ODDS_API_KEY is not set in this environment,
so there is no odds feed to read and nothing below invents one. Every price field is
marked SAM-FILL until a real number is put in it.
The pick is logged as PENDING-LINE in the prediction log and
will not be graded. That rule exists because of a specific past entry: a Pirates
pick conditional on "-130 or better" was graded a loss with
price_condition_met: UNKNOWN, because no line was ever recorded. A pick without
its price cannot be scored, only argued about.
| Market | Price | Notes |
|---|---|---|
| Spread | SAM-FILL | Which side, and by how much. Write the book down next to it |
| Total | SAM-FILL | Watch for 41 / 43 / 44 — the key numbers where a half point is the whole edge |
| Moneyline PIT | SAM-FILL | Home dog or home favourite changes the whole read |
| Moneyline CIN | SAM-FILL |
Everything below is either a fact from a feed the hub already pulls, or reasoning that does not depend on a price. None of it is a pick.
ESPN's Football Power Index, the only free rating in the repo, has CIN +1.63 and PIT -0.87 — Cincinnati about two and a half points better on a neutral field, so roughly a coin flip once home field is added back.
But it is dated 2026-08-18 and ESPN last updated it 2026-06-02. That is before a single 2026 game was played. It is a weak prior about rosters, not a read on two teams three weeks into a season. Treat it as the floor of what is known, not evidence.
This is the part that is actually actionable today, and it is why this game matters twice.
| Player | Team | Whose roster | Proj | Status |
|---|---|---|---|---|
| DK Metcalf | PIT | SE7EN — starting WR | 10.6 | Starting. Locks at 1:00 |
| Mike Gesicki | CIN | Hand Banana — bench TE | 4.9 | Benched, correctly — Fannin (7.4) starts |
Projections are the Sleeper week-3 feed (443 players, generated 2026-09-21), the same source the roster page uses. A rooting conflict worth knowing about: Metcalf scoring is good for SE7EN and good for the Steelers, which for once points the same way.