Our Methodology
Transparent AI analysis. Real verified results. No cherry-picking.
Hybrid: statistics + AI
Every metric has a source. Every formula is documented. Every outcome is verified automatically. The goal is not to pick winners blindly — it's to understand why and how.
Data-first
We start from raw facts: fixture data, xG, odds, squad news. No assumptions before the data says so.
Quantitative core
Probability is the fair probability of the market (median of bookmaker prices, margin removed); Poisson/xG is supporting context. Math runs before the AI opinion.
AI synthesis
AI contextualises numbers — current form, squad quality, venue dynamics — and flags conflicts between math and context.
Everything is verifiable. See the full track record →
The 5-Step Prediction Pipeline
Each prediction goes through the same deterministic sequence. No shortcuts, no skipped steps.
Data Collection
No AI · facts onlyDeterministic data ingestion — no AI involved at this stage. Every number is sourced from external APIs.
- API-Football: fixtures, team stats, H2H, current form, injuries, lineups (700+ leagues)
- Understat xG: expected goals — daily update + our own historical xG base
- Live odds from the bookmakers in our odds feed (the median of their prices is the basis of our probability)
Analysis Generation
AI — expert layerAI generates qualitative match analysis from the collected data: attack vs defence assessment, current form dynamics, H2H on specific venue, squad optimality.
- Attack vs. defence quality comparison for both teams
- Current form dynamics (last 5 matches weighted by recency)
- Head-to-head record on this specific venue
- Squad optimality — key player availability factor
Quantitative + AI Betting Analysis
Math + AITwo parallel tracks that are then compared for consistency.
- 3A — Math: fair probabilities from the market (median of bookmaker prices, margin removed); Poisson/xG shown as supporting context
- 3B — AI: synthesises all prior context and 3A results; can override or confirm the math
- Conflict report: if 3A and 3B diverge significantly, the confidence level is capped at Medium
Final Review
AI — quality gateConsistency check across all prior steps. AI assigns the final verdict, catching contradictions or anomalies; the confidence level is then computed by code.
- Cross-checks Steps 2 and 3A/3B for contradictions
- Confidence level (High / Medium / Low) is computed by code from the checks
Publication
Structured outputHuman-readable analysis rendered on the site plus structured JSON consumed by the frontend for live odds overlay, confidence level, and auto-verification.
- Natural-language analysis visible on prediction page
- Structured JSON: market, prediction value, confidence level
- Auto-verification triggers when match result is available
How confidence is calculated
Confidence is a level (High / Medium / Low) computed by code from the market data and the consistency checks, not a percentage and not an AI opinion.
High
All checks pass: market data is available, the models do not contradict each other, the pick is not priced at an unusually large gap to the market, and for 1X2 it is the market favourite.
Medium
Market data is available, but at least one check fails: the models conflict with each other, the price gap is unusually large, or the pick is not the favourite on 1X2.
Low
Not enough market data: no bookmaker anchor, no real odds, or an unsupported market.
The confidence level is shown on the prediction card but is not a condition for the official pick. We do not claim that a higher level wins more often; the official track record below shows the real results, with the sample size next to every number. Before 2 October 2026 the model used earlier rules (versions v1–v2): a blend of our own Poisson/xG probabilities with the market, an expected-value threshold and a weighted confidence score. Those results stay in the archive with their original numbers and are labelled as such; they are not recalculated.
Profit comes from value, not win rate
We only publish a value bet when the best available price is 5–30% above the fair price of the market and the fair probability is at least 50%.
Value Formula
Fair probability = median of bookmaker prices, margin removed
Fair price = 1 / fair probability
Value % = (best price / fair price − 1) × 100
Published only when the best price is 5–30% above the fair price and fair probability ≥ 50%
Fair probability
Our probability is the fair probability of the market: the median of the prices of all bookmakers in our feed, with the margin removed. The Poisson / xG model is kept as supporting context in the analysis text. It no longer sets the probability. We do not publish stake-size recommendations, and there is no Kelly calculator.
Markets scanned
Updated on a rolling cycle — results verified each time a match concludes.
Example: 1X2 market (illustrative numbers)
The median of bookmaker prices, margin removed, gives a fair probability of 60% for a home win — a fair price of 1.67. One bookmaker offers 1.80.
Value = (1.80 / 1.67 − 1) × 100 ≈ +8% — inside the 5–30% window, and the fair probability is above 50%, so this qualifies.
Because we only publish picks with a fair probability of at least 50%, the win rate of official picks is higher than for longshot-style tips, and the prices are shorter. Win rate alone still says little without the odds at which the bets were placed, so we always show ROI next to it, and ROI only becomes meaningful with a large enough sample.
