Football Squares Example: One Real Board, All Four Winners

It looks like the board below: a hundred squares with somebody's name in each one, a digit from 0 to 9 above every column and beside every row, and four highlighted squares showing who won each part of the game. Twelve people filled this board between them, everyone paid $10 for each square they claimed, and the group collected $1,000 in total.

The setup happened in a fixed order, and the order is what makes the board fair. Everybody claimed and paid for their squares while the columns and rows were still blank, so nobody could see which digits they were buying, and only once all hundred squares were gone did the host randomly assign the digits 0 through 9 across the top and down the left side. New England were the designated home team in Super Bowl LX, so their digits run along the top and Seattle's run down the left, which is the arrangement we use everywhere on this site.

The money was decided before any of that. A $1,000 prize pool split on a common 20/20/20/40 weighting pays $200 at the end of each of the first three quarters and $400 on the final score, and writing those four figures down in advance is what stops any argument later. You could just as reasonably pay $250 four times instead, since every square carried the same chance before the draw, and our breakdown of football squares payouts runs the same arithmetic at every price per square if $10 isn't your group's number.

Take the score at the moment the period ends, drop everything except the last digit of each team's total, find that home digit along the top and that away digit down the left, and the square where the two meet is the winner. Nothing else about the game matters, and a team on 29 counts as a 9 in exactly the way a team on 9 does.

Here is that read performed four times on the board above, once for each prize. Every row is the same operation, and by the second one it stops feeling like arithmetic at all.

One detail in that table is worth pausing on, because it looks like a mistake and isn't. New England were held scoreless until the fourth quarter, so their digit stayed on 0 for the first three prizes, which is why those three winning squares all sit in the same column of the board. A column that looks cold for most of a game can still carry most of the prize pool, and there was no way for anyone to know that while they were claiming squares.

Because a squares board reads the running total, not the points scored inside that one quarter. The score you want is whatever the scoreboard says at the whistle, which already includes everything from earlier in the game, and this is the mistake that costs a first time host the most money.

Super Bowl LX makes the gap unusually easy to see, because Seattle scored in every single quarter. Put the two ways of reading the game side by side and the second quarter is where they come apart.

Read the middle column at halftime and you pay whoever holds a 6, which is the wrong person by $200. Read it after the third quarter and you pay a 3 instead of a 2, wrong again. Only the first quarter gives the same answer both ways, and it does that purely because nothing had been scored before it, which is exactly why the mistake survives the opening quarter and then bites at the break. Whenever a broadcast shows you a line score with points broken out by quarter, add them up before you go anywhere near the board.

The same rule settles a question that comes up in most groups at some point, which is what happens in a game that goes to overtime. The last prize pays whatever the score actually is when the game ends, so overtime points count toward that running total like any others. Our football squares rules reference works through the overtime ruling and the handful of others a host should settle before the draw.

Three of the four were poor squares by any historical measure, and they won anyway. That is the most useful thing this example has to say, because it cuts straight against the advice every other page on this topic leads with.

We analysed the 1,424 NFL games played from 2021 through 2025 to work out how often each digit pairing actually wins. The first quarter square here, a home 0 against an away 3, is a genuinely strong one: it took 125 of those 1,424 first quarters, or 8.78 percent. The other three were nothing of the sort. The halftime square won 9 halftimes out of 1,424, which is 0.63 percent. The third quarter square won 12 of them, 0.84 percent. The final score square also won 12 finals, another 0.84 percent.

So in the biggest game of the year, on a board where the digits are supposed to favour a small handful of pairings, three quarters of the money went to squares that come up less than once in a hundred periods. The pattern in the digits is completely real and it comes from how football scoring works, with touchdowns and field goals pushing totals toward certain last digits far more than others. Even so, the most common pairing across every scoring period in that dataset, a home 7 against an away 0, still only wins 7.33 percent of the time, so the very best square on the grid loses more than nine times in ten. Our football squares odds data breaks every pairing down by scoring period if you want to look properly after your own draw.

