Statistics are a commonly cited aspect of football analysis, but how much do we really know about the numbers we throw around? We got in touch with Big Dave from OPTA ’s data measuring department, to break down the stats.

One major innovation in recent years has been the advent of expected goals (XG), something Dave revealed to have a complex analytical process behind it; “If a team has an XG of 1.5, but they haven’t scored at all, then our statistical models would rate that team as ‘utter shit’. Similarly, if a team has an XG of almost zero but a total goals of exactly 1, we would rate that team as ‘Burnley’. However, if a team has an XG of 2 but they have scored 4, our statistical models would rate that team as ‘probably containing Messi’.”

Possession is another area that Dave classifies as being commonly misunderstood, “A lot of people think we’re just randomly assigning a number to something that objectively has no numerical value. The truth is that possession stats are hugely informative. If a team goes through a spell of only having 34% possession then you can say, with a degree of confidence, that they haven’t had much of the ball, while if they go through a period of having 64% possession you can probably say, again with a degree of certainty, that that team has probably played quite a lot of recent passes in the match, which is certainly meaningful in many important ways.”

He went on to add that if you read through all the data you would know who won the game without looking at the score. We wondered what the point of that was.