Multiplying the legs is the right sum only when the legs are independent. Same-game legs are not.

Free. No account and no subscription. Needs macOS 14 or later on a Mac with Apple silicon (M1 or later). 34.9 MB, notarised by Apple.
A team covering the spread makes the over more likely, not less. A quarterback having a big passing day makes his receiver's yardage prop more likely. Multiply those together as though they were coin flips and you get a probability that is too low. That makes the fair price look higher than it is, and the parlay look better than it is.
These legs are in one game and are not independent. A team covering the spread makes the over more likely, not less. Books price same-game parlays with a correlation adjustment for exactly that reason, so the independent figure above is an upper bound on what this is worth, not an estimate of it.
Books price same-game parlays with a correlation adjustment for exactly this reason. That adjustment is theirs and it is not published, which means the independent calculation you can do is a ceiling on the value, never an estimate of it. Any tool that quotes you a same-game parlay edge from multiplication alone is quoting a number it cannot support.
What survives: correlation cuts both ways, and a negatively correlated pair is the one books misprice more often. That is a real edge and it needs the correlation measured, not assumed.
Reading a same-game price
Work out what each leg is worth on its own, multiply those chances, and adjust for legs that move together. That gives the fair price of the ticket. Then compare it with the book's offer: a book's same-game price is usually shorter than the legs multiplied, and the gap is its margin on the whole ticket, charged once per leg and compounded. Take the ticket only when the offer is longer than the fair price. If it is not, the parlay is a more expensive way to make the same bets.