Pamplemousse Posted 27 October, 2014 Share Posted 27 October, 2014 Below is the average number of points of the opposition each team has faced so far Manchester City 13.556 Aston Villa 13.444 Tottenham Hotspur 13.000 Burnley 12.556 Queens Park Rangers 12.556 Liverpool 12.556 Leicester City 12.444 Crystal Palace 12.222 Hull City 12.222 Newcastle United 12.222 Arsenal 12.111 Everton 11.889 Swansea City 11.778 Sunderland 11.667 West Ham United 11.667 Stoke City 11.556 West Bromwich Albion 11.556 Chelsea 11.444 Manchester United 11.333 Southampton 11.222 So it does confirm that we have had the easiest run in, but it's interesting to note that Man Utd have had it just as easy as us. Here's the average number of points of the opposition for each team's next 5 league games Queens Park Rangers 14.600 West Bromwich Albion 14.400 Sunderland 14.000 Tottenham Hotspur 13.400 Liverpool 12.400 Manchester United 12.400 Manchester City 12.200 Southampton 12.200 Arsenal 12.000 Everton 12.000 Aston Villa 11.800 Crystal Palace 11.800 Hull City 11.800 Swansea City 11.800 Leicester City 11.600 Stoke City 11.600 Burney 11.200 West Ham 10.600 Newcastle United 10.200 Chelsea 10.000 It'll be interesting to see how we fare in our next 5 games in comparison to Man City, who theoretically have the same level of difficulty in their next 5 matches as we do. Whereas Chelsea have it extremely easy in their next matches. Link to comment Share on other sites More sharing options...
musesaint Posted 27 October, 2014 Share Posted 27 October, 2014 Interesting stats but in reality not that much in it overall? As saints gain confidence I don't see the opposition over the next month or so to be unbeatable Link to comment Share on other sites More sharing options...
sandwichsaint Posted 27 October, 2014 Share Posted 27 October, 2014 Good grief, who worked all this out? You can only beat what's put in front of you. I'm driving over 400 miles on Weds night to cheer Saints on in small cup. Newsflash, the season is only 9 games old, if you win 6 out of 9 you will degrade your opponents score quite significantly. Enjoy it while we can, second in the not-EPL on merit and marching on Wem-blee. Where's Ottery these days, always a good poster and I've a very positive mental image of his Daughter! COYRs. Link to comment Share on other sites More sharing options...
sentry Posted 27 October, 2014 Share Posted 27 October, 2014 But arnt their average points lower cause we beat them? Look, we all know the top 3 at the mo (Ars, City, Chelsea) , then there's the other big clubs (Tot, Man U, Everton, Liverpool) We have (in the league ) only played two out of that lot, but we have also played WHU who are doing well, Stoke who beat City, and we still have Leicester and Burley x2 games. Anything is possible and Im hoping for top 6. It would be great to knock a few of those clubs out the toop 4 or 6. Everton are ok with me, but the rest I would love to see fail even for WHU. Link to comment Share on other sites More sharing options...
hypochondriac Posted 27 October, 2014 Share Posted 27 October, 2014 But arnt their average points lower cause we beat them? Look, we all know the top 3 at the mo (Ars, City, Chelsea) , then there's the other big clubs (Tot, Man U, Everton, Liverpool) We have (in the league ) only played two out of that lot, but we have also played WHU who are doing well, Stoke who beat City, and we still have Leicester and Burley x2 games. Anything is possible and Im hoping for top 6. It would be great to knock a few of those clubs out the toop 4 or 6. Everton are ok with me, but the rest I would love to see fail even for WHU. And villa twice who have looked turd. Link to comment Share on other sites More sharing options...
whelk Posted 28 October, 2014 Share Posted 28 October, 2014 We've also hammered resurgent West Ham, Newcastle and..........QPR. Link to comment Share on other sites More sharing options...
