The Difference Between Betting and Investing

Most bettors pick winners. Professional bettors find edge. The difference is mathematics. A bettor who picks the right team 55% of the time but bets at the wrong odds will lose money. A bettor who picks the right team 50% of the time but consistently finds odds that underestimate their true probability will profit long-term. This course teaches you the statistical framework that separates the two — expected value, predictive models, and the data tools that give you a measurable, repeatable edge over bookmakers.

Why Statistics Beat Intuition:

Human beings are wired to see patterns that aren't there, overweight recent events, and ignore base rates. Bookmakers exploit these biases every single day. Statistical models don't have emotions, don't chase losses, and don't get overexcited after a big win. They calculate probability objectively — and over thousands of bets, that objectivity always wins.

What You'll Master:

01 Calculate expected value (EV) and understand why it is the foundation of profitable betting
02 Use Expected Goals (xG) to estimate true scoring probability and find value
03 Apply the Poisson distribution model to predict goals, correct scores, and over/under outcomes
04 Measure your edge using closing line value — the professional standard for performance tracking
05 Compare sharp bookmaker odds to soft bookmaker odds to identify systematic value
06 Build a simple betting model in a spreadsheet that generates selections automatically
07 Apply Kelly Criterion bet sizing to maximise long-term growth while controlling risk

Expected value is the single most important concept in all of betting. It tells you whether a bet will make or lose money over the long run — regardless of what happens on any individual match. Master this concept, and everything else in this course falls into place.

What Is Expected Value?

Expected value (EV) is a mathematical calculation that tells you the average outcome of a bet if you placed it hundreds of times. A positive EV bet (+EV) will make you money long-term. A negative EV bet (-EV) will lose you money long-term. The result of any single bet is irrelevant — EV is about the long run.

Expected Value Formula:

EV = (Probability of Winning × Amount Won) − (Probability of Losing × Amount Lost)

Example:
You back a team at odds of 2.50 (implied probability: 40%)
Your true probability estimate: 55%
Stake: £10

Amount won if correct: £10 × (2.50 − 1) = £15
Amount lost if wrong: £10

EV = (0.55 × £15) − (0.45 × £10)
EV = £8.25 − £4.50
EV = +£3.75

This is a +EV bet. Place it every time it appears.

Implied Probability vs. True Probability

Every set of odds implies a probability. The bookmaker is telling you how likely they think an outcome is. Your job is to estimate the true probability — and when the two don't match, you have found value.

Converting Odds to Implied Probability:

Implied Probability (%) = (1 ÷ Decimal Odds) × 100

Odds 1.50 → 1 ÷ 1.50 = 66.7% implied probability
Odds 2.00 → 1 ÷ 2.00 = 50.0% implied probability
Odds 2.50 → 1 ÷ 2.50 = 40.0% implied probability
Odds 3.00 → 1 ÷ 3.00 = 33.3% implied probability
Odds 4.00 → 1 ÷ 4.00 = 25.0% implied probability

Note: The sum of implied probabilities across all outcomes in a market exceeds 100%. The excess is the bookmaker's margin (overround).

The Overround: How Bookmakers Guarantee Profit

In every market, the bookmaker sets odds so that the total implied probability across all outcomes adds up to more than 100%. This excess is called the overround — and it is how bookmakers build their profit margin into every single bet.

Overround Calculation:

Match Result Market:
Home Win: 1.80 → implied probability: 55.6%
Draw: 3.20 → implied probability: 31.3%
Away Win: 4.50 → implied probability: 22.2%

Total implied probability: 55.6% + 31.3% + 22.2% = 109.1%

Overround = 109.1% − 100% = 9.1%

This means the bookmaker has built a 9.1% margin into this market. You must find bets where your true probability exceeds the implied probability by enough to overcome this margin.

Why You Can Win Individual Bets and Still Lose Money

This is the concept that separates casual bettors from professionals. Winning bets feels good. But if every bet you place has negative expected value, you will lose money over time — no matter how many individual winners you pick.

The Losing Winner Trap:
A bettor backs teams at odds of 1.50 and wins 60% of the time.
Implied probability at 1.50 = 66.7%
True win rate = 60%

Every single bet is -EV. Over 100 bets at £10 each:
Won: 60 bets × £5 profit = +£300
Lost: 40 bets × £10 loss = −£400

Net result: −£100. The bettor won 60% of their bets and still lost money.

