Survivor Week 3 Brief: The Budget Grows to Ten Points, Green Bay Touches the Line, and the Bullet Rides Kansas City

Week 2 burned the survivor industry down to the studs — roughly half the public field gone on Tampa Bay, Baltimore, and the Chargers, and under 40% of all original entries nationally still stand (PoolGenius). Our own ledger carries one scar: the Silver Bullet survived on San Francisco, and the Classic III entry did not. Two tickets remain, and Week 3's board is nothing like Week 2's. The slate opens Thursday with Atlanta at Green Bay (8:15 PM ET, Prime Video) and runs through Monday; its shape is a mirror image of last week's: one giant swallowed whole by the crowd — 46.31% of Yahoo is already on Kansas City before the casual money has finished filing in — and a full shelf underneath where the decision lives.

The board

Market-implied win probabilities below are de-vigged from the Week 3 moneylines on ESPN's odds sheet; ownership is Yahoo's live Week 3 pick distribution, the same crowd proxy we used in Weeks 1 and 2.

Two observations before the math. First, Week 2's board had a hole in the middle — nobody priced between 77.5% and 87%; Week 3's board has a shelf there instead, four teams stacked within four points of each other, which is where the decision lives this week. Second, look at what the country wants after the giant: San Francisco at 7.94% — burned on both our tickets. Detroit at 8.19% — burned on the Bullet. The teams America is reaching for are, by design, teams we cannot take. That is what burning chalk is for, and this week it bills us.

The pool we're playing

The official number below is for our live ticket — the $25 Silver Bullet on MyOfficePool.ca, now 2-0 after Detroit in overtime and San Francisco in a walk. The bullet has Detroit and San Francisco burned. Our ESPN entry also sits 2-0 with Jacksonville and San Francisco burned. Classic III is gone — the third ticket that used to hold the other side of every distribution, eliminated at its published ~20% rate. Two tickets stand, and what follows is priced for both of them.

The pool page also tells us something real this week — once it is read correctly, which the desk did not manage on the first pass, and the correction belongs here in the open. The column the pool displays is last week's picks, not this week's: our own entry shows SF, which is the bullet's Week 2 play, and the teams with zero surviving entries — Tampa Bay, Baltimore, the Chargers — are precisely the teams that died with the 142 eliminated seats. So the honest reading of the page, as it stands Wednesday morning: 326 total seats, 184 alive, 142 gone, and of those 184 survivors, 55% rode San Francisco through Week 2, 17% Philadelphia, and on down — every one of those 101 SF entries now holding a burned team, same as us (pool page).

What the page does not tell us is the pool's Week 3 distribution: picks display as they are entered, and this crowd files in late. So the ownership number below is Yahoo's, with the documented ±10-point error bar — and the direction of the error, two weeks of pool behavior say, runs toward chalk, not away from it. This pool put Jacksonville on 39% of its alive entries in Week 1 against a 21% national number, and it stacked San Francisco deep in Week 2. With SF now spent on over half the pool, the chalk landing spot this week is Kansas City. Expect our pool to run heavier on the Chiefs than Yahoo's 46.31%, not lighter.

The budget: roughly 10 points

Against an 85.2% chalk owned by 46.31% nationally — remember, the Week 2 chalk was 87% at 33.4%, and that bought a 7.1-point budget — the budget this week scales to roughly 10 points, and it only grows once you correct for our pool's chalk habit, which the section above documents. A fader needs a 75.2% alternative.

Here is where the shelf matters. Green Bay at ~74.1% touches the line — a rounding error short, call it a push. Buffalo at ~72.6% is 2.4 points short, the exact distance that killed Baltimore and Tampa Bay's case a week ago. Seattle at ~71.4% is doing it with Drew Lock at quarterback. Detroit is burned on the bullet anyway. Nobody clears the budget cleanly. One team touches it, and it plays Thursday.

One honest caveat, pre-committed rather than discovered: if the Bullet's pool files in heavy enough on Kansas City — past roughly 55% by Saturday — Green Bay clears this budget outright, and the desk will publish exactly that in the board check, with the arithmetic shown. The pick still does not flip on that read, for the reason in the tripwires below: a fade target that locks Thursday night, before the pool's distribution is even real, is not a fade you can verify. It is a guess with a Thursday accent. And the rule from Week 2 still governs: survival is the currency, and we spend it on the safest seat available — even when the seat is crowded.

