Looking for Greatness Part II

I did some work last week about stats to look at when you want to find consistent greatness. Some stats are incredibly fickle from year-to-year, and so doing well in that area one season may not be entirely indicative of improving to the caliber that the numbers would suggest. That piece only covered shooting and finishing goals, though – which is only part of a forward’s job.

Another big piece is creating the shots. While there are several ways to go about creation, today I’m specifically going to focus on creation via passing. Thankfully, there’s a pretty useful stat for shot creation via passing already around called “chances created” – which just looks at passes that lead directly to shots on goal. From there, the stat can be broken into subcategories – key passes and assists. An assist is a pass that leads directly to a goal scored while a key pass is a pass that leads to a shot on target without going in.

To start my initial inquiry I wanted to see if there was any correlation between years of the way chances created were finished. To do this I simply gathered the numbers for La Liga forwards (via squawka.com) with more than 1,000 minutes in both 2013/14 and 2014/15, and did some simple math. To test conversion rates, I simply took assists (i.e. the number of chances created that went into the net) divided by total chances created – below is the graph.

Screen Shot 2015-11-01 at 6.59.21 PM

This had a rather high correlation, in my opinion. I expected that with in randomness associated with finishing there would be some randomness associated with how chances created were finished. While I wouldn’t say this is very consistent stat, there seems to be some predictiveness in the way that players have their chances created turned into goals. It would make sense that some players tend to put their teammates in consistently better positions to finish, though it seems like there would be more statistical noise due to the nature of finishing itself. I didn’t want to stop with an r^2 value of 0.39, though – so I kept going.

Next I wanted to see if there was any correlation between the number of passes made and the number chances created. Again, using squawka.com, I took the total number of chances created and divided it by the total number of passes attempted. The idea here is that better players turn their possession and passing into shots on target more than others. As it turns out, that notion doesn’t seem to have very much evidence to support it.

Screen Shot 2015-11-01 at 7.19.47 PM

In fact, the highest rate over the last two years doesn’t belong to creative phenom Lionel Messi. It doesn’t belong to the Nolito – who had the highest number of chances created in a single season over the last two years. It belongs to Pablo Piatti – who turned nearly 10% of his passes into shots on target in 2014/15.

With the rate at which chances created were finished and the rate in which possession was turned into shots out of the way, I decided to try one more method to see if there was something else that might hold some merit. Perhaps this should have been my first test (it is the most simple, after all), but it wasn’t. It was third.

I wanted to see if there was any correlation in the rate at which chances created were produced. Not compared to how much possession a player had, or anything else, just the rate of chances created per 90 minutes. Perhaps it doesn’t matter how often a player touches the ball, creative players will find ways to open opportunities for teammates. So, once more, it was back to the numbers.

Screen Shot 2015-11-01 at 7.00.20 PM

While this may not look like much, it is. There isn’t as much correlation between the rate in which shots are taken, but there is a solid correlation suggesting that good players will create chances. For instance, in 2013/14 Messi averaged “just” 49 passes per 90 minutes (which is a lot for forwards) and 2.7 chances created per 90 minutes, while in 2014/15 he averaged 61 passes per 90 and 2.5 chances created. It didn’t really matter that he had less possession, creative players produce for others regardless of how often they touch the ball.

There’s definitely more work to be done, but this is at least a start. There are at least two stats to help determine great creative players with solid correlations. There are enough variables in the finishing rate of chances created for me to remain skeptical (especially at a 0.39 r^2), but I am buying into chances created per 90 minutes. There seems to be some serious indication that creative players “get theirs”, and it makes sense logically.


3 thoughts on “Looking for Greatness Part II

  1. Thanks for this, it was an interesting read.

    One suggestion: When the x- and y-axes are measuring something similar, make the graph square. That will make it easier to look at how close the data points are to the center line, and so to tease meaning out of the data.


    1. That’s definitely a good suggestion! There’s probably a lot of “attention to detail” things I could improve along the way, but that’s arguably the biggest one. I usually just get the graph made and move on, but improving the aesthetics is probably a big thing. Thanks!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s