Does a public xG calculator exist?

[a8592]

Curious if there's a publicly available xG calculator of any kind?

For example, somewhere you could input shot data such as location, type, etc. and the xG value would be calculated?

Thanks

Click to read more...

 

[Bear of Bad News]

I'm not aware of anything like that in the public domain.

 

[morehockeystats]

Everyone, including me, has different parameters for xg. Some overlap, but that's it. Even the basic one, how the sectors of the offensive zones are shaped, is different. I am pretty sure nobody does it on a resolution of a foot, that's too low a sample.

 

[Hockey Outsider]

This is (as far as I'm aware) the first article on expected goals that was ever published (by Alan Ryder, in January 2004): http://hockeyanalytics.com/Research_files/Shot_Quality.pdf

This might be overkill for what you're asking about, but he explains how he came up with his model, and runs through the results from the 2002-03 season. There are plenty of other articles about expected goals (or, as he calls it, shot quality) on his website, and models from the past decade or so are much more refined. But this is a really good introduction to the topic.

 

[a8592]

This is (as far as I'm aware) the first article on expected goals that was ever published (by Alan Ryder, in January 2004): http://hockeyanalytics.com/Research_files/Shot_Quality.pdf

This might be overkill for what you're asking about, but he explains how he came up with his model, and runs through the results from the 2002-03 season. There are plenty of other articles about expected goals (or, as he calls it, shot quality) on his website, and models from the past decade or so are much more refined. But this is a really good introduction to the topic.

Thank you for this. I am fairly familiar with rationale behind calculating xG and how models may vary. What i basically meant to ask is that because i am incapable of writing code / creating a model - I was wondering if there was a model I could plug my information in to by shot, then have the model give me an assigned xG value