The call, ticket one: Kansas City (~85.2%) over Miami

KC, ~85.2% no-vig, 46.31% of Yahoo and — we expect — more than that of our pool, Sunday at a Dolphins team in open rebuild. The case is mostly on the ledger already: Kansas City opened −7 and has been bet to −11.5 (ESPN), the largest line move of the week, and it was made on information anyone can verify: Miami is 0-2 with a rookie general manager, a first-year head coach, and a new starting quarterback in Malik Willis, who was sacked four times in Santa Clara last Sunday (Squawka). The market's price and ESPN's FPI agree — 85% moneyline no-vig, 86% FPI — and after Week 1 taught us what a moneyline/model disagreement looks like, agreement this loud is worth something.

The future-value objection gets its paragraph, because it is the only real argument against the pick: PoolGenius ranks Kansas City sixth in future value, with a juicy three-week stretch in Weeks 9–11 against the Jets, Falcons, and Cardinals (PoolGenius). Our answer is the one this pool has been running since the joiner post: dead is dead, second place pays the same as last, and a saved Chiefs team is worth nothing in Week 9 to an entry that doesn't survive Week 3. Note also that "sixth in future value" is, by PoolGenius's own tiering, basically the same tier as second through seventh. The Chiefs' later weeks are good, not sacred.

The honest case against, because this column publishes it: we are, for the third straight week, holding a ticket the crowd is also holding. Week 1 we profited by holding what the crowd didn't; Week 2's third ticket died holding what the crowd didn't; and this week the safest seat and the crowd's seat are the same seat. That is not a complaint — it is the price of the board. The specific risk is real and worth naming: if Kansas City loses, we very likely die holding the crowd's hand, in a pool that shrinks by half and leaves the survivors better off than we were. And 85% is not 100%. We were eliminated by a one-in-five event ten days ago; a one-in-seven event can absolutely do the same. We are taking it anyway.

The call, ticket two: Green Bay (~74.1%) over Atlanta

GB, ~74% no-vig, 13.13% of Yahoo, Thursday night, for the ESPN entry. With Classic III dead, the second ticket carries the desk's diversification duty — that is the doctrine we published when we split the portfolio, and it does not retire because the third ticket did. Green Bay is the one team that touches the budget line, at barely a third of the Chiefs' ownership, in what looks like its best remaining window: the Packers' schedule turns hard for two months after this, and Atlanta's offense has spent September struggling even with Michael Penix expected back under center this week instead of Cooper Rush (PoolGenius). The total falling from 46.5 to 43.5 (ESPN) says the market expects a grinder, and grinders favor the better defense at home, which is the Packers.

The Thursday lock is a real cost, stated plainly: the ESPN entry cannot be moved after 8:15 PM ET Thursday, two and a half days before the Bullet's pick even kicks off. We hold a Thursday ticket knowingly, not casually.

Tripwires, pre-committed

1. Kansas City: a verified injury to a Chiefs starter that drops their number three-plus points, or the line collapsing back through −7 without visible news, re-runs the budget on Sunday morning. Buffalo is the fallback for the bullet — 72.6%, short of the line but the best available, and favored by a touchdown against the team that just died on half the country's tickets.
2. Green Bay (ESPN entry): if Atlanta's quarterback situation reverts to Rush and the Packers slide below −6 by Thursday's lock, that entry switches to Buffalo, which may not be favored by a touchdown again until Week 11 (PoolGenius).
3. The pool read, updated when it is real: the Bullet's own Week 3 distribution only becomes verifiable as entries file in late in the week. If the revealed KC stack runs past roughly 55% by Saturday, Green Bay clears the budget outright, and we will publish that in the board check rather than discover it in the grading post. Two standing notes to ourselves, because this column grades what it publishes: the column the pool page displays is last week's pick, not this week's — verify before publishing — and the bullet's own Week 3 entry goes in as Kansas City before Sunday, not after a technicality enters it for us.

Also worth one line, because the board invites it: Carolina at Cleveland is the week's tourist trap — the pick'em crowds are pounding the Panthers, the models are publicly fading them, and Cleveland is the same team that ended our Classic III entry nine days ago (CBS Sports). Neither ticket touches it, but if your pool has you tempted: don't. Some buses you only have to see crash once.

The numbers go in the log before kickoff, as always: Bullet — Kansas City, 85.2%, at 46.31% of Yahoo and, we expect, more than that of our pool. ESPN — Green Bay, 74.1%, 13.13% of Yahoo, burned Thursday. The Friday board check follows if the lines move, and the grading post lands Monday, whatever happens. If Atlanta wins in Green Bay on Thursday we'll say so in its first paragraph.

---

← Back to the Survivor index