How to Predict Both Teams to Score: A Statistical Approach

How to Predict Both Teams to Score: A Statistical Approach

Learn to predict BTTS outcomes using xG data, team scoring records, and defensive statistics. A data-driven method for both teams to score predictions.

E

Editorial Team

Published 14 April 2026 · Updated 14 April 2026 · 4 min read

Share

Predicting Both Teams to Score

Both Teams to Score (BTTS) might seem like a coin flip — roughly 50% of Premier League matches end with both teams scoring. But with the right data, you can significantly improve your strike rate.

Here’s a statistical framework for predicting BTTS outcomes.

The Core Metrics You Need

1. Team Scoring Rate (Goals For per Match)

Simple but essential. A team that averages 2.0 goals per match is far more likely to score than one averaging 0.8.

2. Goals Conceded Rate

Equally important — you need the opponent to score too. A team with a leaky defence makes BTTS more likely.

3. Clean Sheet Percentage

The inverse of BTTS. If a team keeps clean sheets in 40% of matches, BTTS can only happen in 60% of their games (at most).

4. Failed to Score Percentage (FTS)

How often does a team fail to score? High FTS rate = lower BTTS probability.

Team celebrating together

5. xG For and xG Against

More predictive than actual goals. A team creating 1.8 xG per match will eventually score — even if they’ve been unlucky recently.

The BTTS Probability Formula

A simplified but effective approach:

P(BTTS) = P(Team A scores) × P(Team B scores)

Where:

  • P(Team A scores) = 1 – P(Team A fails to score)
  • P(Team B scores) = 1 – P(Team B fails to score)

Example Calculation

Arsenal (home) vs Newcastle (away)

Arsenal home scoring rate: Scored in 85% of home matches = P(Arsenal scores) = 0.85 Newcastle away scoring rate: Scored in 70% of away matches = P(Newcastle scores) = 0.70

P(BTTS) = 0.85 × 0.70 = 0.595 (59.5%)

If the bookmaker offers BTTS Yes at 1.75 (implied 57.1%), there’s slight value here.

Improving the Basic Model

Adjust for Opponent Strength

A team’s scoring record against top-6 teams differs from their record against bottom-half teams. Weight the data based on the quality of opponent.

Use xG Instead of Actual Goals

Replace actual scoring/conceding rates with xG-based rates. This smooths out short-term luck and gives more predictive estimates.

Factor in Home/Away Splits

Some teams score freely at home but struggle away (and vice versa). Always use venue-specific data.

Check Specific Head-to-Head Patterns

Certain fixtures have unusual BTTS patterns. Arsenal vs Tottenham, for instance, has different characteristics to Arsenal vs Fulham.

Football boots on grass

Consider Match Context

  • Teams needing points (relegation, title race) attack more aggressively → higher BTTS
  • Teams protecting leads sit deep → lower second-half BTTS
  • Teams already safe may rest key players → unpredictable scoring

Red Flags Against BTTS

Watch out for these signals that suggest BTTS No might be the better play:

  1. Defensive specialist team — Teams with 35%+ clean sheet rates
  2. Toothless attack — Teams failing to score in 40%+ of matches
  3. Low xG Against — Teams conceding very few quality chances
  4. Weather/pitch conditions — Heavy rain and poor pitches suppress goals
  5. Key attacker injured — Loss of a team’s primary goalscorer

BTTS Market Variations

Football on a stadium pitch

Beyond simple Yes/No, explore:

  • BTTS & Over 2.5 — Both teams score AND 3+ total goals
  • BTTS & Under 3.5 — Both teams score but not too many goals (think 1-1, 2-1 scorelines)
  • First Half BTTS — More volatile, but higher odds
  • BTTS in Both Halves — Very high odds, very difficult to predict

Tracking Your BTTS Bets

Keep a spreadsheet tracking:

  • Date, match, league
  • Your estimated BTTS probability
  • Bookmaker odds and implied probability
  • Result (BTTS Yes/No)
  • Profit/loss

After 100+ bets, you’ll see whether your model adds value or needs refinement.

18+ only. Gambling can be addictive. BeGambleAware.org

Continue Reading