Back To Projects

Weekly NFL Prediction Tracking

AI Analyzer NFL

A public NFL analytics site following weekly AI predictions, matchup research, and results across the 2026 season.

Tech stack (4)
HTMLCSSJavaScriptGitHub Pages

Product

Weekly NFL Prediction TrackingNFL AnalyticsModel Comparison2026 Season

See It Live

Built To Work On The Screen Most People Reach For.

Most people try a free tool on their phone before they ever open it on a desktop. This is how the app actually looks and works there, screen by screen.

  • Most people reach for a tool on their phone first, so the mobile experience is designed to carry that weight, not squeezed in after the desktop build.
  • Every screen shown here, not a single cropped view, so you can see how the whole app holds together on a real screen.
  • The exact live build, captured straight off the shipped app.
Open the live site
Full mobile page for AI Analyzer NFL
Scroll inside the phone to move through the page

Project Overview

How The Project Came Together.

01

Problem

I built AI Analyzer NFL to follow how ChatGPT, Claude, and Gemini approach the same football matchups. Each model works with a hypothetical $20 budget per game, including the option to make no bets.

02

Approach

The current-season site brings together schedules, team identities, published predictions, and raw response records. Results and reasoning are reviewed separately after games, while the original 2025 experiment stays in its own archive.

03

Result

This is an ongoing CSolutions project and a source for weekly updates about what the models picked, what happened, and which changes to the prompts are worth testing next. It is independently maintained, with no live-score feed or claim of guaranteed returns.

Challenges & Learnings

What It Took, And What Came Out Of It.

01

Challenge

Making weekly matchups and model comparisons easy to follow on phones and desktop screens.

Learning

A correct prediction and a sound explanation need separate review.

02

Challenge

Keeping current-season predictions separate from historical results and corrections.

Learning

Team identity, clear status labels, and consistent matchup pages make dense data easier to navigate.

03

Challenge

Preserving the evidence behind each prediction while publishing through a lightweight static site.

Learning

Weekly publishing needs repeatable checks and preserved source records before it needs more infrastructure.

Like What You See?

Let's Build Yours Next.

Tell me about your business and the kind of site you want. I'll send back scope, timing, and a clear quote.

Stay Connected

Follow Along for Recent Work, Behind-The-Scenes Updates, & New Launches.

Keep up with website launches, client projects, and business updates from CSolutions.