Signal Analysis · 3 of 3
Reading a price
Every price is a probability with a margin on top. Turning one into the other is a division, and it is the only arithmetic you need to read anything we publish.
Break-even
Divide 100 by the decimal odds.
100 / 1.61 = 62.1% 100 / 1.90 = 52.6% 100 / 2.50 = 40.0%
That is the break-even: how often a bet at that price has to win for you to end up where you started. Below it you lose money over time, above it you do not.
It is not an opinion and it is not ours; it comes straight from the number the bookmaker printed.
Why a hit rate on its own says nothing
A 61% hit rate sounds good until you know the price. At odds of 1.57 the break-even is 63.7%, so 61% loses money. At odds of 1.90 the break-even is 52.6%, so the same 61% is well clear.
The same percentage is a win or a loss depending on a number that is not in it. This is why every read we publish carries the price it was sealed at, and why a record without its prices cannot be judged.
Reading the n before the percentage
87% over 23 matches and 55% over 4,000 are not comparable, and the first is not better. A small sample moves a lot: eleven wins out of fourteen looks spectacular, but at those prices you would see it roughly one time in nine by luck alone.
Before reading a percentage, look at how many matches are under it. If the count is not shown, the percentage is decoration.
The margin
Add up the break-even of both players in a two-way match and you will get more than 100%. The excess is the bookmaker's margin: it is how the business works, it is in every price, and it is the reason accuracy alone does not pay.
What we publish
Every match we read gets a probability, published and timestamped before the first serve. It does not move afterwards. When the match ends it is graded, and the losses sit in the same table as the wins, with the price each one was sealed at.
The record is public and needs no account: tt-lossbeater.com/performance
18+ · Analytics, not betting advice.