First of all, I’m a fan of xG — with caveats.
The xG metric, if you’re new here or haven’t been following the stats part of my coverage too closely, is a stylized abbreviation for expected goals, and in its simplest form, it’s a way to gauge how many goals a team should be scoring in a given game. (It’s a metric that also works better statistically as you accumulate data, so it’s a lot better for projecting goals scored over a season than a single match, even though it’s often used in relation to explaining a single game.
The way it works is that each shot gets rated on a percentage scale, so a tap-in of a ball on the goalline is a lot closer to the 1.0 end of the scale, and a shot from distance and near the sideline will be closer to the 0.01 end of the scale. Penalty kicks, because of their special nature, are worth 0.79 xG, as typically about four of five get converted.
So, if you see that Real Salt Lake beat FC Dallas 3-2, in a match in which RSL racked up 2.73 xG and FCD gathered 2.03, you see that the goals scored and the final margin of victory roughly correspond to what the xG shows you.
It doesn’t always work like that, though. In July’s 1-0 Austin FC home loss to the Sounders (the one with the Jordan Morris goal and the Brendan Hines-Ike sendoff), Seattle had a 0.78 xG from nine shots and Verde had an 0.64 xG from 13 shots.
While the 1-0 scoreline isn’t surprising, that single-game xG might lead you to predict a 1-1 draw.
But as we’ll explore, just because you accumulate xG through more shots doesn’t necessarily mean you’ll score more. (But you already know the maxim about missing 100 percent of the shots you don’t take.)
