SuperCoach Scores Twitter SuperCoach Scores Facebook
Results 1 to 3 of 3
  1. #1
    Moderator
    Join Date: 13-09-2014
    AFL Club: North Melbourne
    Posts: 6,442
    Likes: 8,130
    Rep Power: 12419
    Moderator


    4 Not allowed!

    Quote Originally Posted by IDIG View Post
    Barely a clean sheet to be seen and what appears to be more penalties awarded than I can remember???

    I've got Heaton who's done his job and will probably hold his spot in the short term as no one else really appeals apart from Cech with our fixtures. CS are few and far between for us though!
    Even before IDIG’s post above, I had already embarked on an investigation of looking back at the last EPL season stats, with a particular focus on analysing respective team’s CS prospects. The intention was to hopefully identify some useful CS trends through my analysis and somehow come up with a ‘CS predictor or probability’ algorithm to help guide us with our players’ selections when frisking through the upcoming EPL fixtures.

    Before I proceed any further, let me stress that what I’m about to share has no direct impact on predicting individual players’ scores and is not based on the RDT scoring system as outlined on the Sportsdeck website. This is purely and exclusively based on my own understandings and interpretations of the data available. The ‘clean sheet FOR’ (CSF%) and ‘clean sheet AGAINST’ (CSA%) uses a complex formula which takes into account a number of variables to predict a team % likelihood of achieving a clean sheet, either FOR or AGAINST. I’m not even going to attempt to publish or explain this formula here so you all will just have to trust me on this one!

    Final EPL Ladder

    In what was widely described as ‘the greatest sporting upset ever’, the Foxes were crowned champions of the Premier League, their first top-level football championship since the club was founded in 1884…some 132 years ago!

    The final table below showed that they got home comfortably in the end, 10 points clear of their nearest rival in Arsenal with Tottenham Hotspur a further point behind.

    Name:  2015-16 Ladder.jpg
Views: 20
Size:  172.1 KB

    Can we draw any correlation from respective team’s final ladder position to their probability of achieving a CS?

    Breaking the Fixtures into CS Recorded

    I’ve dissected the results of the fixtures to highlight CS achieved which is graphically shown below, including Home & Away goals for the corresponding GW fixtures. Teams achieved a total of 215 CS and slotted 1,026 goals through the course of the season, resulting in a GW average of 5.7 CS and 27 goals.

    Name:  2015-16 CS Fixture.jpg
Views: 22
Size:  260.7 KB

    Please refer to the LEGENDS table for key explanations to the chart. Note that for NIL:NIL drawn games, both ‘Home’ and ‘Away’ teams are deemed to have achieved a CS. As a result, there were a total of 64 CS for 32 games which ended in NIL:NIL draw.

    There was a total of 91 CS achieved by the ‘Home’ teams and 60 CS achieved by the ‘Away’ teams. Arsenal and Manchester United lead the way for ‘Home’ CS with 8, whilst Arsenal, Leceister City & Manchester City all recorded 6 ‘Away’ CS. Should we be surprised that all these teams finished in the Top 5 by season’s end? Conversely, Aston Villa conceded the most CS at ‘Home’ with 7 and Norwich City the most ‘Away’ CS with an astounding11! Needless to say, both these teams are no longer in the Premier League. For completeness, the third relegated team in Newcastle United only managed a total of 5 CS whilst conceding 11 with 8 of them as the ‘Away’ team.

    Can the CS data be a useful predictor for where teams are likely to finish in the overall standing?

    CS Ladder

    I’ve shown you the ‘Final EPL Ladder’ and what the overall fixtures looks like from a CS perspective. I now present you the result of my analysis and findings as revealed in my ‘CS Ladder’, a re-vamp of the ‘Final EPL Ladder’.

    Name:  2015-16 CS Ladder.jpg
Views: 21
Size:  239.7 KB

    Again, please refer to the LEGENDS table for key explanations. Note that for the purpose of this exercise, 0:0 draws are counted as CS for both teams involved. To highlight this fact, both Leicester City and West Ham United recorded 2 & 1 ‘Home’ CSA respectively, but did not concede any goals under the GA column because they were all part of 0:0 drawn games.

    I’ve decided to keep the teams’ final ladder position unchanged in the CS Ladder to make it easier to identify any trends which may be apparent. As an example, the CSD seems to correlate extremely well with teams’ overall ladder position. This is only representative of one season’s worth of data, so I would need to see this replicated over many seasons in order to be confident in it being a useful ladder ‘predictor’ guide.

    As I mentioned earlier, I’m not going into details on how the CSF% and CSA% are calculated, let’s just say that there were too many iterations before I finally settled on this one. Hopefully this will become more apparent when I come to provide progressive tables update for the current season.

    In summary, there was a relative 28% chance ‘Overall’ for a team to record a CS with a ‘Home’ CS being more probable than an ‘Away’ CS at 32% and 24% respectively, a NIL:NIL drawn game has an 8% probability of occurring. Of the 215 CS recorded, a total of 292 goals were scored in these games, an ‘Overall’ average of 1.36 goals per game. Again, the odds favoured the ‘Home’ CS with an average of 1.47 versus 1.21 for the ‘Away’ CS games.

