Most bettors spend 5 minutes on research before a game and then watch it in isolation. This guide shows you how to use KiqIQ from 48 hours before kickoff to the final whistle — with specific AI prompts, tools, and actions at every stage of the matchday. We follow Liverpool vs Chelsea throughout.
⚠️ Responsible gambling note
This guide demonstrates how to use KiqIQ for football analysis and information purposes. Always bet within your means. If you are concerned about your gambling, visit BeGambleAware.org.
Matchday timeline
48–72 hrs before
We will follow Liverpool vs Chelsea throughout this guide.
Actions at this stage
Open the Predictions Hub and find Liverpool vs Chelsea.
Note Liverpool's home xG average (1.9) and Chelsea's away xG average (1.2).
Run both through the Poisson Calculator with home/away splits.
Check the Margin Calculator for Over/Under 2.5 goals and BTTS markets.
AI prompts for this stage
"Liverpool home xG is 1.9 this season, Chelsea away xG is 1.2. Give me the key context I need before modelling this fixture — recent form trends, pressing intensity from both sides, any fixture congestion or mid-week games that might affect intensity."
"Is Liverpool vs Chelsea typically a high-xG game historically, or does the rivalry produce more defensive, compact play that suppresses expected goals relative to both teams' season averages?"
Stage outcome
After this stage: your Poisson model is built, you have edge identified on Over 2.5 goals at 1.80 (58% model vs 55.6% implied), and you have qualitative context from the AI.
24 hrs before
Actions at this stage
Check current odds across 2–3 bookmakers for your target markets.
Run odds through Implied Probability Calculator to confirm margin hasn't increased.
If edge still present, run Kelly Calculator to get pre-lineup stake recommendation.
Note opening odds for CLV tracking — record in your bet log even before placing.
AI prompts for this stage
"Over 2.5 goals in Liverpool vs Chelsea opened at 1.85 yesterday and is now at 1.78. What does this line movement suggest — is sharp money going under, or is this public money pushing the line?"
"I want to bet Over 2.5 goals in Liverpool vs Chelsea. Should I bet now at 1.80 or wait until lineup confirmation tomorrow, and what injury news could significantly affect this market?"
Stage outcome
Decision: bet now at 1.80 if you have strong edge (58% vs 55.6%) or wait for lineup confirmation. Quarter Kelly on a £500 bank at these odds ≈ £6.50.
Lineup day (2–3 hrs before)
Actions at this stage
Lineups are confirmed. Check for any surprises — rested attackers, unexpected injuries.
If a key player is missing, re-run the Poisson model with adjusted xG inputs.
Final AI sense-check before placing your bet.
Place bet and log it in the Bet Tracker with opening odds.
AI prompts for this stage
"Liverpool have rested Salah today — he's on the bench. My Poisson model was based on their full-strength 1.9 home xG average. How much should I adjust Liverpool's expected goals downward without Salah, and does this affect my Over 2.5 edge?"
"Give me a one-paragraph final assessment for Liverpool vs Chelsea: which team has the underlying data edge, what is the most likely match narrative based on xG profiles and pressing intensity, and what is the key risk to my Over 2.5 bet?"
"If this game is 0-0 at halftime, what should I look for in xG data to determine whether Over 2.5 is still likely in the second half? At what xG split at 45 minutes would you consider the over still likely?"
Stage outcome
Bet placed: Over 2.5 goals at 1.80. Logged in tracker: £6.50 stake. Opening odds recorded for CLV. You are ready to watch with data context.
Kickoff — first 15 minutes
Actions at this stage
Watch the early pressing intensity — does it match the PPDA profile you researched?
After 10–15 minutes, if the game is open with early shots, use the AI to calibrate.
If a goal goes in early, ask about in-play implications for your bet.
AI prompts for this stage
"Liverpool are dominating territory in the first 10 minutes of vs Chelsea, pressing high and winning the ball in Chelsea's half. Does this match their seasonal PPDA profile, and does early high-press success typically sustain for 90 minutes or fade?"
"Liverpool just scored in the 12th minute — it's 1-0. My bet is Over 2.5 goals. Does an early goal in a Liverpool home game statistically increase or decrease the probability of 3+ total goals? Does it typically push Chelsea to open up or sit deeper?"
"It's 0-0 after 15 minutes but both teams are creating. Is this a normal pattern in high-xG Liverpool home games where goals tend to come in clusters, or is low-scoring often a risk when Chelsea play defensively at Anfield?"
Stage outcome
The AI gives you context during live action — not to prompt in-play bets, but to understand what's happening against your pre-match model.
Halftime
Actions at this stage
Find the first-half xG data (available from major stats sites during the game).
Compare first-half xG to your pre-match expectations.
Ask the AI to interpret what the half-time xG picture means for the second half.
If in-play betting, use the Poisson calculator with updated xG figures.
AI prompts for this stage
"Halftime Liverpool vs Chelsea: score is 1-0 Liverpool. First-half xG: Liverpool 1.1, Chelsea 0.7. I need Over 2.5. Given 1.8 total xG generated already, what does the Poisson model suggest for probability of 2 more goals in the second half?"
"It's 0-0 at halftime but the xG is Liverpool 1.4, Chelsea 0.4. Liverpool have dominated. Is this the type of game where second-half goals are likely, or does a 0-0 high-xG first half often produce more conservative second-half play from the trailing team?"
