Sport News

By Martin · Published 11th September 2026 · Last updated 11th September 2026

Quick answer: AMpredict publishes its losing predictions alongside its wins because transparent prediction accuracy is the only kind that can be trusted. Football's irreducible randomness guarantees that roughly 10-15% of even High Confidence picks will lose, so a record showing only wins is edited, not accurate. Publishing losses lets subscribers verify the real 89% High Confidence hit rate rather than take it on faith, judge performance across a full sample rather than a highlight reel, and set realistic expectations before they subscribe. A tracked record that includes losses is the single clearest signal separating a legitimate prediction service from a scam.

Most prediction services show you a wall of winning slips.

Green tick after green tick. Screenshot after screenshot. Never a loss in sight. It's designed to look like proof, and it works, because a flawless record feels reassuring. It's also a lie by omission. No prediction service on earth wins every time, because football won't allow it. So a record that shows only wins isn't showing you how good the service is. It's showing you what the service chose to let you see.

AMpredict does the opposite. We publish our losses next to our wins, on purpose, permanently. This is why that policy exists, and why it matters more for your trust than any accuracy figure we could quote.

Why does AMpredict publish its losing predictions?

AMpredict publishes its losing predictions because an accuracy claim means nothing without the losses included. A hit rate is a fraction: wins over total picks. Hide the losses and you've removed the denominator, which lets any service claim any number it likes. Showing every result, win or lose, is the only way a stated accuracy figure can actually be verified.

The principle is simple. If we tell you our High Confidence picks land at 89%, that number is only meaningful if you can see the other 11% that didn't. Without the losses on record, "89%" is just a word we typed.

This is the same reasoning behind how prediction confidence is calculated in the first place. A confidence figure is a claim about how often something lands over a large sample. You can't validate that claim against a record with the failures deleted. Publishing losses isn't modesty. It's what makes the accuracy figure a fact instead of a slogan.

Why can't any football service win every time?

No football service can win every time because football contains irreducible randomness that no analysis can remove. Roughly 8-12% of match outcomes turn on unpredictable single events, red cards, late penalties, goalkeeper errors, deflected winners. Even a perfectly analysed pick carries that residual chance of losing, which is why the realistic accuracy ceiling sits around 88-90%, never 100%.

This isn't a weakness in any particular service. It's the nature of the sport.

We cover the full argument in our piece on why sure wins are mathematically impossible, but the short version is that a game decided by moments of chaos cannot be predicted with certainty. A striker slips. A linesman's flag stays down. A shot deflects in off a defender. None of it was in any dataset beforehand.

So losses aren't a sign that a service is bad. They're proof that it's real. A record with zero losses isn't describing exceptional skill; it's describing an edited highlight reel from a service pretending football works in a way it doesn't.

What does hiding losses let a service get away with?

Hiding losses lets a service inflate its accuracy without limit, because the number can't be checked. When only wins are visible, a service claiming "95% accuracy" and a service actually hitting 55% look identical to a new subscriber. The missing losses are exactly the evidence that would tell them apart.

Concealing losses enables three specific deceptions.

The first deception is the inflated headline number, the kind we break down in our guide to spotting fake prediction guarantees. The second is hiding the normal losing runs every genuine service experiences, so the service looks unnaturally steady. The third is showing a handful of cherry-picked winners as if a small, curated sample represents long-term performance.

All three collapse the moment losses go on the public record. That's precisely why dishonest services never publish them, and precisely why we do.

How does publishing losses help subscribers?

Publishing losses helps subscribers in 3 concrete ways: it lets them verify the real accuracy figure, judge performance across a full sample rather than a highlight reel, and set realistic expectations before they subscribe. Each one protects the subscriber, which is the entire point.

Verification. When every result is on record, our 89% High Confidence figure is something you can check, not something you have to believe. You can count the wins and losses yourself.

Full-sample judgement. A complete record lets you see how the picks perform through good runs and bad, across weeks and months, rather than through a handful of flattering screenshots. That's the only fair way to assess a prediction service.

Realistic expectations. Seeing the losses upfront means you understand what you're subscribing to: a strong, honest hit rate with real losses included, not a fantasy of guaranteed wins. Subscribers who know the real picture make better staking decisions and don't panic during the normal losing runs that every service has.

An informed subscriber is a better-served subscriber. Hiding losses keeps people in the dark; publishing them puts the real picture in their hands.

How does this fit AMpredict's wider methodology?

Publishing losses is the reporting half of the same honesty that runs through AMpredict's whole prediction process. The three-layer methodology of mathematical modelling, AI pattern recognition, and human expert review produces the predictions; transparent results reporting is how we account for them afterwards. One without the other would be incomplete.

The connection matters. A rigorous method that gets reported dishonestly is worthless, because you'd have no way to tell the rigour was real. And honest reporting of a weak method just transparently documents failure. The two only work together: a genuine edge, honestly tracked.

That's why the accuracy figure we publish sits at 89% on High Confidence picks, below the 90% mathematical ceiling, with losses included. The number is deliberately honest rather than impressive, because an inflated figure would undo the entire point of the methodology behind it. You can see how the different confidence tiers and categories are structured on the VIP packages page.

How should you use a service's loss record to judge it?

You should treat a visible, permanent loss record as a minimum requirement before trusting any prediction service, and treat its absence as a decisive warning. The check takes seconds: look for losses in the public record. If there are none, the record has been edited, and nothing else the service claims can be trusted.

Three quick things to look for when you assess any service.

First, are losses visible at all? A service showing only wins has failed the most basic transparency test. Second, does the record go back far enough to include losing runs? A suspiciously short history has usually had its bad stretches deleted. Third, does the stated accuracy sit within the realistic 88-90% ceiling? A figure above that, paired with no visible losses, is a fabrication by definition.

A service that publishes its losses is showing you it has nothing to hide. That single signal tells you more about its honesty than any headline accuracy number ever could.

The Bottom Line

AMpredict publishes its losses because transparent prediction accuracy is the only accuracy worth anything. Football guarantees that even elite High Confidence picks lose sometimes, so a record showing only wins is edited, not accurate. Publishing every result, win or lose, is what lets subscribers verify our real 89% hit rate, judge performance across a full sample, and know exactly what they're signing up for.

Hiding losses is how dishonest services inflate their numbers and manufacture the illusion of a flawless run. Showing losses is how an honest service proves its figure is real. It's the clearest single line between a legitimate prediction service and a scam, and we've chosen which side of it to stand on.

When you assess any prediction service, look for the losses first. If you can't find any, you've already learned the most important thing about it.

Want a prediction service that shows you the full picture? See AMpredict membership options and get honestly-tracked predictions, wins and losses both, built on the full three-layer methodology before your next weekend kickoff.