News:

Use a VPN to stream games Safely and Securely 🔒
A Virtual Private Network can also allow you to
watch games Not being broadcast in the UK For
more Information and how to Sign Up go to
https://go.nordvpn.net/SH4FE

Main Menu


Fulham stats and stuff 2019-20

Started by MJG, July 21, 2019, 01:59:25 PM

Previous topic - Next topic

toshes mate

The only definition I can find about the 'long ball' is that it is "a forward pass of 35 yards or more that is kicked into space rather than a precise pass aimed at a particular team-mate".  Is there a more precise definition than this since the word 'forward' is ambiguous as it makes a potential arc of only slightly less than 180 degrees, and, when does it become a 'precise pass' to a team-mate when that team-mate is running into the space where the pass has been aimed? 

I am genuinely curious.

MJG

Quote from: toshes mate on January 20, 2020, 08:56:49 AM
The only definition I can find about the 'long ball' is that it is "a forward pass of 35 yards or more that is kicked into space rather than a precise pass aimed at a particular team-mate".  Is there a more precise definition than this since the word 'forward' is ambiguous as it makes a potential arc of only slightly less than 180 degrees, and, when does it become a 'precise pass' to a team-mate when that team-mate is running into the space where the pass has been aimed? 

I am genuinely curious.
I'm led to belive its a pass (any direction) of 25 yards or more, which i always thought it was too small. But whatever it is does give a indication of how a team play changes at times, be it 25/35/50 as long as it stays the same for all.
Just the views of a long term fan

toshes mate

Quote from: MJG on January 20, 2020, 10:30:37 AM
Quote from: toshes mate on January 20, 2020, 08:56:49 AM
The only definition I can find about the 'long ball' is that it is "a forward pass of 35 yards or more that is kicked into space rather than a precise pass aimed at a particular team-mate".  Is there a more precise definition than this since the word 'forward' is ambiguous as it makes a potential arc of only slightly less than 180 degrees, and, when does it become a 'precise pass' to a team-mate when that team-mate is running into the space where the pass has been aimed? 

I am genuinely curious.
I'm led to belive its a pass (any direction) of 25 yards or more, which i always thought it was too small. But whatever it is does give a indication of how a team play changes at times, be it 25/35/50 as long as it stays the same for all.
Thanks, MJG.  I got the definition by looking for quotes from the statisticians themselves about how they define things and this, believe it or not, was the only comment I could find and it was, indirectly (i.e. secondhand), a quote from Opstat.  My observation is precisely trying to identify how these measures are made, since  as you say - 'as long as it stays the same for all' is a hugely and overbearing function of the efficacy of analysis.


MJG

Quote from: toshes mate on January 20, 2020, 10:37:04 AM
Quote from: MJG on January 20, 2020, 10:30:37 AM
Quote from: toshes mate on January 20, 2020, 08:56:49 AM
The only definition I can find about the 'long ball' is that it is "a forward pass of 35 yards or more that is kicked into space rather than a precise pass aimed at a particular team-mate".  Is there a more precise definition than this since the word 'forward' is ambiguous as it makes a potential arc of only slightly less than 180 degrees, and, when does it become a 'precise pass' to a team-mate when that team-mate is running into the space where the pass has been aimed? 

I am genuinely curious.
I'm led to belive its a pass (any direction) of 25 yards or more, which i always thought it was too small. But whatever it is does give a indication of how a team play changes at times, be it 25/35/50 as long as it stays the same for all.
Thanks, MJG.  I got the definition by looking for quotes from the statisticians themselves about how they define things and this, believe it or not, was the only comment I could find and it was, indirectly (i.e. secondhand), a quote from Opstat.  My observation is precisely trying to identify how these measures are made, since  as you say - 'as long as it stays the same for all' is a hugely and overbearing function of the efficacy of analysis.
The change, be it very slight, has been Rodak going in goal. The season and graph can almost be split in two where he went in
Just the views of a long term fan

toshes mate

I was actually referring to the definition of a long ball (and for any other measure used in football data) since if that is inconsistent among the various companies allegedly collating the numbers of long balls then the measurement itself is in even more doubt than it would be from checker error alone.  To be trustworthy I would have thought all companies claiming to maintain in game statistics would have clear definitions about what each item is.  That seems not to be the case. 

MJG

Quote from: toshes mate on January 20, 2020, 12:08:11 PM
I was actually referring to the definition of a long ball (and for any other measure used in football data) since if that is inconsistent among the various companies allegedly collating the numbers of long balls then the measurement itself is in even more doubt than it would be from checker error alone.  To be trustworthy I would have thought all companies claiming to maintain in game statistics would have clear definitions about what each item is.  That seems not to be the case.
I'm 80% sure Opta do it at 25, but like you I would have thought they all stick to same size. Personally if I had to give a figure for 'long ball' or lumping it Id say 35 Yards should be starting point fo that, so I might have misread it.
Just the views of a long term fan


MJG

Championship Game 28 last 12 seasons

Table below shows how the top 6 stood at game 28 compared to where we are now.

