-148 is not a prediction. It is a bill. The sportsbook is telling you how often the wager must win before the payouts cover the losses. An American odds converter translates that bill into a win rate.
The cited market is clean enough to do the entire lesson. For the Week 1 game in Melbourne, the source row lists the Rams at -148 and the 49ers at +124 . The conversion is mechanical. The betting judgment is not.
Read the sign before touching the formula
A minus price is the favorite side of the payout. Use the absolute value of the price as the numerator, then divide by that value plus 100. The Rams quote is -148 . The implied break-even rate is 148 divided by 248, or about 59.7% .
A plus price is the underdog side. Divide 100 by the quoted number plus 100. The 49ers quote is +124 . The implied break-even rate is 100 divided by 224, or about 44.6% .
The decimal equivalents are about 1.68 for the Rams and 2.24 for the 49ers . Different notation, same payout relationship. The odds-reading guide covers the formats; the converter puts them on one scale.
Strip the margin before calling it fair
The two implied rates add to roughly 104.3% . The amount over 100% is the book’s overround, about 4.3 percentage points in this quoted pair .
Normalize each side by the combined total and the market-only split lands near 57% for the Rams and 43% for the 49ers . That is a no-vig view of this pair. It is still not an independent forecast.
This is the point most converter copy rushes past. The offered break-even rate and the normalized market rate answer different questions. One prices the ticket. The other describes how the quoted pair divides after its margin is removed. The vig guide is where to slow down and keep those columns separate.
The spread and total are not conversion inputs
The same source row lists the Rams as a 2.5-point favorite and the total at 48.5 points . Those numbers describe the handicap and scoring line. They do not go into the American-moneyline formula.
A spread or total wager needs its own attached price. Without that price, there is no break-even win rate to calculate. Treating -2.5 as if it were -250 is not a shortcut. It is a category error with a calculator attached.
The market row also records the Melbourne broadcast for 8:35 p.m. ET on Netflix . That is useful event context. It does not alter the payout math.
Travel can inform a handicap, not the conversion
The source set says San Francisco is scheduled for 38,105 travel miles, described as a single-season record . It lists the Rams at 34,847 miles . Those are real schedule facts and plausible modeling inputs.
They are not automatic downgrades, and they do not change what -148 means. A model would have to define the travel feature, keep it available before kickoff, and test whether it improves out-of-sample estimates. Until then, the converter should stay in its lane.
That is the broader lesson: a source-backed number can be true and still be irrelevant to the calculation in front of you. Good analysis does not reward a statistic merely for showing up in the same paragraph.
What the price still cannot tell you
The converter cannot decide whether the Rams win more often than 59.7% at the offered price . It cannot decide whether the normalized 57% market view is right . It cannot decide whether the current board still offers either quote.
Those answers require a timestamped market, an independently built estimate, and a pass rule when the inputs disagree. Use the betting desk to record that comparison and the graded ledger to review published decisions later.
The American odds converter does one job well. Let it do that job, then make the handicap earn the rest.
Breakeven win rate at recorded American prices
Breakeven probability is calculated only from American prices that were actually captured in the odds-history table.
Observed market vig versus no-vig fair probability
Each bar is derived from a recorded two-sided moneyline snapshot and removes the paired market overround before comparing fair probability.
