Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals) ?? (2025)

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (2)

By now in footballing circles, ranging from the higher echelons of professional football to passionate fans of different clubs, people have been familiar or know the existence of Expected metrics, starting from the simple xG (Expected Goals). But for the sake of even the simple reader this article may find, xG is defined as the statistical likelihood of a shot ending up as a goal from any one point of pitch, ascertained by machine learning models trained on loads of shots take across the world in football.

While xG has been so useful in predicting trends across a larger sample, mere xG/npxG numbers over/under performance arent enough to give a clear picture of certain situations. Say for Example Potter’s Chelsea or Inzaghi’s Inter have underperformed highly on their xG yet that point isnt alone conclusive without looking at shot quality of these teams.

Here’s where post shot models comes in which calculate PSxG/xGOT, and according to Opta it’s defined as follows: “It measures the likelihood of an on-target shot resulting in a goal, based on the combination of the underlying chance quality (xG) and the end location of the shot within the goalmouth”

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (3)

Let’s start diving into the specifics: The viz above shows teams in Italy wrt xG generated & the goals scored. As we talked about Inter Milan earlier, they rank the highest in xG accumulated with 55 but have scored only 48 goals showcasing a big difference of 7 in underperformance. But this alone doesnt conclude anything on their goalscoring woes because higher xG alone wont quantify good chances created. To get a deeper idea, we next take a look at how well teams are getting shots on Target:

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (4)

Now things get Interesting. The team that accumulated the highest xG in the League, Inter ranks 13th in getting shots on Target, meaning that many of their shots that they take are either blocked or go wayward. Lazio meanwhile get the most shots on Target amongst teams in Italy, but once again the question begs that how do we qualitatively assess these shots?? Meaning how do we know which team/teams test opposing Keepers better than others?? This is where usage of Post shot models make more sense.

Post shot models can help describe how well teams are able to enhance from the position they take shots from, in other words how well teams better test opponent Keepers, whereas sole xG values can tell about positions a team can get into, when both are taken on a per shot basis.

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (5)

When we look at Post Shot xG accumulated, Inter rank as 48.1, very close to the Goals they have scored. Consequently Napoli who lead in PSxG, have outperformed it when you look at their Goals scored at 66. Here again we have to understand that in a game a team can take X number of shots & accumulate xG but if all X of them are not on Target they accumulate 0 PSxG. Consequently a team on a bad day may accumulate large PSxG but a Great Keeper performance may keep them from scoring. Hence PSxG is also the same metric that is used to measure shot stopping abilities of Keepers. A good example of use of PSxG alongside xG is given below:

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (6)

In the recently concluded Europa League Quarter Final Tie, Juventus won 1–0 thanks to a Closing stages goal from Federico Gatti. The game ended with an xG scoreline of 0.9–1.7 for Juventus-Sporting respectively. But on PSxG scoreline it ended as 1.01–2.26, with Mattia Perin’s 2 saves in close range clearly proving the difference in scoreline. Now what this means is Sporting enhanced shot Probability of scoring from the locations they took but were met by excellent saves which means their players arent at Fault for them not scoring any Goals that night. Juventus though managed to enhance their shot to the highest value possibel of 0.97 with the way Gatti took his shot, a shot that’s 97% probability of going in, got their goal.

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (7)
Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (8)

Another Example is the Inter vs Monza game, where the xG scoreline finished 2.1–0.7 respectively for both, but when you look at PSxG they accumulated on the shots on target they took, you can see in Pic 6 & Vid 1, Lautaro martinez reduce the Probability of Goal with his shot taken from a location that has 0.45xG but after the shot it’s probability reduced to 0.2, 20% chance. Similarly Joaquin Correa’s 0.13xG shot had reduced chance of scoring by almost Half with a shot which was only probable to go 7% (0.07 PSxG).

With xG Numbers alone we can say Inter had a finishing Problem, but with PSxG added to it, we are able to pinpoint specifics & perhaps where does larger blame lie.

As discussed before PSxG is also beneficial to find games where a Keeper may have let down wrt a result while also talking about the finishing struggles we may have highlighted before.

