The Road Ahead

The current CFB Ranks algorithm evaluates teams based on wins, opponent strength, location, margin, and recency. It's transparent, it's testable, and it produces credible results.

But we know it can be better.

What follows are ideas - starting points, not solutions. None of these have been implemented or tested. They're concepts we believe could improve the ranking process, shared here to spark conversation and invite input. Some may work. Some may not. That's what the consensus model is for - figuring it out together.

What we're exploring:

• - The Eye Test - Capturing structured human judgment from people who actually watch the games
• - Player Health - Accounting for who was actually on the field
• - In-Game Context - Filtering garbage time and weighting competitive situations
• - Weather - How extreme conditions affect teams differently

The "eye test" is real. Anyone who watches football knows that box scores don't tell the whole story. A team can win ugly or lose impressively. The question isn't whether human judgment matters - it's how to capture it transparently.

Here's one approach:

Structured evaluation, not gut feeling

Approved participants commit to watching a game and then submit a rating on a simple scale:

Participants can rate the full game, each half separately, or all three. Along with the score, they provide feedback explaining their evaluation.

Aggregated and accountable

Ratings are collected across all participants who reviewed a game. The results are tabulated into an average that becomes a point modifier for that team's game score.

Each participant's individual rating is compared against the group average. Over time, you can see who rates consistently and who's an outlier. Bias becomes visible - and that's not inherently wrong. Bias is human nature. A Longhorns fan will always hope for Texas to succeed. An Oklahoma fan will see things differently. That's expected. The goal isn't to eliminate bias - it's to make sure it doesn't unbalance the results. By aggregating across many participants and tracking individual patterns against the average, natural biases tend to balance out while extreme outliers become obvious.

Weighted like everything else

The Eye Test modifier will be weighted within the broader algorithm, just like margin, opponent strength, and recency. The community will help determine how much weight it deserves.

This isn't about replacing data with opinions. It's about adding structured human judgment to the equation - openly, accountably, and in a way that can be questioned and improved.

A win is a win. But was it a win with your starting quarterback, or your third-string backup?

Current rankings treat all wins equally regardless of who was on the field. We're exploring ways to incorporate player availability data:

• - Key player injuries and their impact on team performance
• - Games where starters sat (blowouts, rest before playoffs)
• - Depth chart changes that significantly affect team strength

The challenge is doing this fairly and consistently. A torn ACL to a Heisman contender is obvious. But how do you quantify the impact of an injured left guard? This is where community input will be critical.

Not all points are equal.

A touchdown in the fourth quarter of a blowout doesn't mean the same thing as a touchdown in a tie game. Current rankings don't distinguish between the two. We're exploring:

• - Garbage time filtering - Adjusting margin calculations when games are already decided
• - Game state context - Weighting plays and points based on competitive situation
• - Performance under pressure - How teams play in close games versus blowouts

The data for this exists - our current source (CollegeFootballData.com) provides play-by-play data with garbage time flags and expected points metrics. The question is how to weight these factors fairly and whether the added complexity improves the rankings.

A 35-10 win in perfect weather isn't the same as a 17-14 grind in a blizzard.

Extreme conditions affect how games are played - but they don't affect all teams equally. A team from Wisconsin playing in snow is different from a team from Miami playing in snow. Home teams in extreme weather have a natural advantage that goes beyond the usual home field bump.

We're exploring:

• - Extreme weather flags - Identifying games played in conditions that significantly impact play (snow, heavy rain, extreme cold, extreme heat)
• - Climate familiarity - Teams accustomed to certain conditions may deserve different treatment when playing in those conditions
• - Neutral site adjustments - Championship games and bowls often move southern teams north or vice versa

The data exists for game-day weather. The question is whether the added complexity produces fairer rankings or just adds noise.

These are starting points, not a complete list. The consensus model exists precisely because we don't have all the answers.

If you're a coach, analyst, or expert with ideas about what should factor into rankings, we want to hear from you. The best improvements will come from the people who know the game.