How We Find Value Bets
Transparent, math-driven, no cherry-picking. Every value bet passes the same filter — here's exactly what that means.
Fair probability
Our probability is the fair probability of the market: the median of the prices of all bookmakers in our feed, with the margin removed.
Odds scan
We scan current bookmaker odds for each analysed market and take the best available price. Excluded bookmakers are filtered per region and licensing requirements so only legitimate, accessible prices are considered.
Value calculation
A bet has value when the best price is higher than the fair price of the market:
Formula
Fair probability = median of bookmaker prices, margin removed
Fair price = 1 / fair probability
Value % = (best price / fair price − 1) × 100
Strict filters (honesty gate)
Every opportunity passes a hard filter before appearing on the page. We drop anything that doesn't meet all criteria — no cherry-picking:
- Best price is 5–30% above the fair price
- Fair probability is at least 50%
- Every opportunity expires at kickoff
No stake sizing
We do not publish stake-size recommendations, and there is no Kelly calculator. We report our track record with flat 1-unit stakes.
What you see on the odds page
Probability ring
Fair probability of the outcome — the market median without margin, not a hype score or editorial rating.
Value %
How far the best price sits above the fair price of the market.
Confidence level
High / Medium / Low, computed by code from the checks above. We don't publish a stake-size recommendation.
Value betting is long-term and probabilistic
Mathematically sound value bets still lose. The edge plays out over hundreds of bets, not individual outcomes. No guarantee of profit per prediction. Bet responsibly and size your stakes by your own bankroll rules; we do not recommend stake sizes. 18+.
See live value betsHow the Track Record is calculated
Not every prediction we publish is counted as a staked bet. Here is exactly which picks qualify, and the formula used to turn them into ROI.
What counts as an "Official Tip"
The headline ROI and win rate above — and everywhere else on this site — only count picks that pass all of these conditions (the same rule since 2 October 2026):
Primary pick
The main recommendation for the match — not an alternative market or a side note.
Markets and real price
Only 1X2, Over/Under 2.5 and BTTS, with a real bookmaker price and market support from the feed.
Fair probability ≥ 50%
We only stake on outcomes the market itself rates as more likely than not.
Best price 5–30% above fair
The best available price must sit 5–30% above the fair price of the market.
The official pick of the day follows the same rule: the primary pick must have a fair probability of at least 50% and the best price 5–30% above the fair price. The confidence level is shown on the prediction card but is not part of the selection. Only these picks count towards the official track record, with flat 1-unit stakes. If nothing qualifies, we show no pick. Other picks are still published as analysis on the site, but they are not counted as staked bets in the official ROI.
Before 2 October 2026 the model used earlier rules (versions v1–v2): a blend of our own Poisson/xG probabilities with the market, an expected-value threshold and a weighted confidence score. Those results stay in the archive with their original numbers and are labelled as such; they are not recalculated.
ROI Formula
Stake = 1 unit, flat, on every official tip — no progressive or variable staking.
Win profit = (decimal odds − 1) × 1 unit
Loss = −1 unit
Push (void/cancelled) = 0, excluded from win rate
ROI = (Σ profit / Σ staked) × 100
Win rate = wins / (wins + losses) × 100
Flat 1-unit staking is used here so the headline ROI reflects pure picking accuracy, not stake-sizing skill — we don't publish stake-size recommendations. ROI is only meaningful on a large enough sample; small samples are flagged.
We show the losing markets too
Per-market performance is published openly, including markets that are currently unprofitable — we do not hide or exclude a market just because its ROI is negative. The same qualifying rules (primary, fair probability ≥ 50%, best price 5–30% above fair) apply identically across every market; nothing is curated after the fact to flatter the headline number.
Verified Track Record
Live per-market breakdown from the same data powering our home page — official tips only, as defined above. No cherry-picking; every completed bet is counted.
Honest about what we are and aren't
Automated verification
Results are verified automatically — every hour we check for completed matches and update win/loss/push status. The full daily cycle runs once per day. No manual selection.
No guarantees
Football is fundamentally unpredictable. Even mathematically sound value bets lose. We provide a transparent edge over time — not a guarantee per prediction.
System evolving
Models improve as we collect more data. Past ROI does not guarantee future ROI. We document methodology changes and do not rewrite history.
Bet responsibly
Treat each tip as one input, not a copy-paste decision. Never bet more than you can afford to lose. We don't recommend stake sizes — size your bets according to your own bankroll rules.
Frequently asked questions
Now You Know How It Works — See the Results.
Browse today's AI-powered predictions built on the exact methodology above. Every tip backed by a transparent, verified track record.