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (9)

We did talk about Inzaghi’s Inter & their struggles & were able to conclude to a better degree about their problems on getting shots on target & eventually PSxG numbers to see how well Inter forwards make most of their chances. Now Let’s talk on Graham Potter’s chelsea (during his time as coach for most of the season) . Chelsea this season have had serious Goalscoring problems & from Pic 7 it’s fair to say that Chelsea rank 10th in xG geenrated with 39.5, but what’s more intrguing will be they have underperformed on their xG with only 30 Goals scored.

Unlike Inter’s case this is different because degree of underperformance is larger & volume of xG generated low. If we once again look back to how we dived into Inter’s woes taking Shots on Target % as a tool, we see that

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (10)

Chelsea rank low in Shots taken Per90 which obviously will translate a lot into lower xG accumulation, but at same time when SOT% is lower, iminent result is also that their Goals have dried up & consequent underperformance. But a look at PSxG in Pic 9 Below will reflect

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (11)

the fact once again that it’s closer to Goals scored than xG Chelsea generated. But once again we do have to remember the fact taking Arsenal’s case or Napoli’s in earlier Serie A study who both have a larger gap between PSxG & their Goals scored that Teams may have found route to Goal better than some teams, a simpler example can be illustrated below in Pic 10 when Chelsea played Aston Villa:

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (12)

For example in the Pic above we can see the shot data for the game after which Potter was relieved of his duties when he his Chelsea lost to Aston Villa. John Mcginn took a shot from distance at a position valued with 0.02 xG, Post shot models show he enhanced the probability of it to 31% & it went into Goal. Could the Chelsea Keeper have done better to save it?? Maybe, here it’s difficult to water it down to Mcginn’s enhancement technique or Kepa must have done better or not. I will leave the Goal Video below for you to decide:

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (13)

But maybe one look at Joao Felix’s shot from closer distance, a shot that had 0.05xG, difficult position to score was enhanced to 0.39 PSxG value but was saved by Martinez probably asks questions of whether Chelsea met a Keeper that performed very well & also whether the Chelsea’s route to goal depended largely on players enhancing shots from very low goalscoring probability positions:

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals)?? (14)

Perhaps the situations like above are where we get the idea of drawing precedence for usage of xG & PSxG combined. On a visual level it’s a great save, on a statistical level only with xG it’s a low probability position, but with PSxG value, we can combine all 3 & say Chelsea took difficult routes to Goal & were met with Great keeper meaning it’s now down to player, keeper or coach but all chipping in contributions.

CONCLUSIONS

If you made it till here without getting confused as much, the idea around this article is simple: Simple xG Numbers over a season or a game shows one dimension of understanding Goalscoring problems if any team has. Addition of Post shot metrics opens the other dimesnions where accountability can be more concisely put on or spread amongst the factors. We saw 2 great cases in European football with Inter & Chelsea about their underperformance wrt xG & Goals scored.

Simple case can be seen with 2 examples in Vid 1 & Vid 3 of Lautaro Martinez’s & Joao Felix’s chances where route taken to Goal was better by Inter but Lautaro’s shot technique decreased probability of scoring from an already high probable position. It was opposite in case of Felix, who enhanced probability from a low probability (xG) position. Hence Coaches should take more precedence in how they design a team’s ability to accumulate xG, PSxG take larger precedence on Players (with some on coach’s own methods to improve players) to finish chances.

With that we can conclude with greater degree that Maybe Potter factors in more than his players in Chelsea’s scoring woes whereas in Inter’s case, perhaps Inzaghi factor isnt the greater problem. Of course end of the day, there’s no one single factor to explain any woe in football, it’s always a consequence of Multiple Facors!!

Does PSxG/xGOT (Post shot Expected Goals) needs to take as much Precedence as xG (Expected Goals) ?? (2025)

References

Top Articles
Latest Posts
Recommended Articles
Article information

Author: Kelle Weber

Last Updated:

Views: 6331

Rating: 4.2 / 5 (73 voted)

Reviews: 88% of readers found this page helpful

Author information

Name: Kelle Weber

Birthday: 2000-08-05

Address: 6796 Juan Square, Markfort, MN 58988

Phone: +8215934114615

Job: Hospitality Director

Hobby: tabletop games, Foreign language learning, Leather crafting, Horseback riding, Swimming, Knapping, Handball

Introduction: My name is Kelle Weber, I am a magnificent, enchanting, fair, joyous, light, determined, joyous person who loves writing and wants to share my knowledge and understanding with you.