Can see that top is below average but in fact within a point or the same as a third of the past 12 seasons.

2nd place is almost where teams usually are over that period, but against last 6 years is 2-3 points behind.

As a total number of points amassed by the top 6 this is the 9th highest against the last 12 years.

In essence about 50% of the time a team runs away at this point (59+)

Only one has a team in 2nd had more than 55.

Currently the gap between 2nd and 3rd is joint 2nd highest, but that could change if team wins game in hand.

Maybe a below average season is best way to describe how the table stands.

Just the views of a long term fan

Jim©

Thanks MJG, what's also interesting is how many teams in 3rd/4th/5th/6th managed to break into that top 2 by season end. I have a feeling that this one will go to the wire...

MJG

#188






]

Just the views of a long term fan


MJG

There are a couple of posters who like to know how accurate the system I use is and I'm always looking at ways to improve it and also show where its working or not.

As i always say it's a moving target. Since Game 7 2nd place has peaked at a target of 95 and is now at its lowest target of 86.
3rd for example over same period has been 87 at its highest and 80 at its lowest.

There will always be ranges and I consider it a success if I'm a point out either way.

The Top 2's form has been shocking really over last 8 games, 9 for Leeds and 7 for WBA. I doubt there has ever been a run at the same time where two teams fell off a cliff like this. But its good news for us.

I have had a look back at where my predictions showed the highest total needed. That was at game 21.

The table below shows what my system predicted to be the table now and how far I'm out in each position.

Across the whole table I was 1 point out. Actual points in the current table is 950 and I would have predicted 951

Top 2 and bottom are the areas I have not got it right, but overall fairly happy at this stage even with a few 2points out.





Just the views of a long term fan

toshes mate

The metrics you have used do show that in any statistical methodology it is the extremities that provide the greatest variance between predictions and outcomes. 

You would expect a considerable variance at the top end of the table since the points total is already larger than all that below it and the margin of error is always likely to be large.  The middle of the table should conform to reality since the closer you get to the statistical mean the more likely you are to hit a sweet spot.  What is more interesting is the variance at the bottom end because the points total must be lower than all that above it but not by any significant margin.  The current error should narrow at the bottom end and be much closer by season's end. 

In short your model seems to me to be a pretty good one that works as well as it could do for anyone trying to make a living out of predictions.  Find an answer to top end variance (which would also impact on the rest of the table) and you could have a gold mine.     

MJG

Quote from: toshes mate on January 29, 2020, 08:58:32 AM
The metrics you have used do show that in any statistical methodology it is the extremities that provide the greatest variance between predictions and outcomes. 

You would expect a considerable variance at the top end of the table since the points total is already larger than all that below it and the margin of error is always likely to be large.  The middle of the table should conform to reality since the closer you get to the statistical mean the more likely you are to hit a sweet spot.  What is more interesting is the variance at the bottom end because the points total must be lower than all that above it but not by any significant margin.  The current error should narrow at the bottom end and be much closer by season's end. 

In short your model seems to me to be a pretty good one that works as well as it could do for anyone trying to make a living out of predictions.  Find an answer to top end variance (which would also impact on the rest of the table) and you could have a gold mine.     
Cheers for that its appreciated.

I do have data which im trying to build in that shows the majority of the time the top half of the table scores less points than the bottom half in the send part of the season. If WBA or even Leeds played as a normal team thses last half a dozen games it would not be so bad.....but its good for us tho.
Just the views of a long term fan


MJG

Championship situation for @FulhamFC after game 29 over last 3 years in this division.

[/url]
Just the views of a long term fan

MJG

Championship top 6 @ 30 games.

A pretty average position for all of the league apart from top 2 where pretty much worst performing in last 10 years.

Just the views of a long term fan

MJG

Just the views of a long term fan


MJG

I was asked to do a points per game for the starters...
[/url]
Just the views of a long term fan

MJG

Number of points each position made in last 16 games

[/url]
Just the views of a long term fan

MJG

The championship with 16 games left and potential points for each team based on...

-being the best team for last 16 games (2pts per game)
-being the 6th best team for the run in
-Just being bang average

My current prediction for 6th is 76 points.

Just the views of a long term fan


MJG

Put simply we need to pretty much fall to 8/9 points every six games to fail to make playoffs.
To make 2nd we need to be the best team over next 16 games, do that and we make it.
Just the views of a long term fan

The Rational Fan

Quote from: MJG on February 05, 2020, 08:51:46 AM
Put simply we need to pretty much fall to 8/9 points every six games to fail to make playoffs.
To make 2nd we need to be the best team over next 16 games, do that and we make it.

Being the best team over next 16 games is not easy, as we have the hardest draw apart from Derby.