As prediction is closely related to probability.
I have used Poisson distribution to estimate the probability of the goals being scored in a match.
As Poisson Distribution expresses the probability of a given number of events occurring in a fixed interval of time or space if these events occur with a known constant rate and independently of the time since the last event.
Known constant rate can be estimated as the average goals scored by each opponent team based on their home/away schedule and deducing their Attack and Defence Strength in comparison with the average League goals combined by all the teams.
Using the above Attack/Defence strength matrix we can estimate the probability of the team to score a specific number of goal(s).
By individually estimating the discrete prob using Poisson Distribution of the Home and Away team for different events (goals) , then further selecting the maximum Joint probability of the two independent events we can predict the final score/outcome.
Example-
Consider a following match between Real Madrid (Home) and Barcelona (Away).
Following Matrices can be used to estimate the final match score
Average Home Goals Scored (HAG) = Goals Scored / # Games played
Average Home Goals Conceded (HAC) = Goals Conceded / # Games played
Average Away Goals Scored (AAG) = Goals Scored / # Games played
Average Away Goals Scored (AAC) = Goals Conceded / # Games played
Avg League Home Goals (LHG) = Total Goals Scored / Total # Games
Avg League Away Goals (LAG) = Total Goals Scored / Total # Games
Home Attack Strength (HAS) = HAG / LHG
Home Defence Strength (HDS) = HAC / LAG
Away Attack Strength (AAS) = AAG / LAG
Away Defence Strength (ADS) = AAC / LHG
Home Goal Prediction (HGP) = HAG X ADS X LHG
Away Goal Prediction (AGP) = AAS X HDS X LAG
Final Score = Poisson Dist ( Max (HGP,AGP) ) for events/goals 0 - 5
Home - Real Madrid (RM) | Away - Barcelona (BL)
HAGRM= 1.89 | AAGBL = 1.315
LHG = 1.492 | LAG = 1.20
HASRM = 1.842 | AASBL = 1.095
HGPRM = 1.235 | AGPBL= 0.881
Poisson Distribution (HGP,AGP) for events k—-> 1 to 5.
Goals 0 1 2 3 4 5 RM 19.73% 32.02% 25.99% 14.06% 5.07% 1.85% Barca 43.86% 36.14% 14.89% 4.09% 0.84% 0.14%
Max (HGP,AGP) = RM (1) * BL (0) = 14.04 % or 0.1404
Hence the final score/outcome will be
RM (1 - 0 ) BL
The same algorithm and matrices can be used to predict the Final score between a Home and Away fixture based on its historical data.