Jump to content

A Rate Carry On...


Recommended Posts

Thanks to @Emerson Thome's sterling work compiling Owlstalkers' player ratings for each game, we now have a set of data which reflects our collective evaluations of our players so far this season.

 

I thought it might be interesting to look at how this compares to a more objective metric, namely whoscored.com's stat-based ratings, to see whether we have a tendency to overrate or underrate certain players.

 

So here we go - ranked by the difference between the stat-based ratings and the average ratings dished out by us Owlstalkers. Basically, the closer you are to the top of the list, the kinder we've been to you; the lower you are, the harsher we've been...

 

1427114776_Screenshot2021-11-16at13_19_00.thumb.png.726d22f60ee5f95515c6dea194ec69cc.png

 

So what does it tell us?

 

Firstly, we've underrated each and every player thus far, with each stat-based rating being higher than the scores handed down by the Owlstalk hive-mind!

 

Having said that, it seems that our newer recruits are getting a particularly raw deal from us, with ten of the bottom 11 players (ie those most underrated by us fans) all having joined in the summer.

 

A player's stat-based score being high or low doesn't seem to play a huge role in how far fans misjudge them, either: a player such as Wing scores slightly above Liam Palmer on the stat-based ratings, but is way down in comparison based on the Owlstalk ratings, for example.

 

I'd be interested to see people's thoughts on this: why do we have a tendency to underrate our players' performances, and why is this particularly pronounced for those who've only recently joined the club?

  • Like 3
  • Love 1
  • Thanks 5
Link to comment
Share on other sites

New recruits bring expectations, I suppose. Then reality appears on the scene (manager, opponents, etc) and we find new players are n't meeting those expectations, which are usually too high anyway.

 

Do the above stats account for the number of minutes played? And whether these are from the start or the bench?

 

How do they account for coaching decisions like putting square pegs in round holes? This must affect a players performance

  • Like 1
Link to comment
Share on other sites

5 minutes ago, edmontonowl said:

New recruits bring expectations, I suppose. Then reality appears on the scene (manager, opponents, etc) and we find new players are n't meeting those expectations, which are usually too high anyway.

 

Do the above stats account for the number of minutes played? And whether these are from the start or the bench?

 

How do they account for coaching decisions like putting square pegs in round holes? This must affect a players performance

 

No - both whoscored.com and Owlstalk ratings are based on what happens in whatever minutes a player has on the pitch, so if a player makes a 10 minute cameo, they get what they get based on what happens in that time.

 

Perhaps Owlstalkers try to account for coaching decisions when scoring players, but whoscored.com doesn't - it's a purely stat-based rating.

 

I think you're right in what you say: there does seem to be a pattern of fans expecting more from new players and then judging them harshly when they fail to meet those expectations, whereas most of the players who've been here a while are being rated more closely in-line with what the objective stats are suggesting.

  • Like 1
Link to comment
Share on other sites

I've never particularly understood or payed much attention to player ratings.

 

Following a game, can appreciate good performances, good things done in games, usually fairly easy to identify a MOM. But, assigning an arbitrary number doesnt really appeal to me. 

  • Like 3
Link to comment
Share on other sites

To make an informed judgement we really need to know how whoscored.com evaluate the ratings they give out, doubt they watch every game closely.

 

Lewis Wing for example having a rating of 6.75 is way off.

 

But the Owlstalk ratings are fueled by emotion, we're all disappointed for example that Wing, Berahino and Brown have been pants, the ratings reflect this.

  • Like 1
Link to comment
Share on other sites

Just now, striker said:

I've never particularly understood or payed much attention to player ratings.

 

Following a game, can appreciate good performances, good things done in games, usually fairly easy to identify a MOM. But, assigning an arbitrary number doesnt really appeal to me. 

 

Solid post. 

 

I'd give it a 7.5.

  • Like 1
  • Haha 8
Link to comment
Share on other sites

Just now, Weshallovercome said:

To make an informed judgement we really need to know how whoscored.com evaluate the ratings they give out, doubt they watch every game closely.