"I had Over 2.5 pre-match at 1.80. It's 1-0 to Liverpool at halftime. In-play Over 2.5 is now available at 2.20. My Poisson model, updated with first-half xG of 1.8, suggests roughly 55% chance of 2 more goals. Is 2.20 value for a second-half double in-play position?"
Stage outcome
Half-time is the most data-rich moment for recalibration. The AI turns raw xG into actionable second-half context.
60–75 min
Actions at this stage
Check if the tactical picture has changed — substitutions, injuries, formations.
If the scoreline creates a specific situation (e.g. 2-0 team chasing), ask for context.
If your bet is already looking good (2-1 and Over 2.5 covered), no action needed.
AI prompts for this stage
"Liverpool have brought on a defensive midfielder at 60 minutes, switching from 4-3-3 to 4-5-1 while leading 2-0. Does this typically compress their second-half xG generation, and does it change the probability of a 3rd goal for my Over 2.5 position?"
"It's 1-0 to Liverpool at 70 minutes with 60 minutes of total xG sitting at 2.2 (Liverpool 1.6, Chelsea 0.6). Roughly how much additional xG is typically generated in the final 20 minutes in games with this profile?"
"Chelsea are chasing the game at 2-0 down, 65 minutes. They've pushed players forward. Does this typically benefit my Over 2.5 position — will the open game increase total xG — or does Chelsea defending forward actually create counter-attack opportunities that increase goal probability?"
Stage outcome
The second half companion helps you understand whether what you're watching matches the data — and makes the post-match review more valuable by tracking your real-time reads.
Final whistle
Actions at this stage
Update the bet result in the Bet Tracker (won or lost).
Find the closing odds and calculate CLV.
Ask the AI for a post-match xG debrief — was the result consistent with the data?
Note anything your model missed for next time (injury impact, tactical change, etc.).
AI prompts for this stage
"Liverpool beat Chelsea 2-1. Final xG: Liverpool 2.1, Chelsea 1.4. Total xG was 3.5. My Over 2.5 bet won. Does the 3.5 xG suggest the result was fair, or did Liverpool underperform their xG while Chelsea got lucky to keep it close?"
"I modelled this as Liverpool 1.89 xG, Chelsea 0.82 xG. The actual xG came in at Liverpool 2.1, Chelsea 1.4. Chelsea significantly outperformed my model estimate. What might explain the gap — better Chelsea performance away, Liverpool rotation effect, or something else?"
"I bet Over 2.5 at 1.80 pre-match. The closing line was 1.72. My CLV is +4.65%. Over 50 bets my average CLV on Over/Under is now +2.8%. What does this suggest about my xG-based model's ability to beat the closing line on goals markets?"
Stage outcome
Every match now feeds your learning loop. The model gets sharper, the CLV tells you if it's working, and the Bet Tracker turns individual games into portfolio data.
Running this playbook consistently across a season builds data assets that compound.
A CLV record that tells the truth
After 50+ bets, your average CLV shows whether your Poisson model is actually beating the market. Positive CLV means your pre-match edge is real. This is faster and more reliable than ROI.
Market-specific insight
The tracker lets you segment CLV and ROI by market. You may find your Over/Under model is strong (+3% CLV) but your match result model breaks even. This tells you where to focus.
A model that improves
Each post-match debrief identifies where your xG inputs were inaccurate. Over 10–20 matches, you learn when to use recent form weighting, how to adjust for rotation, and which leagues your model handles best.
Better AI conversations
The more context you give the AI (your model outputs, your observations from watching the game, your CLV history), the more specific and useful the responses become.
The same matchday structure applies if you are using KiqIQ for FPL — with different prompts at each stage.
48–72 hrs before — Gameweek planning
"Which players in my FPL squad have the highest xG in the next 3 gameweeks based on fixture difficulty? Who should I captain, and is there a differential pick under 10% ownership with strong underlying data?"
Lineup day — Transfer and captain decision
"Salah has a tough fixture this week but Saka has an excellent matchup at home against a low-block side that concedes 1.5 xGA. Based on fixture difficulty and xG, who is the higher expected-points captain pick?"
During the game — Live points tracking
"My captain Salah is playing in a 0-0 game at 60 minutes. Liverpool's first-half xG was 1.2. Based on his individual xG in this game and his historical points return in matches with this profile, should I expect his captain points to come late?"
Post-match — Next gameweek prep
"My FPL captain got 12 points this week. He faces a tough double header over the next 2 gameweeks before a run of very good fixtures. Based on xG form and ownership trajectory, is this a hold, sell, or transfer-in moment for managers who don't have him?"
Predictions Hub
Pre-match xG data and team stats for any fixture.
Poisson Calculator
Model scoreline and market probabilities from xG.
Margin Calculator
Check bookmaker margin before placing any bet.
Implied Probability
Convert any odds to probability for edge comparison.
Kelly Calculator
Size your stakes based on your model edge.
Match Analysis Tool
Pre-match template with form, xG, and captain notes.
How to Research a Football Bet
The complete pre-match workflow — predictions hub to Kelly stake.
How to Model a Football Match
Build a Poisson probability model step-by-step with xG inputs.
Track Closing Line Value
How to use CLV to validate your betting edge over time.
Track Betting Performance
Log bets, measure CLV, and find where your edge leaks.
Post-Match xG Analysis
Read match data after full time to improve your next model.
Sample Questions & Prompts
80+ AI prompt templates across 12 categories.