    So which teams are likely to score heavily in CSF games and which teams are more prone to conceding in CSA games?

    The following 2 tables summarises the scoring frequencies for such games, the first showing the frequency of scores in CSF games followed by the frequency of scoring in CSA games.

    Name:  2015-16 CS FOR Team.jpg
Views: 22
Size:  62.7 KB


    Name:  2015-16 CS AGAINST Team.jpg
Views: 20
Size:  66.6 KB

    From these tables we can see that of the 11 CSF games (excluding 0:0 games), Manchester City recorded scores of 3 or more goals 7 times, followed by Arsenal and Tottenham Hotspurs, each with 5 from 14 & 10 CSF games respectively. Swansea City only CSF ‘Away’ was a 0:0 draw in GW19 against Crystal Palace. They are also amongst teams with the lowest total CSF games with only 6 with all bar 1 where they managed to score more than a goal.

    At the other end of the spectrum, Norwich City conceded 3 or more goals in 6 of its 15 CSA games. Aston Villa and Stoke City both recorded 5 games of conceding 3 or more goals from 15 & 10 CSA games respectively.

    Arsenal, Leceister City, Manchester City, Tottenham Hotspur & West Ham United all represented difficult fixtures for opposing teams to score a CS, with each conceding less than 3 CSA for the season. Leicester City and Tottenham Hotspur were the pick of the group with 1 & 3 CSA respectively and all were 1:0 losses.

    As I alluded in my opening statement, these analyses are not based on any ‘fantasy’ related statistics, just my own interpretation of data arisen from personal interest and intrigue. It may not directly provide you with information on which players to trade, but hopefully will at least point you towards which teams you should be considering and/or avoiding based on their fixtures. As we are all too well aware, the question of owning or not owning a player is one part of the equation, catching them at the start of a hot streak or vice versa, is akin to sorting the wheat from the chaff. THIS is ultimately what potentially make or break our season.

    I will continue to record data for the current season with a view of providing updated tables ONLY at regular interval, probably 3-4 times through the course of the season. I’m interested to monitor trends which have been identified through analysis of the 2015-16 EPL season and assess how the CSF% and CSA% ‘predictor’ performing and whether it needs any further tweaking.

    With the apparent lack of ‘useful’ data out there to help enhance our EPL fantasy experience, I’m relatively happy with the outcome of this investigation, despite the labourious nature of the exercise. I hope that you are able to take something away yourself and come to your own conclusion. If not, I hope that at least you’ve found it to be an interesting way to past some time.

    As always, happy to answer any questions or clarify anything…just don’t ask me about the CSF% or CSA% formula!
    SuperCoach:  15,799 (2016)  1,033 (2015)  6,367 (2014)  8,442 (2013)  AFL Dream Team:  2,706 (2016) 

    BBL|05 Major Prize Winner
    BBL|06 Member of #1 Ranked League

  2. #2
    Moderator
    Join Date: 08-03-2012
    AFL Club: Essendon
    Posts: 23,743
    Likes: 11,820
    Rep Power: 9241
    Moderator


    1 Not allowed!

    Wow mate what a piece of work! Took me a while to digest it all (welll try to) but don't think i'm quite understanding your definitions correctly.

    In the CS Ladder, for Leicester's home games. It shows 9 CSF, 2 CSA, 13 goals for, 0 goals against. So they have given up 2 clean sheets but not conceded a goal? Is that right?

    Also, as an EPL betting man i reckon there can be some good info taken out of this when looking at scores to zero, and to keep a clean sheet market! Surprised Arsenal kept so many clean sheets AND played so many 0:0 draws lol

    Edit: the 2 CSA were 0:0's
    Last edited by IDIG; 30-09-2016 at 8:54pm.
    SuperCoach:  1,895 (2017)  4,853 (2016)  5,202 (2015)  479 (2014)  7,684 (2013) 


  3. #3
    Moderator
    Join Date: 13-09-2014
    AFL Club: North Melbourne
    Posts: 6,442
    Likes: 8,130
    Rep Power: 12419
    Moderator


    0 Not allowed!

    Quote Originally Posted by IDIG View Post
    Wow mate what a piece of work! Took me a while to digest it all (welll try to) but don't think i'm quite understanding your definitions correctly.

    In the CS Ladder, for Leicester's home games. It shows 9 CSF, 2 CSA, 13 goals for, 0 goals against. So they have given up 2 clean sheets but not conceded a goal? Is that right?

    Also, as an EPL betting man i reckon there can be some good info taken out of this when looking at scores to zero, and to keep a clean sheet market! Surprised Arsenal kept so many clean sheets AND played so many 0:0 draws lol

    Edit: the 2 CSA were 0:0's
    That's correct...feel free to ask any questions and I'll try to explain.

    Great minds do really think alike as that was exactly my thought re EPL betting. I don't usually bet on scores over/under but I might just use these data to try my luck on a few multis.

    Hmmm...maybe I should consider taking royalties for winnings this becomes useful in increasing your chances of beating the odds!
    SuperCoach:  15,799 (2016)  1,033 (2015)  6,367 (2014)  8,442 (2013)  AFL Dream Team:  2,706 (2016) 

    BBL|05 Major Prize Winner
    BBL|06 Member of #1 Ranked League



Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •