Reviewed 2026-09-05. Educational research; no wager required.
Your first five minutes
Start with NFL or college football. Choose one upcoming game on the league page and open its analysis. Your goal is a written explanation you can revisit, not a bet slip.
Check the event date, start time, and update time before reading the forecast. Confirm you are looking at an upcoming game rather than a completed matchup. If coverage is unavailable or stale, stop there; use the example below instead.
1. Separate the projection from the price
Find the available model projection and the displayed market line. These answer different questions: the projection describes a model's estimate; the line and odds describe an offered market. Neither is a settled result.
Illustrative example only: a model total of 46 against a market total of 44 is a two-point disagreement. That alone does not establish positive expected value. You still need the relevant outcome probability, the offered odds, and confidence that the inputs are current. These numbers are not a live Olympus prediction.
2. Read one reason and one uncertainty
Inspect the context actually available for that matchup. For football, consider quarterback availability, injuries, weather, expected pace and opponent quality. A factor missing from the page is something to verify separately, not something to assume the model incorporated.
Write two sentences: My current read depends on ____. I would reconsider if ____. If the explanation cannot support those sentences, keep researching or pass.
3. Check the price you can actually get
A displayed line can move. Record the line, odds and time you observed, and compare them with the available offer. Do not treat an old quote as executable or a model label as a guarantee.
Passing is a complete research decision. No wager is required to use this workflow, complete onboarding, or learn from the result.
Copy this research note
Game / research time / source update time / projected outcome / displayed market / available price / supporting factor / unresolved uncertainty / decision: research further or pass / what would change my view.
On another day, revisit the note. What changed in the inputs or market? After the game, distinguish the quality of the reasoning from the outcome. One winning or losing game proves very little; inspect a dated, broader record including losses.
Player research or an AI assistant?
For player questions, start with expected role and opportunity, then opponent and availability. Do not assume every sport has the same player-market coverage. A missing player projection is missing information.
For AI-assisted research, use the MCP setup guide linked below. Ask the assistant to identify timestamps and missing fields, and keep forecasts distinct from settled results. Never paste a personal token into a public prompt or social post.
Make the next session easier
After signing in, choose your sport and research style in the welcome guide. Your first-week lessons use the same workflow. Board-email delivery is a separate preference you can manage in your account.
If a paid feature remains locked, reopen your welcome email and use the sign-in instructions. Contact support before buying a second subscription. Keep your access code private.
Your next step
Evidence and limitations
Prices and features can change. Competitor details reflect the sources linked here, not independent performance testing. Verify current coverage and billing terms before subscribing.
Research does not remove uncertainty. No outcome or profit is guaranteed. Never wager money you cannot afford to lose.