How we calculate odds, value and forecasts
Every figure on a competition card is worked out the same way everywhere on CompWatch: on cards, in the table view, in alerts and in the emails. This page says how, in plain words. Trust grades are a separate thing: see how we rate sites.
Odds now
1 in 250
Tickets sold so far, plus yours, divided by the number of winners. It only gets longer as more tickets sell.
Odds if it sells out
1 in 1,000
The total number of tickets, divided by the number of winners. The fairest way to compare two draws.
Prizes per £1
£0.85
If every ticket sells, the prize value each £1 of tickets buys, on average. Most entries win nothing.
Example figures, to show the idea.
Prizes per £1
We take the main prize's value (the cash alternative the operator states, or our own estimate when it does not state one, marked (est.)), multiply it by the number of winners, add the value of any instant-win prizes still to be won, and divide by what the whole ticket run would raise: total tickets times the ticket price.
We use the odds-if-sold-out basis on purpose. A draw that has only just opened looks generous on the figures so far, and that generosity disappears as tickets sell. When a prize value looks wrong (more than the tickets could ever raise, for example) we leave the figure out rather than show a number we do not believe. Free entries and draws with an unknown ticket cap have no figure.
The sell-through forecast
For a competition that is still open, the forecast estimates what share of its tickets will have sold when it closes, and so what the final odds will be. It learns from how earlier draws with the same shape sold: the same operator and kind of prize where we have enough history, with how far through the draw's life it is. Each forecast carries a confidence label (high, medium or low) that reflects how much history sits behind it.
We test the forecast against two simple rules of thumb. The first assumes the draw finishes where that operator's draws usually finish: their average over the last 12 weeks, and never less than what has already sold. The second assumes no more tickets sell. Where the forecast has clearly beaten the operator's usual finish in testing, we show the forecast. Where it has not, we show the operator's usual finish itself, provided that has clearly beaten assuming no more tickets sell. A forecast built on how the operator's draws usually sell is labelled that way on the competition. Where neither has earned its place at a confidence level, we show no forecast.
We have not published an accuracy test yet. Until we have, forecasts are not shown anywhere on the site.
A forecast is an estimate from history. A draw can sell faster or slower than similar ones did, so treat it as a guide and rely on the odds if it sells out when you want a firm number.
Frequently asked questions
- Why do you show the odds if a competition sells out?
- The odds right now flatter whichever draw launched most recently, because few tickets have sold. If every ticket sells, your chance is one in the total number of tickets (divided by the number of winners), so that is the fairer way to compare competitions.
- What does £1.00 of prizes per £1 mean?
- If every ticket sells, each £1 of tickets buys that much of the prizes on offer, counting the main prize and any instant wins still to be won. It is an average across everyone who enters: most entries win nothing, and a figure above £1.00 does not mean you will win.
- How accurate is the sell-through forecast?
- We have not published an accuracy test yet. Until we have, forecasts are not shown anywhere on the site.