None of this is something a player can act on anyway, because the digits don't exist while the board is being filled in. Nobody picks their own numbers, and the only thing anyone chooses is a position on a blank grid. Where the data genuinely helps is on the host's side, as a decent argument for paying more than one prize and for adding the reverse score rule, both of which spread the money across more of the room.

The square where the two zeros cross wins the full prize, and a square that wins twice gets paid twice. Neither case needs a special rule, and inventing one after the fact is how a friendly board turns into an argument.

A scoreless quarter feels like it should be void, and it very much isn't. A 0 is a last digit exactly like a 7 or a 3, and the pairing of two zeros is the single most common way a first quarter ends in our dataset at 14.47 percent of them, so the person holding it deserves the prize as much as anyone. Super Bowl LX came close to demonstrating this, since New England sat on 0 through three quarters and it was only Seattle scoring that moved the winning square down the column each time.

Repeat winners work the same way. If one square happens to match at two different breaks, pay it both times, because that square had an identical chance to the other 99 at the moment somebody claimed it and it did nothing unusual to win twice. Some experienced hosts prefer moving a repeat prize to a random square or shifting it by a pre-agreed digit offset, which are reasonable choices in particular situations rather than better defaults, and either way the decision has to be written down before the digits are drawn.

The case actually worth planning for is a winning square nobody bought. The prizes you advertised still have to be paid in full, so the cleanest fix by a wide margin is filling the board before the draw, with the host or the players between them taking whatever is left at the price everyone else paid. When that genuinely isn't possible, roll the unclaimed prize into the next scoring period so it grows a later one, and decide in advance what happens if it is the last prize that lands empty, either splitting it across the squares that were claimed and played or drawing a random square and paying whoever holds it.

The reading works identically, but each square covers four different score combinations instead of one, because the digits get grouped two to an axis position rather than one. A 25-square board is five across and five down, so each column carries a pair of home digits and each row carries a pair of away digits.

Run Super Bowl LX's final score through one and you can see the difference straight away. Say the columns are grouped 0 and 1, then 2 and 3, then 4 and 5, then 6 and 7, then 8 and 9, with the rows grouped the same way. New England finished on 13, so their digit 3 falls in the second column, and Seattle finished on 29, so their digit 9 falls in the fifth row. The square where those two meet takes the last prize, and that same square would also have won on a 12 to 28 final, a 12 to 29 final or a 13 to 28 final, because all four of those score endings map to the same cell.

That is the honest trade-off of a smaller board rather than a flaw in it. Fewer squares means each one wins more often and the group fills the board faster, and it also means the prize pool is smaller at the same price per square, since 25 squares at $10 collects $250 rather than $1,000. Plenty of groups prefer that, and the choice should come from how many squares you want to sell and how your group likes to play rather than from a headcount rule. Our guide to how many squares are in a football pool covers all four standard sizes, including the ten-square format that works off the combined score of both teams instead of a grid.

Everything above works on a sheet of paper taped to a kitchen wall, and that is how most people first met this game. Set the price, write the four prizes down as actual dollar figures, get all hundred squares claimed and paid for, draw the digits once in front of everybody, and then leave the board alone until the first quarter ends. If you want the mechanics at more length than a worked example goes into, our walkthrough of how to play football squares covers the same board from the beginning.

What a paper board can't do is answer questions while the game is on, and that is where a host's evening actually goes, fielding the same three messages from thirty people who want to know which squares they hold, whether they just won, and who took the first quarter. Running a football squares board on Play Squares moves all of that off the host. The score updates on its own, the grid highlights the square currently on track to win the period being played rather than only the settled winner once a quarter closes, and everybody can pull the board up on their own phone whenever they feel like checking.

The board at the top of this page is that product showing a finished game, which is why it can mark all four winners and name the prizes they collected. Yours would look the same by the end of the fourth quarter, with the digits randomized at the press of a button in front of the group and one central record of who has paid. Work through this football squares example once and you already know everything you need to run one for your own crowd.