A better Super Bowl pool

It's the last week of January, and that means you've likely dropped $10 or $20 to pick a square in the office Super Bowl pool. The typical pool involves each person getting a pair of numbers, and if the last digit of the scores at the end of a quarter matches your numbers, you win $200.

By Friday, about half of you will regret joining the pool, when you get "quality" numbers like 2, 8, and 5.

I remember, years ago, playing a different pool, where at the beginning of the playoffs, you put money on any of the teams. For each team that makes the Super Bowl, anyone who put money on that team shares part of the pot (say 70% for the winner, 30% for the loser). This leads to an interesting choice: do you put money on the team that's sure to make the Super Bowl, and receive less of a reward, or put a dollar on the #6 seed, and share a larger portion of the pot if the team makes it? Normally, the #6 seed is a sucker bet, but not this year!

So if you're staring at the power combination of 5 and 8 in your Super Bowl grid, consider a new pool next year, combining the traditional grid with this second pool.

Instead of the traditional grid with the numbers for each determined randomly after the fact, let each person choose their own spot on the grid! Have each person's entry be divided among 101 known buckets: each of the AFC X, NFC Y combinations, and Not Chosen. You could drop $5 on 0-0, $2 on 3-0, $2 on 0-3, and $1 on 3-3.

Scoring would be as normal, at each quarter and at the end of the game (20% each quarter, 20% for the final score). For the winning square, the prize would be divided among all people who dropped money into the square, in proportion to their bet. (If 1 person bet $2 and 2 people $1 on the square that wins $200, one person would get $100 and the others $50.)

If no one chose the winning square, then everyone who selected the Not Chosen bucket would split the prize.

This kind of pool would not rely on being lucky and getting 3-3 or 0-0. You would rely on your prognostication. You could choose any combination you thought likely to come up. But then comes the "metagaming" prognostication. What is the return really worth for your 0-0 pick, considering the number of people who will select it? Do you play for the rarer combinations (8s, 1s, 4s) that are more likely to come up in the final score? And how many squares are likely to be unpicked for the Not Chosen bucket?

This pool can also run with any number of participants, for any level of contribution from $1 up.

If you try it, let me know how it goes!

Update: Welcome, Carnival of the Vanities readers.

In an interesting twist, I got the desirable 3-3 in a slightly modified grid pool (it pays on each scoring change, so if one team gets a field goal, there's a chance for payoff on both a field goal and two touchdowns (13-3 before 14-3)).