Redslo Posted 28 October, 2014 Share Posted 28 October, 2014 (edited) Below is the average number of points of the opposition each team has faced so far Manchester City 13.556 Aston Villa 13.444 Tottenham Hotspur 13.000 Burnley 12.556 Queens Park Rangers 12.556 Liverpool 12.556 Leicester City 12.444 Crystal Palace 12.222 Hull City 12.222 Newcastle United 12.222 Arsenal 12.111 Everton 11.889 Swansea City 11.778 Sunderland 11.667 West Ham United 11.667 Stoke City 11.556 West Bromwich Albion 11.556 Chelsea 11.444 Manchester United 11.333 Southampton 11.222 So it does confirm that we have had the easiest run in, but it's interesting to note that Man Utd have had it just as easy as us. Here's the average number of points of the opposition for each team's next 5 league games Queens Park Rangers 14.600 West Bromwich Albion 14.400 Sunderland 14.000 Tottenham Hotspur 13.400 Liverpool 12.400 Manchester United 12.400 Manchester City 12.200 Southampton 12.200 Arsenal 12.000 Everton 12.000 Aston Villa 11.800 Crystal Palace 11.800 Hull City 11.800 Swansea City 11.800 Leicester City 11.600 Stoke City 11.600 Burney 11.200 West Ham 10.600 Newcastle United 10.200 Chelsea 10.000 It'll be interesting to see how we fare in our next 5 games in comparison to Man City, who theoretically have the same level of difficulty in their next 5 matches as we do. Whereas Chelsea have it extremely easy in their next matches. I believe your methodology is flawed. Part (but not all) of the reason we appear to have faced an easy schedule is that our opponents have had to play us. Sunderland's schedule looks tougher because they opponents got to play them. If someone who understands statistics wants to do some sort of regression analysis the result would be more meaningful. Of course, your look ahead doesn't suffer from that flaw. Edited 28 October, 2014 by Redslo Link to comment Share on other sites More sharing options...
Joneth Posted 28 October, 2014 Share Posted 28 October, 2014 I believe your methodology is flawed. Part (but not all) of the reason we appear to have faced an easy schedule is that our opponents have had to play us. Sunderland's schedule looks tougher because they opponents got to play them. If someone who understands statistics wants to do some sort of regression analysis the result would be more meaningful. Of course, your look ahead doesn't suffer from that flaw. Surely you just disregard the match involving the team in question? Link to comment Share on other sites More sharing options...
angelman Posted 28 October, 2014 Share Posted 28 October, 2014 I believe your methodology is flawed. Part (but not all) of the reason we appear to have faced an easy schedule is that our opponents have had to play us. Sunderland's schedule looks tougher because they opponents got to play them. If someone who understands statistics wants to do some sort of regression analysis the result would be more meaningful. Of course, your look ahead doesn't suffer from that flaw. I've done Spearman's Rank Correlation Coefficient to see whether ease of opponent and table position have any correlation. Ranked Saints as 1st with ease, Man U 2nd, etc etc and got a Correlation Coefficient of + 0.395 The results are always between -1 and +1. The closer to -1 you get, means that there is a negative correlation. At 0, it means there is NO correlation, and the closer to +1 you get, means there is a positive correlation. So while the hypothesis about ease of players and league position, has a slight positive correlation (ie is a little bit true), it isn't very strong (ie a little tenuous). Overall, I say that unless you have a positive correlation of over +0.5, and in reality more like +0.7, then the theory falls down. Link to comment Share on other sites More sharing options...
Charlie Wayman Posted 28 October, 2014 Share Posted 28 October, 2014 Wonder why the OP doesn't scoop the pools jackpot every week? How many scores has he predicted so far? Link to comment Share on other sites More sharing options...
Griffo Posted 28 October, 2014 Share Posted 28 October, 2014 I've done Spearman's Rank Correlation Coefficient to see whether ease of opponent and table position have any correlation. Ranked Saints as 1st with ease, Man U 2nd, etc etc and got a Correlation Coefficient of + 0.395 The results are always between -1 and +1. The closer to -1 you get, means that there is a negative correlation. At 0, it means there is NO correlation, and the closer to +1 you get, means there is a positive correlation. So while the hypothesis about ease of players and league position, has a slight positive correlation (ie is a little bit true), it isn't very strong (ie a little tenuous). Overall, I say that unless you have a positive correlation of over +0.5, and in reality more like +0.7, then the theory falls down. F*cking hell. Link to comment Share on other sites More sharing options...
Saint_Tony Posted 28 October, 2014 Share Posted 28 October, 2014 A few of these teams totals are influenced by the fact we have already beaten them, don't forget that. Yet another pointless statisticsl exercise. We"ve got people trying to show how we shouldn't be 2nd, and others trying to make us top. Think I might stick to the real premier league table, or perhaps start a new thread on it Link to comment Share on other sites More sharing options...