How to Estimate True Probability

Estimating true probability is the core skill of statistical betting. There is no single perfect method — but combining multiple data sources gives you the most accurate estimate.

Sources for Probability Estimation:

  • Expected Goals (xG): The most powerful single metric for estimating goal-scoring probability. Covered in detail in Lesson 2.
  • Sharp bookmaker odds: Exchanges and sharp books price markets most efficiently. Their implied probability is often very close to true probability.
  • Historical base rates: How often does this type of outcome actually happen? Home teams win roughly 45% of Premier League matches — this is your starting point.
  • Team form (last 6-10 matches): Recent form provides context, but should not override statistical baselines.
  • Team news: Injuries and suspensions can shift probability by 5-10 percentage points for key players.

Practical EV: A Quick Decision Framework

You don't need to run a full calculation for every bet. Here is a quick mental framework to assess whether a bet is likely +EV or -EV.

Quick EV Check:

Step 1: Convert the bookmaker odds to implied probability
Step 2: Estimate the true probability using your research
Step 3: If true probability > implied probability → likely +EV ✓
Step 4: If true probability < implied probability → likely -EV ✗

Rule of thumb: You need at least a 3-5% edge (true probability exceeding implied probability by 3-5 percentage points) to account for estimation error and bookmaker margin. A 2% edge is borderline. A 1% edge is noise.

Key Takeaways:

  • Expected value determines whether a bet makes or loses money long-term — not the result of any single match
  • Implied probability is calculated by dividing 1 by the decimal odds
  • The overround is the bookmaker's built-in profit margin — it exists in every market
  • You can win 60% of your bets and still lose money if every bet is -EV
  • True probability must exceed implied probability for a bet to be +EV
  • Aim for at least a 3-5% edge between your probability estimate and the bookmaker's implied probability
  • EV is a long-term concept — short-term results will vary, but +EV bets always profit over time

Expected Goals (xG) is the single most powerful statistical tool available to football bettors. It does what raw goal counts cannot — it measures the quality of chances, not just the results. This lesson teaches you exactly what xG is, how it's calculated, and how to use it to find value across multiple betting markets.

What Is xG and Why Does It Matter?

xG assigns a probability (0 to 1) to every shot taken in a match, based on how likely that specific shot was to result in a goal. A shot from the centre of the box with no defenders might have an xG of 0.35. A shot from 30 yards out with two defenders might have an xG of 0.02. Adding up all the xG values for a team gives you their Expected Goals for the match — a measure of how many goals they "should" have scored based on the quality of their chances.

Why xG Beats Raw Goals:

Team A scored 3 goals. Team B scored 0 goals.
Raw interpretation: Team A dominated completely.

But xG tells a different story:
Team A xG: 1.2 (they scored 3 but only "deserved" 1.2)
Team B xG: 2.1 (they created chances worth 2.1 goals but scored 0)

Reality: Team B actually had the better chances. Team A got lucky.
Prediction: Team B is likely to revert toward their xG in future matches.
This is where the betting edge lives.

How xG Is Calculated

xG models use machine learning trained on hundreds of thousands of historical shots. Each shot is evaluated based on several factors that determine how likely it was to go in.

Key Factors in xG Calculation:

Shot location: Distance from goal, angle to the posts
Shot type: Foot (left/right), header, or other
Assist type: Open play, cross, set piece, through ball
Defenders in path: Number of defenders between shooter and goal
Goalkeeper position: How well-positioned the keeper is

More advanced models also factor in: time pressure, game situation, and player-specific finishing ability (post-shot xG or xGOT).

