Source: https://www.pro-football-reference.com/ | Tools: Excel heat map : )

I could be misunderstanding this. How is there a 0.9% probability of a 1-0 game (for Eagles)? I thought, from Scorigami, that was impossibe.

How do you get these numbers from the win probability?

Thank you for the question, and sorry for the misunderstanding. Those digits represent the final digit of each team's final score. So [1,0] could mean [Eagles 21, Chiefs 20], [Eagles 31, Chiefs 20], [Eagles 41, Chiefs 30], etc. In these examples and others like it, whoever owned the [Eagles 1, Chiefs 0] square would win the pool.

Ohhhh okay. So like 0-0 isn't a tie it's like 10-0 or something. Got it. Thanks for the explanation.

Those are based on current betting markets, which price the Eagles at -125 and the Chiefs at +105. -125 implies a 55.56% chance to win, and +105 implies a 48.78% chance to win. That sums to more than 100%, as the sportsbooks factor in a cut (or "vig") for themselves. So you have to adjust those back down to 100% to determine actual sportsbook-implied probabilities.

[Edit: Added a missing word]

Do you have this code somewhere? I think this is interesting, and I'd like to see how you did it

Not sure where you're located, but the way betting odds are presented can vary quite a bit from place to place. -125, +105 are "American odds." The tool I've linked to below is a nifty little way to convert these to decimal odds or British odds (both of which are more common internationally), as well as to win probabilities.

https://www.covers.com/tools/odds-converter

Yes, [0,0] would be like 10-0, 20-10, 30-20, etc. I should consider how to make this more clear on the standalone image because right now it sort of requires prior knowledge of how "Super Bowl Squares pools" work.

Most of these pools do offer smaller prizes if you have the correct square after the first, second, or third quarters as well. (The final score is the big pot.) So practically speaking, 0-0 is actually conceivable if no team scores in the first quarter, for instance.

So. I want combinations of 7, 4, 3 and 0. Just like every year since 1982? Shocking.

I will end up with 2 and 6 somehow...

place your bets on 7-0 or 0-7

Haha, won't we all with the 2 and the 6. Yes, you still want 0, 7, 4, and 3... but as compared with the rest of the Super Bowl era (since 1966) prior to the time period shown above, the proliferation of the NFL's weird score era has made these Squares more equitable, shifting over seven percentage points' worth of probability from the traditional stalwarts (0,7,4,3,1) to less-common numbers (2,5,6,8,9). So getting 2s and 6s like you've mentioned actually isn't as bad as it was from 1966 to 2014 (or 1982 to 2014).

With the addition of the 2 point conversion, increase of the yardage for the extra point conversion, and more teams trying to convert on 4th and short as opposed to kicking a FG, it makes sense.

For example, 4 and 7 are each only about 80-85% as common as they were before 2015. (In 2015, the NFL moved the extra point distance back from 20 yards, where kickers converted 99% of the time, to 33 yards, where they now make only 94% of their extra-point kicks. Combined with near-record levels of field goals and 2-point conversion attempts, scores have gotten a lot weirder than they used to be.)

It's usually a random draw, but some of the sportsbook apps actually offer these now with the opportunity to pick.

Unfortunately, of course, they assign different payout odds to each square, which would be less (or a lot less) for the more common number pairs. And they do it in a way, mathematically, whereby they can pretty much guarantee that they make money.

For example, if a square with a 3.0% probability were fairly priced, it would be priced at +3233 (i.e., risk $100 to make $3,233). They'd probably price that square around +2900 instead, essentially creating negative expected value for bettors (who run the math) in every single square.

have no clue how to read this :/

Blue is best probabilities

It’s very literal. The percentages are % likelihood of that particular square being the final result

So, would my Birds winning 32-12 be Scorigami????

And each number represents the final digit of each team's final score. Which I realize now is not super clear when just looking at the standalone image if someone is not already familiar with how these pools work. Will look to remedy that next year.

[2,2] shows up as 0.0% due to rounding. It has actually happened 3 times as a "final digit of each team's final score" number pair in the 13,557 regular season or playoff games in the Super Bowl era (since 1966). I've linked to the games below.

None of those instances was 32-12, which according to the Scorigami website, has never happened before, and would therefore be Scorigami. (Not sure how far back Scorigami looks; I always assumed it covered "all" of NFL history.)

Buccaneers 42, Raiders 32 Nov 4, 2012 https://www.pro-football-reference.com/boxscores/201211040rai.htm

Bills 42, Dolphins 32 Dec 5, 2004 https://www.pro-football-reference.com/boxscores/200412050mia.htm

Chiefs 42, Bengals 22 Oct 26, 1969 https://www.pro-football-reference.com/boxscores/196910260kan.htm