Saint IQ Posted 28 October, 2014 Share Posted 28 October, 2014 Making the simple game complicated springs to mind Link to comment Share on other sites More sharing options...
scotty Posted 28 October, 2014 Share Posted 28 October, 2014 I've done Spearman's Rank Correlation Coefficient to see whether ease of opponent and table position have any correlation. Ranked Saints as 1st with ease, Man U 2nd, etc etc and got a Correlation Coefficient of + 0.395 The results are always between -1 and +1. The closer to -1 you get, means that there is a negative correlation. At 0, it means there is NO correlation, and the closer to +1 you get, means there is a positive correlation. So while the hypothesis about ease of players and league position, has a slight positive correlation (ie is a little bit true), it isn't very strong (ie a little tenuous). Overall, I say that unless you have a positive correlation of over +0.5, and in reality more like +0.7, then the theory falls down. What in the actual f*ck??? Link to comment Share on other sites More sharing options...
eurosaint Posted 28 October, 2014 Share Posted 28 October, 2014 " A mathematician called Hall, had a hexacadromical ball, the square of it's weight was two thirds of five eights, of two tenths, of four ninths of f*ck all " !! I rest my case Link to comment Share on other sites More sharing options...
angelman Posted 28 October, 2014 Share Posted 28 October, 2014 What in the actual f*ck??? Oh come on, it's easy. Basic maths. Shows that, I guess in layman's terms, that the OP's theory is 60.5% bollix. Link to comment Share on other sites More sharing options...
Barry the Badger Posted 28 October, 2014 Share Posted 28 October, 2014 Well I thought it was interesting. It's not black and white of course but it is interesting nonetheless and does confirm that we've perhaps had a slightly easier starts, although I knew that just from reading the fixture list! Link to comment Share on other sites More sharing options...
Patrick Bateman Posted 28 October, 2014 Share Posted 28 October, 2014 Some serious over analysis on this forum. Part of me wants to but part of me is entertained by how absolutely stark raving bonkers you all are! Link to comment Share on other sites More sharing options...
simondo Posted 28 October, 2014 Share Posted 28 October, 2014 I've done Spearman's Rank Correlation Coefficient to see whether ease of opponent and table position have any correlation. Ranked Saints as 1st with ease, Man U 2nd, etc etc and got a Correlation Coefficient of + 0.395 The results are always between -1 and +1. The closer to -1 you get, means that there is a negative correlation. At 0, it means there is NO correlation, and the closer to +1 you get, means there is a positive correlation. So while the hypothesis about ease of players and league position, has a slight positive correlation (ie is a little bit true), it isn't very strong (ie a little tenuous). Overall, I say that unless you have a positive correlation of over +0.5, and in reality more like +0.7, then the theory falls down. what this means is: you can legitimately make statistics say whatever you want, depending on the answer you are looking for... Link to comment Share on other sites More sharing options...
angelman Posted 28 October, 2014 Share Posted 28 October, 2014 what this means is: you can legitimately make statistics say whatever you want, depending on the answer you are looking for... LOL. I would normally agree with you, but this one actually can't really fudge the answer. It just shows whether two sets of figures are tied in together and how much of a connection there is between them. I think that I best get my coat and get a life! Link to comment Share on other sites More sharing options...
trousers Posted 28 October, 2014 Share Posted 28 October, 2014 Link to comment Share on other sites More sharing options...
Doctoroncall Posted 28 October, 2014 Share Posted 28 October, 2014 what this means is: you can legitimately make statistics say whatever you want, depending on the answer you are looking for... That's the bias, it's why you need to ask the question first rather than start with the answer! Link to comment Share on other sites More sharing options...
Redslo Posted 28 October, 2014 Share Posted 28 October, 2014 Some serious over analysis on this forum. Part of me wants to but part of me is entertained by how absolutely stark raving bonkers you all are! You should read the things people write about Baseball. Link to comment Share on other sites More sharing options...
nta786 Posted 28 October, 2014 Share Posted 28 October, 2014 I've done Spearman's Rank Correlation Coefficient to see whether ease of opponent and table position have any correlation. Ranked Saints as 1st with ease, Man U 2nd, etc etc and got a Correlation Coefficient of + 0.395 The results are always between -1 and +1. The closer to -1 you get, means that there is a negative correlation. At 0, it means there is NO correlation, and the closer to +1 you get, means there is a positive correlation. So while the hypothesis about ease of players and league position, has a slight positive correlation (ie is a little bit true), it isn't very strong (ie a little tenuous). Overall, I say that unless you have a positive correlation of over +0.5, and in reality more like +0.7, then the theory falls down. some people need to get out more. then again, i am typing on an internet forum! Link to comment Share on other sites More sharing options...
cellone Posted 28 October, 2014 Share Posted 28 October, 2014 I love it when people geek out. Not capable myself so find it interesting. Link to comment Share on other sites More sharing options...
