I did some work on what I’m calling “Primary Chance Creators.” There has been a lot of talk in Austin media/fan discussions about need for a classic 10 or “primary chance creator,” so I wanted to see to what degree teams are reliant on one or two chance creators to a great extent, as well as how a team with a large degree of reliance on one or two players fares in goal scoring / xG.
All stats are from FBRef.
Buckle up, this is a lot.
I took every team from the 2023 and 2024 seasons and found their total number of shot creating actions (SCA), which can include any live ball pass, dead ball pass, take on, shot, foul drawn or defensive action that leads to directy to a shot. (This means that a player can have an SCA and a shot on the same play.)
I also found the top two players on each team in SCA, as well as their SCA/90 and the percentage of the team’s total SCA each player had. My final added metric was the delta between a team’s highest and second highest SCA players.
Here's the Google Sheet with the data and analysis - Correlation results is the top level stuff. Master Sheet has all of the data, including the specific players, if you want to see that. Each tab with titled with the metric (explained in Glossary tab) has the graphs for that metric and how it correlates to xG, xGD, Goals, Goal Difference, and Points.
Some conclusions:
There is very little correlation between just *having* a “primary chance creator” and any of the measures of success in my study: xG, xGD, goals, goal difference and points. However, the teams who have the best chance creators on a per 90 basis have a much better chance of being successful by any of those measures of success. (This is probably obvious).
With regard to “primary chance creators,” the correlation between the SCA % of the TWO highest SCA leaders and the measures of success is much higher than the % of the single highest SCA, but still not a significant correlation, hovering around .2 in most of the measures.
Another obvious thing I found (confirmed?) is that the teams with the highest total volume of SCA tend to be the best teams in any of the measures of success, especially in xG/xGD, which both had a correlation right around .79.
One more note of interest: The biggest outlier in any of my correlation sets is “SCA/90 of a team leader” with “goals,” which had a correlation of 50% higher than xG and twice as high as xGD. I think this is more evidence in the theories we have that the top end of talent players in MLS have an outsized impact on goal scoring, and can play a role in overperfomlng xG, like what we saw with Miami this year (79 goals from 53.8 xG).
My main takeaway right now is I don’t think you need a *primary* chance creator, in the sense that there should be one or two main players who create most or a lot of the chances. It might even be more of advantage to have two players who share that role. But there need to be players who create chances, and having one of the players who is the absolute the best at it is a huge advantage, and, obviously, the more chances a team creates, the better their outcomes are.
What does this mean for Austin FC in 2025? I think it means we don’t really need a “10”, but we do need players that create chances (obviously!). The findings here suggest that it’s possible that the total number of chances created is more important than one player who creates most of them. I wanted to see what we could expect from the new players coming in, but the data just doesn’t really exist publicly, especially with FB Ref dropping Liga MX advanced stats.
The surest way to correlate chance creation with success is by having one of the very best ones, like Reynoso, Almada, Lucho, Gil, Puig or Messi, but it doesn’t guarantee it as we can see with Gil. And we know how difficult it is to acquire the other kind of guys.
Comments
Post a Comment