Maintenance Mindset: What is the hidden connection between system reliability and Super Bowl squares?

Super Bowl squares are often described as a perfectly fair game. One hundred squares. Four payouts. Randomized numbers. Buy a square and you have a one percent chance per quarter. That explanation is clean, intuitive, and incomplete.

Recently, one of my childhood friends reached out to a text string that was made up of 7 boyhood friends. The ask was if anyone wanted in on Super Bowl squares. (For those new to this, squares are a betting pool where you buy a block or more, and that block gets assigned a number position that is on a grid with each team assigned an axis. After each quarter if your block corresponds to a value in the column or row then you win a certain amount.)

If you buy one block, in theory you could win up to 4 times, and in theory the odds all seem the same. However, they are not even close to being the same. Football scoring is not random in the way dice are random. It is discrete, biased, and time-dependent. Once you account for how scores actually accumulate over a game, the odds are no longer flat across time, even if the board itself is randomized.

Understanding that difference is the key to understanding how many squares actually matter.

Why the numbers are not equally likely

Football scores move in chunks. Touchdowns, extra points, field goals, missed kicks, two-point conversions, and safeties create strong bias in the final digit of a team’s score.

Over many games, certain digits appear far more often than others. Zero and seven dominate. Three appears frequently. Four and six emerge from common score combinations. Digits like two, five, eight, and nine appear far less often.

This means that the probability of landing on a given square is not driven by uniform randomness. It is driven by how football is played.

What randomized boards actually do

Randomizing the numbers on the grid is often described as eliminating bias. That is true only in the aggregate. Randomization spreads scoring bias evenly across the board before kickoff, but it does not change the fact that the game itself evolves through constrained states early and more diverse states later.

The board is static. The game is dynamic:

• At the end of the first quarter, scores are tightly clustered. A small set of outcomes accounts for a large fraction of real games. Probability is concentrated.
• By halftime, the score space widens, but not evenly. Bias remains.
• By the end of the third quarter, dispersion increases again. The probability mass spreads across more combinations.
• By the final score, the distribution is the widest it will be. Bias still exists, but it is diluted by time and accumulation.

This matters because a square that has little chance early can become viable later, not because the square improved, but because the system evolved.

How many squares actually matter, and what actually improves your odds

Owning more squares does not change which outcomes are likely. It changes how much of the evolving outcome space you occupy.

Early in the game, probability is concentrated. Owning additional squares helps only marginally because most outcomes live in a narrow band.

Later in the game, probability is dispersed. Additional squares become more valuable because they cover a broader range of plausible score states. In other words, coverage matters more as time progresses.

It is important to separate two ideas that are often conflated. Expected value scales linearly. If you own ten percent of the board, your expected share of the total payout is ten percent. That does not change with scoring bias or time. What does change is the probability of experiencing a win at least once. That is a variance problem, not a value problem.

From a purely mathematical standpoint, buying more squares always increases your odds proportionally. From a practical standpoint, the benefits are not linear in how they feel.

• One square leaves you almost entirely dependent on a narrow early distribution.
• Two or three squares still leave most outcomes uncovered.
• Around five squares, the chance of seeing at least one favorable outcome across four quarters becomes psychologically meaningful.
• Beyond ten squares, gains continue but with diminishing experiential returns relative to cost.

At some point, you are no longer buying better odds. You are simply buying more certainty at a higher price.

For most boards, the best balance sits between five and ten squares. Below that range, your probability of seeing a win remains low enough that outcomes feel arbitrary. Above that range, cost rises faster than perceived benefit. This is not optimization for profit. It is optimization for engagement under uncertainty.

The broader connection to reliability

This article is not really about football.

It is about understanding systems that are biased, time-dependent, and path-driven. Early states are constrained. Later states are diverse. Coverage becomes more valuable as uncertainty unfolds.

Whether the system is a football game, a reliability program, or a portfolio of decisions, the principle is the same. You do not win by assuming uniform randomness. You do better by respecting how outcomes actually emerge over time.

Super Bowl squares look simple because the grid is simple. The game beneath it is not. The odds are shaped by scoring structure, by time, and by accumulation. Buying squares does not change the game. It changes how exposed you are to its evolution. The smartest question is not “Which square is best?” It is “How much of the game do I want to own as it unfolds?”

By the way, I bought 5 blocks.