Dees “Effective Kicks” Data

[Watson11 2,240]

7 hours ago, Roost it far said:

I have no idea what that is saying. What does a threat rating of 18 and an expected threat of 0.31 mean? 

Expected threat of 31 means that compared to every kick in his database (170000 kicks) when that player takes a kick from a similar position, to a similar area, under similar conditions, it would be expected to end up in a score 31% of the time.  The threat rating of 18 means compared to the expected threat, the specific players actual kicks resulted in a score 18% more than expected.  Chandlers kicks would be expected to result in a score fairly often (31% of the time) but actually result in a score 49% of the time. 

It would be far better if he also include the opposite ie turnovers and scores against by the same players.  

[layzie 33,740]

1 minute ago, Watson11 said:

Expected threat of 31 means that compared to every kick in his database (170000 kicks) when that player takes a kick from a similar position, to a similar area, under similar conditions, it would be expected to end up in a score 31% of the time.  The threat rating of 18 means compared to the expected threat, the specific players actual kicks resulted in a score 18% more than expected.  Chandlers kicks would be expected to result in a score fairly often (31% of the time) but actually result in a score 49% of the time. 

It would be far better if he also include the opposite ie turnovers and scores against by the same players.  

Great stuff. 

I also just liked seeing 31 mentioned a lot.

[old55 23,356]

I think threat rating is skewed favourably for players in the forwardline, there's less possessions in a scoring chain involving their kick compared to backline players.

[Bitter but optimistic 22,218]

I've made a number of posts recently stating how totally confused/mystified I am by rules interpretation and umpiring in games.

Now along comes yet another stat ..............!

FMD !!

[old55 23,356]

29 minutes ago, Bitter but optimistic said:

I've made a number of posts recently stating how totally confused/mystified I am by rules interpretation and umpiring in games.

Now along comes yet another stat ..............!

FMD !!

People frequently rightly complain about the meaningless disposal effectiveness stats as pointed out in the article.  It's good to see someone trying to do it better.

[Watson11 2,240]

43 minutes ago, old55 said:

I think threat rating is skewed favourably for players in the forwardline, there's less possessions in a scoring chain involving their kick compared to backline players.

Not entirely.  It’s the expected threat rating (y axis) that is skewed. The threat rating is relative so it compares players taking kicks from similar positions against the average. Ie the 0 threat measure is average.

Bowey was low on both measures.  He is a defender but it basically says he takes safe kicks and they result in scores to us less than the average. In my view, it would be better if expected threat also subtracted how often the kicks resulted in score to the opposition, and the threat did the same.  I could imagine Frosty would look ok with just the offensive numbers, but if you included turnovers not so good.