Reading xG: What the Numbers Mean

Understanding xG values in context is essential. Not all xG numbers mean the same thing.

xG Reference Guide:

Team xG per match:
0.8 or below = Weak attacking output
0.9 - 1.2 = Average
1.3 - 1.6 = Strong attack
1.7+ = Elite attacking output

Goals vs. xG (over 10+ matches):
Goals significantly above xG = Team is overperforming (likely to regress)
Goals significantly below xG = Team is underperforming (likely to improve)
Goals close to xG = Team is performing at their true level

Using xG to Find Value Bets

xG gives you an objective measure of a team's true attacking and defensive quality. When bookmaker odds don't reflect xG reality, you've found value.

xG Value Betting Process:

Step 1: Find Team A's home xG per game: 1.5
Step 2: Find Team B's away xGA (goals conceded) per game: 1.3
Step 3: Estimate likely goals for Team A in this fixture: ~1.4
Step 4: Use this to estimate probability of Over 2.5 goals, BTTS, match result, etc.
Step 5: Compare your probability estimate to the bookmaker's implied probability
Step 6: If your estimate is higher → +EV bet exists ✓

xG and Regression to the Mean

One of the most profitable applications of xG is identifying teams that are due for a correction. When a team's actual goals diverge significantly from their xG over 8-10 matches, the gap almost always closes. Teams scoring well above their xG will slow down. Teams scoring well below will improve.

Regression Example:
A team has scored 12 goals in 8 matches but their xG total is only 7.2.
They are scoring at 1.67x their expected rate.

Historically, teams performing this far above xG see their scoring rate drop back toward xG within 4-6 matches.

Betting: Back Under goals or BTTS No in their upcoming matches. The bookmaker odds will still reflect their high goal count — but xG tells you the truth.

Where to Find xG Data

Several free and paid platforms provide reliable xG data for football matches.

Recommended xG Sources:

  • Understat.com: Free xG data for all top European leagues. Match-level and season-level breakdowns.
  • FBref.com: Free advanced stats including xG, xA, and shot-level data for major leagues.
  • WhoScored.com: Match ratings and xG data with good visual presentation.
  • FootyStats.org: Comprehensive stats including xG trends, home/away splits, and predictive rankings.

Key Takeaways:

  • xG measures the quality of chances created, not just results — it is the most predictive single stat in football
  • Teams scoring well above their xG are overperforming and likely to regress
  • Teams scoring well below their xG are underperforming and likely to improve
  • Use xG to estimate true goal-scoring probability, then compare to bookmaker implied probability
  • xG is most useful over 8-10+ matches — single-match xG is less reliable
  • Always split xG into home and away — the same team can have very different xG in each context
  • Regression to the mean is one of the most reliable patterns in football — xG identifies it before the market does

The Poisson distribution is the mathematical model most commonly used to predict football scores. It is the foundation of every serious betting model and the reason sharp bookmakers price their odds the way they do.

Coming Soon

What the Poisson distribution is, how to calculate goal probabilities for any scoreline, building a Poisson model in a spreadsheet, predicting correct scores, over/under, and BTTS outcomes, and the model's strengths and limitations.

Win/loss records are unreliable over small samples. Closing line value is the professional standard for measuring whether you actually have an edge — regardless of results.

Coming Soon

What closing line value is, how to calculate it, why it predicts long-term profitability better than ROI, tracking CLV in a spreadsheet, and using it to calibrate your betting strategy.

Not all bookmakers are equal. Sharp bookmakers price markets efficiently. Soft bookmakers lag behind. The gap between them is where systematic value lives.

Coming Soon

The difference between sharp and soft bookmakers, which exchanges and books to use as reference points, how to compare odds systematically, line movement analysis, and building a value-finding workflow.

A betting model automates the analysis — removing emotion and ensuring you evaluate every fixture consistently. This lesson walks you through building a functional model using nothing but a spreadsheet.

Coming Soon

What a betting model does, the data you need to feed it, building the model step by step in Excel/Google Sheets, backtesting your model on historical data, and interpreting output to generate selections.

You've found a +EV bet. Now the question is: how much should you stake? Too little and you underutilise your edge. Too much and a short losing streak can devastate your bankroll. The Kelly Criterion is the mathematically optimal answer.

Coming Soon

The Kelly Criterion formula, full Kelly vs. fractional Kelly, how edge size determines stake size, practical staking examples, and why Kelly maximises long-term bankroll growth while controlling ruin risk.

📈 Core Lessons Available

The first 2 comprehensive lessons are available now, covering expected value fundamentals and mastering xG. Remaining lessons releasing weekly!

Join Community