Why we publish losing days

Every settled day is published, whether it won, lost or voided. The archive endpoint has no status filter and no way to ask it for the wins. That is not an oversight anyone forgot to add — it is the feature.

Because the alternative proves nothing

A record a bad row can be lifted out of is not a record. Every tipster site on the internet can show you a winning history; producing one requires nothing but a text editor and the nerve. The only thing that makes a published history worth reading is that the publisher had no way to remove the parts they did not like — so the losses are the load-bearing rows, not the wins.

Because it costs us nothing to be honest about it

A day comes off your package only when the main pick wins. A losing day is recorded, published, and your counter does not move: it costs you nothing, so we absorb it. That is the deal the product is sold on, and it is what makes publishing the losses cheap for us and useful for you at the same time.

It is also only affordable because the picks are real. A product inventing them could not afford to give the losing days away.

What you will and will not find

• Every settled day, newest first, one row per day, no gaps.
• Losing days with the fixture, the market, both prices and the final score.
• Voided days with the reason printed beside them.
• No deleted rows, no edited rows, no "best of" page, no filter for the good ones.
• The same rows for an anonymous reader and a paying one — byte for byte.