For any individual match, xG can be used to compute the probability that each team "deserved" to win based on the quality and volume of their chances — independent of the actual result. A team with 2.8 xG vs. their opponent's 0.7 xG had a strong process advantage; the simulation can estimate that they would have won roughly 75-80% of replays of that exact match. This reframes the match narrative from "we drew" to "we should have won X% of the time."
Simulate match outcomes by (1) treating each shot as an independent Bernoulli trial with probability equal to the shot's xG; (2) running 10,000 simulations of the full shot sequence; (3) computing the % of simulations each team won, drew, or lost. Report as: "If this match were played 100 times with the same chances, Team A would win 58 times, draw 22 times, lose 20 times." Supplement with: "Actual result was a 2-2 draw, which occurred in 22% of simulations — an unlikely but possible outcome." This separates process from luck in a single game.
Total xG comparison understates variance. A team with 1.0 xG from a single penalty (0.76 xG shot) has much higher variance than a team with 1.0 xG from ten 0.10 xG shots. The latter will score approximately 1 goal in almost every simulation; the former will score 0 or 1 in wildly different proportions. Shot-by-shot simulation captures this distribution difference that aggregate xG comparison misses.