Statistical Analysis Types
The 10 statistical result types that the Oxiano model analyses, with their associated methodology and calculation factors.
XGBoost + Poisson model · 225,000 matches · 10 European leagues
Final Result — 1X2
Primary modelThe tripartite classification of a football match result into one of three possible outcomes: home team win (1), draw (X) or away team win (2). This is the foundation of any statistical football analysis.
The Oxiano model achieves 74–86% accuracy on this analysis type at confidence ≥65%, depending on the league. La Liga and Bundesliga show the highest degree of statistical predictability.
Goal Volume — Over/Under 2.5
Offensive volumeAnalyses the probability that the total goals scored in a match will exceed or remain below the 2.5 goal threshold. This is the most widely used volume threshold in statistical football match analysis.
Matches between teams with combined xG >2.8 per game have a statistically high probability of exceeding the 2.5 threshold. The model integrates this variable as a primary classification factor.
Mutual Scoring — BTTS
Defensive activityBoth Teams To Score (BTTS) quantifies the probability that both teams will score at least one goal during the match. It reflects the balance between each team's offensive capacity and the opponent's defensive vulnerabilities.
Teams with a BTTS rate >65% in their last 10 matches show high statistical consistency for this classification. The model weights offensive and defensive form separately.
Double Chance
Composite probabilityQuantifies the composite probability of two of the three possible outcomes simultaneously: 1X (home does not lose), X2 (away does not lose) or 12 (neither team draws). Represents the sum of individual probabilities from the 1X2 matrix.
Double Chance is derived mathematically directly from the 1X2 analysis. A model with high 1X2 accuracy automatically produces correct estimates for the composite variant.
Asian Handicap
Adjusted analysisThe Asian handicap system eliminates the draw possibility and redistributes probabilities across a continuous spectrum. The model calculates adjusted probabilities by applying a relative strength offset (Elo differential) to the raw 1X2 distribution.
Asian handicap reduces analysis variance in unbalanced matches. The Oxiano model calculates adjusted probabilities for standard lines: -0.5, -1, -1.5, +0.5, +1.
Correct Score Projection
Poisson distributionEstimation of the complete distribution of possible scores through the bivariate Poisson statistical model. The model calculates the probability of each individual score (0-0, 1-0, 1-1 etc.) based on the attack and defence rates of both teams.
The Poisson distribution produces a complete probability matrix for all possible scores. Scores of 1-0, 1-1 and 2-1 statistically cover ~45% of all European matches.
Low Goal Threshold — Over/Under 1.5
Minimum volumeAnalyses the probability that the match will produce at least 2 goals (Over 1.5) or remain at a maximum of 1 goal (Under 1.5). The 1.5 threshold is relevant in matches featuring highly defensive teams or in high-stakes tactical encounters.
Statistically, ~88% of major European league matches exceed the 1.5 goal threshold. Under 1.5 is a lower-frequency classification but carries increased predictive power when identified by the model.
Half Time Result
Temporal analysisStatistical classification of the result at the end of the first half, independent of the final result. The model analyses match-opening trends, tactical starting styles and the correlation between the half-time and full-time result.
HT/FT correlation varies significantly between leagues. In the Premier League, ~55% of half-time results hold at full-time. The model calibrates this correlation per league.
Win to Nil
Defensive-offensive efficiencyQuantifies the probability that a team wins the match without conceding any goals. Represents the statistical intersection between offensive capacity (scoring at least one goal) and complete defensive solidity (conceding no goals). One of the model's most selective classifications.
Win to Nil combines two independent probabilities: win and Clean Sheet. The model calculates their intersection, producing a low-frequency classification (~15–25% of matches won) but with high predictive power.
Clean Sheet
Defensive solidityAnalyses the probability that a team concedes no goals during the match, regardless of the final result. It is a pure indicator of defensive solidity and the opponent's inability to convert created chances. The model analyses Clean Sheet separately for the home and away team.
Home team Clean Sheet occurs statistically in ~30–35% of major league matches. Correlated with defensive solidity (GAA <1.0) and low xG Against (<0.8), the probability increases significantly and becomes statistically relevant for the model.
All 10 analysis types, updated daily
The Oxiano model runs automatically at 07:00 and 13:30 and publishes analysis results for the day's matches.
The content of this page is for exclusively educational and informational purposes. Statistical analysis · Not betting advice · oxiano.com