I like Shot Creating Actions. It’s a good way to look at how teams create their opportunities. And on the flipside, how they allow opportunities against them. There are six types of Shot Creating Actions tracked over at fbref.com, and it’s important to note out that these are:
The key there being that they are the TWO offensive actions. This covers essentially things like the ‘hockey assist’ that isn’t normally tracked, which is nice. There are six types of Shot Creating Actions (SCA) you will see in the datasets over at FBREF:
- PassLive – These are when a pass is made during live play
- PassDead – These are set piece situations, be it free kicks, corner kicks, throw-ins, and goal kicks
- Drib – These are successful dribbles that lead to a shot attempt
- Sh – These are shots that lead to ANOTHER shot attempt (important distinction there)
- Fld – These are fouls drawn that lead to a shot attempt
- Def – These Defensive actions that lead to a shot attempt
It’s important to note that while all of these occur, the Live Pass SCAs are by far the most frequent and common, which makes sense anyway, given that most of a match is spent during live play. Below is a combination of the above six SCA types stacked for each league’s club average during a match:
Most of the leagues are pretty consistent, although La Liga definitely has less overall. Now, if we were to look at say the Top 8 sides for each league in the table, then things change up just a bit:
Notice how much higher the values are for Live Pass in the Top 8, and also that while Dribble and Shot go up a bit, Dead Pass, Fouled and Defensive Action really stay pretty much the same. Most teams have similar quantities of dead ball situations, and get fouled similarly. Dribbling and Shooting do improve a minor bit with higher quality sides, but it’s the Pass Live category that really jumps. Serie A, for example, increases nearly 24%. La Liga increases 27%. This is where real variance occurs in these leagues. Better teams have the ball more, which equals more live passing opportunity, and they also simply take more shots in the process.
Now, none of this is possession adjusted, or anything like that, because that’s not my point today. My point is to show how much better these teams are at this, AND, in the next part, I’m going to show how much better they are at limiting these by their opponents as well. Again, possession time and all that, but this is precisely why better teams keep the ball.
This below is how the teams average by league top to bottom for allowing SCAs:
Wait, you are saying, in that Spider-Man pointing at Spider-Man moment, this looks just like the first chart. Sure, it is, because that makes logical sense. But I wanted another copy down here for reference, when I show you the Top 8 one here:
Notice not only the big improvement on the Pass Live category, but also how the other categories improved a little, but again, not by much. In Serie A, for example:
- Pass Live decreased by 2.1
- The other five decreased by a combined 0.48
Needless to say, only around 20% of the reduction came from those other five types, and around 80% from Pass Live. It’s the biggest factor in either direction. Also note that the gap from SCA FOR to SCA Against with the Top 8 filter is, in Serie A, a whopping 5.45. So, on average, for those top 8 sides, they create 5.45 more Pass Live SCAs than they concede. Almost 50% more.
Now, I also did a little digging, I wanted to see for each league, how this compared, and it was fascinating how much of a gap there is between the top and the rest normally. Let’s start with Serie A:
These are JUST Pass Live SCAs, for and against. I included a trend line and average lines, because of how neatly squads seem to cluster together. Notice:
- Great Attack, Great Defense, bottom right: 1,2,3,4,5,6,7 (2 barely). This is as expected.
- Good Attack, Bad Defense: 9,12. Again, makes sense
- Bad Attack, Good Defense: 11,14,15,18.
- Bad Attack, Bad Defense: 8,10,13,16,17,19,20. I wouldn’t expect 8 and 10 there, but it happens. Hellas Verona is really hurting those Top 8 averages up above, too. Kind of canceling out the ridiculous numbers Atalanta have.
If we look at the Premier League:
The PL is more all over the place, but there have been a lot of both underperforming and overperforming sides this year, which is impacting that. There’s no real ‘clumps’ of four, like in Serie A, but rather a simple hugging of the trend line. Here are some interesting outliers, tho:
- Liverpool – Obviously everyone’s favorite to crash and burn, unless you are a Liverpool fan. Doing the right things, but clearly neither finishing or saving effectively.
- Brighton and Fulham – Head scratching, and folks have been all season. And why is Leicester there?
- Man United up there with Leeds and Aston Villa. Ehh? Oddity.
- What I’m going to call the Foggy Five: Everton, Tottenham, West Ham, Arsenal, Wolves. All five are not in a spot folks thought they would be, for better or worse.
- Southampton. LOL.
- And lastly, what really feels like should be the bottom five, of West Brom, Burnley, Newcastle, Sheffield UTD, and Palace.
Okay, I’ve got three leagues to go! Ligue 1 next:
This is definitely a ‘haves’ and a ‘have nots’ league this year.
- Lille, PSG, Lyon, Monaco, Lens, and … Rennes? Yeah. I’ve got no problem with that. Given all that’s going on, it’s not surprising that they look better than the results right now at Rennes.
- It feels like there’s two mini clusters with 9/14/16 and 7/8/13, and then the rest kind of forming the bottom. Marseille is a bit of an odd one.
On to the Bundesliga!
A bit like the PL this year (how many ways is the Buli going to be described like that, by the way?), and I get the feeling if Bayern actually cared about defending they would be even better on this chart.
- You see Bayern by themselves, RBL by themselves, and then 3/4/5/6/7/9/11 all kind of loosely grouped.
- Freiburg and Schalke are just bad, and the rest hugging that line like die guten kleinen Jungs, die sie sind
Lastly, La Liga. This one is a fun one.
This is quite literally like one of those math problems from algebra, where there are nearly two asymptotes here dividing up the group at those average lines.
- Somehow Getafe and Eibar have found a way to defend really well, while not creating either, in two of the odder placements of the entire five charts.
- Beyond those two, you’ve got 1-9 all in the ‘good box’
- And you have 10-14, 16,17,19,20 all in the ‘meh box’.
While there definitely are outliers, it’s my opinion that these outliers help to actually define that in most cases, if you keep your Pass Live SCAs high, and you minimize the Pass Live SCAs of your opponent, you are going to find success. Seems an obvious thing, but it’s quite cut and dry, I think.
If we look at all five leagues together, and use THAT average:
- Rank 1 – All in Lower Right Box
- Rank 2 – 3 in LR Box, 2 in Upper Right Box (near tho)
- Rank 3 – All in LR Box
- Rank 4 – All in LR Box
- Rank 5 – 3 in LR Box, 1 in Lower Left Box (near), 1 in Upper Left Box (near)
- Rank 6 – 3 in LR Box, 2 in UL Box (near)
- Rank 7 – 2 in LR Box, 1 in LL Box (near), 2 in UL Box (near)
So from the top 7 ranks in all five tables, that’s 26 of a possible 35 in that Lower Right box, with the other 9 not too far away.
Notice also, how little there is in the “High Pass Live For, High Pass Live Against” quadrant. I assume because it’s simply difficult to do. Pushing for shots yet being exposed. Sounds bad tactically.
Similarly, notice how few are in the “Low For, Low Against” quadrant. Again, difficult, someone’s gotta shoot at some point. Clubs that simply aren’t offensively talented, yet trying super hard defensively.
Overall this has been an interesting metric to dig into, and review. I was curious how impactful it is, and it’s clearly playing a massive role for how sides’ results are in the table.