How we check football prediction sites

Every site goes through the same 10 checks. Each one is something you can confirm yourself in a few minutes, without paying or creating an account.

On this page

Why a checklist and not a score

Most prediction site rankings give each site a number out of 10 or 100. The number hides the reason. Two sites can both score 7 out of 10 for opposite reasons: one keeps a full archive but hides the odds, the other shows odds but deletes old tips.

A checklist shows the reason. When a site misses check 3, you know the odds are missing and you know what to ask before you follow it. Every review on this site lists the checks a site passed and the ones it failed.

The 10 criteria

  1. C1
    Public results archive

    Past predictions stay visible after the match, including the losing ones.

  2. C2
    Tips published before kick-off

    Each tip carries a publication time or appears on the site before the match starts.

  3. C3
    Odds recorded with each tip

    The site shows the odds a tip was given at, so profit can be worked out.

  4. C4
    Long-term statistics

    Win rate, yield or profit is shown over a defined period, not only recent wins.

  5. C5
    Sample size shown

    The number of tips behind each statistic is visible.

  6. C6
    Method explained

    The site says how predictions are produced or how tipsters are ranked.

  7. C7
    Free access

    Some predictions or statistics can be read without paying.

  8. C8
    Clear paid terms

    Anything paid shows what it includes before payment and can be cancelled.

  9. C9
    No guaranteed-win claims

    No "100% sure", "fixed match" or "sure win" promises.

  10. C10
    Identifiable operator

    The site names who runs it and gives a working contact.

What each check looks for

C1. Public results archive

We look for a page where past predictions stay up after the match, including the losers. A site that only shows “yesterday’s winners” or a screenshot gallery fails. Deleting losing tips is the most common way prediction sites make their record look better than it is.

C2. Tips published before kick-off

A tip only counts if it existed before the match started. We look for a publication time on each tip, or we watch tips appear on the site ahead of matches. Tips posted or edited after kick-off fail.

C3. Odds recorded with each tip

Without the odds, a winning tip tells you nothing about profit. A 1.20 favourite and a 4.00 outsider both count as “one win”. We check that the odds at the time of the tip are shown next to it.

C4. Long-term statistics

We look for win rate, yield (profit per unit staked) or total profit over a stated period, such as the last 30 days or the whole season. A “last 10 tips” streak does not count.

C5. Sample size shown

A 40% yield over 15 tips can be luck. The same yield over 600 tips is far harder to explain away. We check that the number of tips sits next to each statistic.

C6. Method explained

For a model-based site, we look for a description of what the model uses: form, goals, expected goals, odds. For a tipster platform, we look for how tipsters are ranked and over what period. “Our experts analyse every match” is not a method.

C7. Free access

We check whether a visitor can read some predictions, rankings or statistics without paying or registering. Free access lets you test a site before you trust it with money.

C8. Clear paid terms

If a site sells a subscription, VIP group or premium tips, we check that it says what is included and how to cancel before you pay. Payment only through a private message or a bank transfer to a person fails.

C9. No guaranteed-win claims

Any promise of “100% sure”, “sure odds”, “fixed matches” or “no loss” fails the site, whatever else it shows.

C10. Identifiable operator

We look for a company name or a named team, terms of use and a working contact. An anonymous site with only a Telegram handle fails.

Verdict thresholds

Checks passedVerdictWhat it means
8 to 10PASSEDYou can see how the site has performed before you rely on it.
5 to 7CAUTIONPart of the record is visible, but something important, such as odds or sample size, is missing.
0 to 4FAILEDYou cannot verify the record, or the site makes guaranteed-win claims.

How the order inside a list is set

Sites with the same verdict are ordered by editorial assessment. We look at how much of the record a visitor can check for free, how many leagues and markets are covered, whether statistics are broken down by period or by tipster, and how easy the site is to use on a phone. The order is our judgement, not a measured accuracy figure.

Run the checks yourself in five minutes

  1. Open the site’s results or history page. If there is none, stop: check 1 fails.
  2. Pick a match from last week. Was the tip there before kick-off, and is the losing result still shown?
  3. Look for the odds next to each tip and for a profit or yield figure.
  4. Find the number of tips behind that figure.
  5. Search the site for “100% sure”, “fixed” and “VIP guaranteed”.
  6. Scroll to the footer and look for who runs the site and how to contact them.

If a site passes all six steps, it is worth a closer look. If it fails steps 1 or 5, move on.

How often we re-check

We re-run the checks when a site changes how it shows results, rankings or prices, and at least every three months. When a verdict changes, the review says what changed.

Corrections

If a check is wrong or out of date, write to us through the Contact page with a link that shows the correct information. We fix factual errors and say what we changed.

FAQ

What makes a football prediction site trustworthy?
A trustworthy prediction site keeps its losing tips visible, records the odds of every tip and shows long-term results with the number of tips behind them. If you can only see recent wins, you cannot judge the site.
Why does a guaranteed-win claim fail a site automatically?
No one can guarantee a football result. A site that sells “100% sure” tips or “fixed matches” is either misleading its visitors or describing match fixing, which is a crime in most countries. We mark these sites FAILED whatever else they show.
Is a high win rate enough to trust a tipster?
No. A tipster who backs short-priced favourites can win 70% of bets and still lose money. Check the odds and the profit or yield as well, and how many tips the figures are based on.
How many tips do I need to judge a tipster?
A few dozen tips tell you very little. Results start to mean something after a few hundred tips, and more at high odds. A site that shows the sample size next to each statistic passes our check 5.
Do you measure the accuracy of each site?
No. We check whether a site lets you see and verify its own record. We do not publish accuracy or profit figures that we have not measured ourselves, and the order inside a list is an editorial assessment.