xG BTTS system

shrewd. · November 2, 2019

Just trying something out, just a little idea.

I'm going to try and identify matches in which both teams are underachieving offensively & overachieving defensively according to xG (expected goals) data, and see how they do in the 'both teams to score' market. If there's more than one selection, then it could be worth combining for multis... and who knows?

The rule:

- Both teams must be overperforming xGA (expected goals against) and underperforming xG (expected goals scored) by at least 2.0 respectively.

*Note the rules could be tinkered/altered depending on system results/performance.

What is xG?

Well, xG is a data analysis recording that records a player or teams shots/chances whilst considering several variables to determine how likely was it to have scored/conceded. Variables such as the location (most crucial), the angle the pass came from, the shot type (for example, headers will accumulate lower goal probability/xG than a strike across the floor from the same position, by default), how many players were in front of the shot, etc. It's most generally used to get a better picture as to how sustainable a teams form is (negative or positive) or players goal run - or indeed, lack of. Basically, football's law of averages.

I touched on the location being the main factor. It's common knowledge the clearer cut chances you create or surrender, the likelier you are to score or concede. From the below image we have it split into 6 main zones, ordered from 1 to 6 in order of where you want to shoot from. Zone 1 & 2 are the "ideal" zones; short and central. Shots/chances from here will obviously accumulate the highest xG figures (measured 0-100; 100 probably being a ball sat dead on the goal-line with the keeper in a coma; I've yet to see a 100/100 xG chance - penalties have a 0.75/100 probability for reference). The 6 zones; wide, far and acute - will accumulate lottery-esque xG figures. Any team who scores from here should fine their keeper a weeks wages.

A team or player who's under-performing underlying xG numbers generally are creating/getting chances in the 1/2/3 zones, but for whatever reason, aren't scoring. Whether that's due to superman keepers, slack finishing or rotten luck - they're not scoring as many as the quality of their chances/chance creation suggests.

Conversely, a team who are "over-performing" xGA (expected goals against) are generally getting out of jail a lot, and surrender more high quality chances than their goals-against stat may suggest.

My logic behind using this angle for a 'Both Teams To Score' market system is if you find 2 teams in the same match who are have been getting lucky defensively and unlucky offensively, it has the recipe for goals.

Stay tuned.

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------