A Punter Kicked the Extra Point, a Kicker Hit Six Field Goals, and Half of a $25 Million Survivor Contest Died Before 4:30 p.m.
Entering Sunday, 2,908 Circa Survivor entries were alive and 55 percent of them were on two teams. The Rams lost 18-16 to Arizona as a 12.5-point favourite. The Colts led Tennessee 13-0 at the half and lost 26-20 to a team that had not won a game. By the late window, more than 1,600 entries were gone and every survivor's share of $25,017,000 had roughly doubled. The question is why 84 percent of DraftKings' money was on Arizona.
October 18, 2026 at 11:53 PM EDT
6 min read
Thirty-one percent of the surviving entries in the largest survivor contest ever run picked the Los Angeles Rams on Sunday. Another 24.2 percent picked the Indianapolis Colts. Covers' Ryan Gilbert had the count: 2,908 entries alive entering Week 6 of Circa Survivor, 901 on the Rams, 703 on the Colts. By the time Chad Ryland's sixth field goal of the afternoon went through from 43 yards with no time left in Inglewood, both teams had lost, more than 1,600 entries were dead, and the implied value of each survivor, which Covers put at $8,603 on Sunday morning, had roughly doubled before dinner.
The arithmetic is the easy part. The interesting part is that the betting market was telling the contest something on Friday, and 901 entries did not listen.
Two games that should not have been close
The Rams were 12.5-point favourites at DraftKings on Friday, per Covers' Brad Senkiw, and ESPN's AP recap calls Arizona "a double-digit underdog." Los Angeles came in leading the league in total offence. It gained 286 yards. Matthew Stafford threw for 222 and two touchdowns, the second to Colby Parkinson with 47 seconds left, and the Rams turned the ball over twice in Arizona territory in the fourth quarter. Their kicker, Blake Grupe, hurt his knee in the first quarter making a tackle on a kickoff, which is why punter Ethan Evans kicked the go-ahead extra point with 47 seconds to go, and why there was no field goal attempt available for most of the game. Arizona then went 36 yards in three snaps, helped by Jalen Brooks's catch and an unnecessary roughness penalty on Josh Wallace, and Ryland, who had already hit from 34, 30, 22, 38 and 27, hit from 43. The Cardinals are 2-4, both wins at SoFi Stadium. Mike LaFleur, who was Sean McVay's offensive coordinator for three seasons, beat him in their first meeting.
The Colts were 6.5-point favourites at DraftKings per ESPN's odds page, and -7 with a -325 moneyline in Sports Illustrated's preview. They led 13-0 at the half and had out-gained Tennessee 232 yards to 64. Then Jeffery Simmons stripped Jonathan Taylor on the first play of the second half, Chimere Dike returned a punt 58 yards to set up the tying touchdown, Oluwafemi Oladejo returned a blocked punt 14 yards for the lead with 10:17 to play, and Joey Slye kicked from 49 with 2:03 left and 44 with 43 seconds left after the Colts turned it over on downs. Daniel Jones threw two interceptions. Tennessee's defence had six sacks and three takeaways. The Titans are 1-5, Robert Saleh has his first win with the franchise, and the AP says the remaining Indianapolis fans booed the home team off the field.
What the money said on Friday
Here is the number that should have been in every survivor pool's group chat. At DraftKings on Friday afternoon, per Covers, the Rams had 51 percent of spread bets. Arizona had 84 percent of the money.
That split is what a sharp side looks like. Ticket counts follow the public; dollar counts follow the people who bet in sizes a book notices. When half the tickets and five-sixths of the cash are on opposite teams, the big money has decided the number is wrong. It does not mean Arizona was going to win outright, and it did not mean the Rams were a bad survivor pick on any single-week basis. A 12.5-point favourite wins the game most of the time. But a survivor entry is not a single-week bet. It is a share of a pool, and the value of the share depends on how many other shares are on the same team.
Nine hundred and one of 2,908 entries on the Rams meant that if the Rams won, every one of those 901 entries kept a share worth whatever the other 2,007 entries left behind. If the Rams lost, the 2,007 who were elsewhere gained 901 competitors' worth of equity in one afternoon. The sharp money was pricing the first outcome as less likely than the public thought. Nobody in the pool was obliged to agree, but the signal was free and public, and it was in a Covers story two days before kickoff.
The $100,000 room
Circa's Grandissimo, the $100,000-entry version of the same contest, started with 77 entries and a $7.7 million pool. It entered Sunday with 15. Six were on the Rams and two on the Colts, per Covers, and it left Sunday with 12. One of the three eliminated entrants says he missed the submission deadline and that his pick had been sent on Friday. We do not know which team he had picked, and Circa has not said anything publicly about the dispute.
Twelve entries into $7.7 million is $641,667 apiece if the pool were split today, which it will not be. Covers puts the implied value at $550,000, which is the figure for 14 entries; we print ours and theirs. Either way, every surviving Grandissimo entry now carries a position in the mid-six figures that can be hedged with a bet against its own pick at any Nevada counter, and the players still alive in that contest will have been doing exactly that sort of maths since the Colts' punt was blocked.
The case for picking the biggest favourite anyway
The honest counterargument is that survivor strategy is not about the week you are in. The optimal survivor pick maximises the probability of being alive at the end of the season, which means saving good teams for weeks when there are no good options, and it means that in a week with a 12.5-point favourite you take the 12.5-point favourite, because the alternative is burning a better team you will need in December. Covers' own September story noted that last season's contest ended with five perfect 20-0 entries splitting $18,718,000, which is to say that the winners survived twenty weeks by not getting cute. On that view, 901 entries on the Rams were correct on Friday and unlucky on Sunday, and the eliminated players lost to variance, not to a reasoning error.
That argument is right about the maths of a single entry and wrong about the maths of a pool. If a 12.5-point favourite wins 85 times in 100, the 901 Rams entries were all taking the same 15 percent risk at the same time, which means the pool's survivor count was not 2,908 independent bets but one large bet with 901 tickets on it. When a correlated pick fails, the pool does not lose 15 percent of its field. It loses 31 percent. The Chargers did this to 8,000 entries in Week 2. The Rams and Colts did it to 1,600 in Week 6. The contest started with 25,017 entries; after Sunday, by our arithmetic on Covers' count, fewer than 1,300 are alive with fourteen Circa weeks to go, Thanksgiving and Christmas counted separately.
What we do not know is how much of Sunday's damage was sharp survivors deliberately fading the field. Circa does not publish who picked what beyond the aggregate. What we know is that the one place the field's consensus was visible, DraftKings' handle, disagreed with it two days early, and that nobody who read the number can call Sunday a surprise.