Goodguy AI Labs · Open weekly experiment
The Fantasy Fruit Fly
A side experiment: Are you smarter than a fruit fly?
An experiment to see whether a fantasy-football expert can emerge from a measured map of fruit-fly wiring. We supply the public NFL facts. The calculations—and the calls—are the Fly’s.
Software fetches, joins, and checks the record book. A fly-derived pathway learns how to turn those facts into PPR, Half PPR, and D/ST rankings. It does not copy an analyst list or ask an AI for the answer.
Week 1 is a blind reconstruction: results from games already played were not fed back into the calculation. The depth chart is a role reference, not a promise about snaps.
- Board
- 288 rankings
- Scoring
- PPR · Half PPR · D/ST
- Answer key
- None
Rankings stay blank until a complete release passes the page’s checks.
The Fly’s player rankings
Week 1 rankings by position.
A real rankings board, built from the Fly’s own calculations. The starting universe is one QB, two RBs, three WRs, and two TEs per team, plus every D/ST. Open a row for its range and role reference.
Loading the released board…
| Rank | Player | Matchup | Projected points | Likely range | Role reference |
|---|---|---|---|---|---|
| The board appears after JavaScript verifies the release contract. No third-party rankings or Week 1 results are used. | |||||
The released board scrolls inside this frame. Search or filter first, then scroll to inspect more players.
No released players match those filters.
Not numerically ranked
Unrated players
Players without enough applicable history are kept out of the numerical board instead of receiving a made-up estimate.
View unrated players
Human vs. model
Take the five-call challenge.
Choose one player from each same-position pair before the first listed kickoff. Pairs come only from the current release. This browser-only practice round has no verified leaderboard and is not personal roster advice.
Waiting for a current board before setting a lock time.
Five pairs will appear only when the current released rankings contain enough same-position comparisons.
Make a call first. Final outcome records are needed before this round can be scored.
Honor system: we do not send or verify your picks. If browser storage is unavailable, the page will say so rather than pretending your calls were saved.
Past practice rounds saved in this browser
ELI5: the method
The wiring is evidence, not an answer key.
Imagine a marble machine. Software puts clean NFL facts in the hopper, keeps each player’s identity and role attached, and lets the Fly-derived pathway decide what those facts are worth. The result is a transparent experiment—not a magic animal and not a copied expert list.
- 01 · FactsPublic NFL record bookSchedules, stable IDs, prior workload, team context, scoring rules, and cutoff-safe depth slots.
- 02 · LearnTwo routes study the same historyA lagged factual companion and the full BANC-derived path see the same record. Neither sees a fantasy answer.
- 03 · EarnUntouched weeks set the influenceAn out-of-sample selector decides whether the fly-derived route gets a voice for each component.
- 04 · RankSimulate the slate 256 timesThe Fly’s component draws become PPR, Half PPR, and D/ST distributions and ranks.
- 01
Facts go in clean.
Schedules, prior production, team/opponent context, scoring rules, and a depth-chart role reference are joined by software. Sourcing the facts is our job; football expertise is not smuggled in.
- 02
The Fly earns its say.
The lagged companion and measured BANC path learn from the same untouched history. A selector can give the Fly-derived route influence, a little or a lot, only when the record earns it.
- 03
The slate gets played on paper.
The selected component models generate 256 deterministic draws. Those draws become the points, ranges, and within-position ranks on this board.
- 04
Week 1 stays out of the answer.
Scores and plays from games already underway were not fed back into this release. That is why this page can be inspected as a blind reconstruction, not sold as a proven forecast.
It is not a living animal or a complete fly brain. The software gathers the marbles; the fly-derived calculation decides how to weigh them. Treat the board as a clearly labeled weekly experiment, not proof of superiority.
Data notes
Show the release, show the limits.
The page reads one same-origin release file. If it is unavailable, blocked, incomplete, or old, it says that instead of filling the board with stand-in rankings.
Release information
- Status
- Checking for a release…
Primary sources & attribution
- Sources appear after the current release passes its contract.
This release’s files
- Sources will appear with a released dataset.
Limitations
- Limitations will appear with a released dataset.
Evaluation context
How this release was checked
These are release-supplied retrospective checks, not a promise about an individual player or week.
Fair read: compare the primary model with every declared control and retain negative results.