A probability estimate is not a betting rule until it becomes a price. If your model puts an outcome at 62%, the fair decimal price is 1 divided by 0.62, or 1.613. Any market price above 1.613 pays more than your estimate justifies; anything below it pays less. That single division turns a probability into a decision line.
The conversion in both directions
The odds converter turns any price into its implied probability with 1 divided by the decimal price, and the same formula runs backward: your probability estimate, divided into 1, is the decimal price that exactly breaks even on your estimate. At 62%, that is 1.613. At 45%, it is 2.22. The break-even price moves fast at the short end: 80% converts to 1.25, while 55% converts to 1.82.
American equivalents follow the standard rules the converter documents: at decimal 2.00 or above, multiply the profit part by 100; below it, divide -100 by the profit part. Your 1.613 floor is about -163 in American terms, per the converter's rules, checked 2026-10-02.
Why the market's margin changes the comparison
A raw market price is not a fair price. It carries the book's margin, so comparing your break-even price against the posted price flatters the book: the market's 1.65 on your 62% outcome looks like value, and may still be value, but the honest comparison is against the market's fair price with the margin removed. If the fair price is 1.69 and your floor is 1.613, the market prices the outcome shorter than you do, and the posted 1.65 is not enough.
The sequence is always the same: remove the margin from the market's prices first, then compare your probability against the market's fair probability. The no-vig calculator does the removal on a complete market in one step.
Worked example, start to finish
Say your model rates a tennis player at 62% for the match. Your break-even decimal is 1.613. The book posts 1.65 on that player and 2.40 on the opponent. The implied probabilities are 60.61% and 41.67%, summing to 102.27%, an overround of 2.27%. Proportional removal gives fair probabilities of 59.26% and 40.74%, fair prices of 1.688 and 2.455.
Now the comparison is clean: the market's fair price for your player is 1.688, your break-even is 1.613, and the posted 1.65 sits between them. The price is worse than fair, so the bet fails your rule even though it beats your floor. Your estimate, not the posted number, sets the bar, and the fair price tells you which side of the market you are on.
Where this breaks
The floor is only as good as the probability behind it. A 62% estimate with wide error bars produces a confident-looking 1.613 that deserves no confidence. Convert the estimate honestly, price the complete market, and let the fair price, not the raw one, arbitrate.
One division in, one comparison out. Probability to floor, market to fair, and the decision makes itself.