save for later. Built for survivor pools and Circa Survivor. Our projection, not a bet.

Ranked by the model’s win probability for the favorite. Tap a game for the full breakdown. Remember: the safest pick and the rightpick aren’t always the same once you factor in saving teams for later.
| # | Pick | Opponent | Win % | |
|---|---|---|---|---|
| 1 | vs CAR | 71% | STRONG | |
| 2 | vs TEN | 66% | STRONG | |
| 3 | vs LAC | 62% | SOLID | |
| 4 | vs MIN | 61% | SOLID | |
| 5 | vs BAL | 60% | SOLID | |
| 6 | vs CLE | 59% | SOLID | |
| 7 | at MIA | 59% | SOLID | |
| 8 | vs CHI | 58% | SOLID | |
| 9 | vs KC | 58% | SOLID | |
| 10 | vs ATL | 57% | LEAN | |
| 11 | vs ARI | 56% | LEAN | |
| 12 | vs IND | 55% | LEAN | |
| 13 | at WAS | 55% | LEAN | |
| 14 | vs LV | 54% | LEAN |
These are the strongest teams by full-season projection. They’re tempting picks every week, but the survivor game is won by saving them for a spot where your other options are thin. Here’s each one’s best upcoming matchup, so you know when to spend them.
Every pick above is ranked by the model’s straight-up win probability, computed from six seasons of play-by-play efficiency. That’s the right starting point: in a survivor pool you only need your team to win, so the highest win probability you can get is usually the safest path through the week. But the biggest favorite isn’t automatically your best pick, because you can only use each team once.
The most common way to lose a survivor pool is running out of good options in December. If a top team is a big favorite this week but you have three other safe picks, take one of the others and save the stud for a week when your board is thin. The “save for” notes above point you at each strong team’s best remaining spot so you can plan the season, not just the week.
Survivor is single-elimination, so as entries drop, how many other people are on your team starts to matter. If everyone picks the same chalk favorite and it loses, the whole field resets together and your pick did nothing for you. Picking a slightly less popular team that still wins gains you ground on the field. Early on, take the safe favorite; deep in the money, weigh being contrarian.
A 78% favorite still loses about one time in five. Over a full season, upsets are not a question of if but when, which is why team management matters more than squeezing an extra point of win probability out of any single week. Use the model to stay on the right side of the odds, and plan so a normal upset doesn’t catch you with no strong teams left.
The safest play and the teams to save, in your inbox before the slate locks. Free, every week of the season.
Also bet the games? The Edgeis the small set of games our model flags as actual bets each week, with sizing and the model’s read on the price.
SEE THE EDGE →Projections are the model’s own, derived from team efficiency, not betting lines. Survivor is single-elimination. Past results. Not financial advice.