A moneyline bet is a bet on which team wins; deciding whether the price is worth taking requires more than choosing the likely winner.
That is the trap in “they only need to win.” A favorite can win more often than it loses and still be too expensive at the offered price. Before making a football argument, work out what that price asks you to believe.
This is a price-reading guide, not a current pick. The sourced examples below explain odds notation. They are not live offers, suggested stakes or predictions for an NFL team.
The price has a question hidden inside it
For a bet that either wins or loses, the break-even win rate is:
Amount risked ÷ (amount risked + net winnings if successful).
Here, “net winnings” excludes the returned stake. Taxes, separate fees, ties, voids and special settlement terms are outside this simple calculation; read the actual market rules before applying it.
FanDuel’s NFL betting explainer uses −120 and +120 to demonstrate American odds. Using those published teaching examples gives this arithmetic, checked November 3, 2026:
| Source example, not a live quote | Example amount risked | Net winnings if successful | Calculated break-even win rate |
|---|---|---|---|
| −120 | $120 | $100 | 120 ÷ 220 ≈ 54.55% |
| +120 | $100 | $120 | 100 ÷ 220 ≈ 45.45% |
Read these as separate notation examples, not as both sides of a captured game market. Do not use them to estimate a bookmaker’s margin.
The minus sign tells you the amount risked for the quoted net winnings. The plus sign tells you the net winnings for the quoted stake. Neither sign is a team-quality grade, and neither calculated percentage is a forecast.
“More likely to win” is not the same as “priced well”
In the sourced −120 example, a team would need to win more than approximately 54.55% of comparable settled win-or-loss bets for positive expected net winnings under the stated assumptions. Winning more than half is not enough.
The difficult part is not the division. It is supporting a probability estimate that deserves to be compared with that threshold.
Ask where the estimate came from. Does it predict an outright win or a different outcome? Was it recorded before the event, using information available then? Does it perform sensibly on games it did not learn from? If those answers are missing, a confident story about the matchup does not fill the gap.
Do not present the quoted price’s own implied percentage as an independent model opinion. That merely compares the price with itself.
A close game and an upset are different bets
An underdog can lose narrowly and still beat a positive point spread. Its moneyline selection needs the outright win, subject to the market’s settlement rules. A favorite can win the game yet fail to cover the points it gives away.
So “this game will be close” does not, by itself, establish a moneyline case. You need to distinguish the chance of a close loss from the chance of a win.
The same caution applies to model output. A projected margin is not automatically a win probability. Turning one into the other needs an explicit model of possible results and evidence that the conversion is useful. There is no universal spread-to-moneyline shortcut supplied here.
When comparing the two markets, write the selected team and the sign of its spread in full. Do not silently swap a home-team margin for a selected-side line. If the source convention is unclear, stop the comparison rather than guess which team the number favors.
Keep the football case and the receipt together
A useful decision note contains:
- The exact market: team, opponent, event date and whether the quoted market includes overtime.
- The offered price: sportsbook and capture time, not a remembered number from earlier in the week.
- The estimate: outright-win probability, its source and the information cutoff, or an honest statement that none is available.
- The uncertainty: which missing input or plausible alternative would change the conclusion.
- The alternative: the actual spread and its price, assessed under the same game assumptions.
This makes a change visible. If the price moves or the underlying information changes, the old comparison is stale. Do not keep the original recommendation and quietly attach the new number.
What the Number Says
The price gives you a threshold. It does not supply the evidence that your team clears it.
Treat a no-bet conclusion as a complete answer when the estimate is missing, the market cannot be matched or the uncertainty swamps the claimed difference. There is no need to turn every game into an opinion with a stake attached.
Where these numbers come from
FanDuel Research’s NFL Betting Explained, checked November 3, 2026, supplies the −120 and +120 teaching examples and their stake/net-win meanings. The break-even percentages are derived above, not observed results or model predictions. The example amounts explain notation; they are not stake recommendations.
No current odds, team win probability, profitable betting record or historical moneyline advantage is claimed. For any actual comparison, the next thing to check is whether the recorded price, settlement rules and pregame information still match.
Breakeven win rate at recorded American prices
Breakeven probability is calculated only from American prices that were actually captured in the odds-history table.
Questions worth answering before kickoff
Is moneyline better than taking the points?
There is no automatic winner between the two. They settle on different outcomes and have different prices. Compare a supported outright-win estimate with the moneyline, and a supported cover estimate with the spread. Do not use one probability for both.
Should I always take a small underdog on the moneyline?
No. A narrow spread does not establish how often that team wins outright at the offered moneyline price. A close-loss scenario can matter to a spread selection without helping the moneyline.
Does the implied percentage tell me the true chance of winning?
No. It translates a price into a break-even threshold under specified settlement assumptions. It is not an independent forecast. If you want to estimate fair probabilities from a full market, you also need the other outcomes’ prices and an explicit treatment of the bookmaker’s margin.