 

Lewis Wing for example having a rating of 6.75 is way off.

 

But the Owlstalk ratings are fueled by emotion, we're all disappointed for example that Wing, Berahino and Brown have been pants, the ratings reflect this.

 

The statistics whoscored.com generate are pretty impressive if they don't watch each game closely!

 

:duntmatter:

 

You can read a bit about what they do on their website if you're interested: https://www.whoscored.com/Explanations

Link to comment
Share on other sites

It's more to do with how it's scored.

 

The sites start a player with a 6, then add or subtract based on actions.

Subs start with a 6 too, so most short sub appearances hover around there.

 

Wing gets good marks as he's involved all game, and doesn't lose 6 marks for missing a shot as he does on OT.

Same as Dunkley and Paterson, it's based on what they do, not how comfortable they look doing it.

  • Like 3
Link to comment
Share on other sites

4 minutes ago, areNOTwhatTHEYseem said:

 

That's a good point.

 

I suspect a few ratings suffer from a lack of 'style points'.

 

:duntmatter:

More than a few I reckon.

 

Both have been among our most consistent performers.

 

Dunkley suffered for not being called Iorfa or Hutch, forgetting that the one he didn't play in was the 3-0 at Plymouth.

Pato suffers for looking like the third Mario brother.

  • Like 2
Link to comment
Share on other sites

3 minutes ago, Weshallovercome said:

If you take the 5 players in the red area of your list I don't really care how whoscored.com get their ratings, they're just wrong.

 

That's kind of the point of the thread: some of our subjective ratings don't tally with the objective stat-based ratings.

 

You can disagree with the weighting they give to certain actions when calculating their algorithm perhaps, but ultimately their rankings are based on concrete, measurable events which happened during our games.

  • Like 1
Link to comment
Share on other sites

5 minutes ago, FreshOwl said:

How are whoscored.com’s stats collected? 
 

We watch our team week in week out for 90 mins, whereas they don’t 

All of the sites and TV etc. get them from Opta.

Pretty sure Opta get the data from the bras that the players wear too.

Link to comment
Share on other sites

2 minutes ago, FreshOwl said:

How are whoscored.com’s stats collected? 
 

We watch our team week in week out for 90 mins, whereas they don’t 

 

Of course they bloody do!

 

How else do you think they compile all the stats? They don't stop fans as they leave the ground and ask them to estimate how many aerial duels Liam Palmer won, FFS!

 

:duntmatter:

 

They use Opta for their raw data. Here's a snippet of how it works:

 

'Opta hires teams of full-time and part-time analyst to watch every single game that takes place and, with the use of their data collection video software, notate all the various events that occur on the field, often capturing up to 2,000 pieces of data per match. Opta's data collection software operates similarly to a video game, where each combination of buttons will represent an action by a certain player allowing the analyst to press the appropriate combination of buttons as they watch the live game. Three analyst will be involved in each game: one for the home team, one for the away team and a third one to double check the data. The data is then checked by a post-match team to ensure 100% accuracy.'

Link to comment
Share on other sites

8 minutes ago, areNOTwhatTHEYseem said:

 

Of course they bloody do!

 

How else do you think they compile all the stats? They don't stop fans as they leave the ground and ask them to estimate how many aerial duels Liam Palmer won, FFS!

 

:duntmatter:

 

They use Opta for their raw data. Here's a snippet of how it works:

 

'Opta hires teams of full-time and part-time analyst to watch every single game that takes place and, with the use of their data collection video software, notate all the various events that occur on the field, often capturing up to 2,000 pieces of data per match. Opta's data collection software operates similarly to a video game, where each combination of buttons will represent an action by a certain player allowing the analyst to press the appropriate combination of buttons as they watch the live game. Three analyst will be involved in each game: one for the home team, one for the away team and a third one to double check the data. The data is then checked by a post-match team to ensure 100% accuracy.'


Ahh I see. I thought they just went off the match stats and/or individual stats, and gave a rating based off that 

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...