The9 Posted 28 October, 2014 Share Posted 28 October, 2014 I believe your methodology is flawed. Part (but not all) of the reason we appear to have faced an easy schedule is that our opponents have had to play us. Sunderland's schedule looks tougher because they opponents got to play them. If someone who understands statistics wants to do some sort of regression analysis the result would be more meaningful. Of course, your look ahead doesn't suffer from that flaw. There's an element of truth to this, but Man City having the most difficult opponents and still being in the top 3 somewhat overshadows it. Link to comment Share on other sites More sharing options...
Deano6 Posted 28 October, 2014 Share Posted 28 October, 2014 I've done Spearman's Rank Correlation Coefficient to see whether ease of opponent and table position have any correlation. Ranked Saints as 1st with ease, Man U 2nd, etc etc and got a Correlation Coefficient of + 0.395 The results are always between -1 and +1. The closer to -1 you get, means that there is a negative correlation. At 0, it means there is NO correlation, and the closer to +1 you get, means there is a positive correlation. So while the hypothesis about ease of players and league position, has a slight positive correlation (ie is a little bit true), it isn't very strong (ie a little tenuous). Overall, I say that unless you have a positive correlation of over +0.5, and in reality more like +0.7, then the theory falls down. How about calculating Pearson's rho on it, or Kendall's tau? +0.4 is pretty significant, HTH. It's not a question of needing to be 100% correlated to have merit. Plenty of things have small positive correlation, such as the number of people who post about statistics vs the number of people who understand the subject. Link to comment Share on other sites More sharing options...
cellone Posted 28 October, 2014 Share Posted 28 October, 2014 Geek off! Link to comment Share on other sites More sharing options...
angelman Posted 28 October, 2014 Share Posted 28 October, 2014 (edited) I disagree that +0.4 is significant, but that is opinion. But I am curious why you would use Pearson's r? I thought Spearman was more appropriate as we were dealing with ranked variables, but do tell why you think it mightn't be. I am sure that you could have used Kendall's tau, but as I was taught by Spearman's grand son, I rather favour him, as well as it being the more widely used. And also, let's face it, Spearman is pretty damn easy to calculate. Edited 28 October, 2014 by angelman Link to comment Share on other sites More sharing options...
cellone Posted 28 October, 2014 Share Posted 28 October, 2014 It's on now. Link to comment Share on other sites More sharing options...
SarniaSaint Posted 28 October, 2014 Share Posted 28 October, 2014 You should read the things people write about Baseball. LOL........ but the MLB season is 162 games so most of the stats do make sense. Hey this is a soccer (LOL) forum so wont go into the percentage game........ Link to comment Share on other sites More sharing options...
ericb Posted 28 October, 2014 Share Posted 28 October, 2014 jesus christ some of you lot really need to get a life, no wonder Stoke fans think we've got a lot of new fans, do you also work out the relative probability of getting laid on a night out and never actually enjoy life for what it is too!? Link to comment Share on other sites More sharing options...
sadoldgit Posted 28 October, 2014 Share Posted 28 October, 2014 Wibble Link to comment Share on other sites More sharing options...
angelman Posted 28 October, 2014 Share Posted 28 October, 2014 jesus christ some of you lot really need to get a life, no wonder Stoke fans think we've got a lot of new fans, do you also work out the relative probability of getting laid on a night out and never actually enjoy life for what it is too!? I'm married (with kids), so well know the relative probability of getting laid. Link to comment Share on other sites More sharing options...
cellone Posted 28 October, 2014 Share Posted 28 October, 2014 jesus christ some of you lot really need to get a life, no wonder Stoke fans think we've got a lot of new fans, do you also work out the relative probability of getting laid on a night out and never actually enjoy life for what it is too!? But they enjoy it Bud. Link to comment Share on other sites More sharing options...
whelk Posted 29 October, 2014 Share Posted 29 October, 2014 How about calculating Pearson's rho on it, or Kendall's tau? +0.4 is pretty significant, HTH. It's not a question of needing to be 100% correlated to have merit. Plenty of things have small positive correlation, such as the number of people who post about statistics vs the number of people who understand the subject. Surely worthy of a poll? Link to comment Share on other sites More sharing options...
whelk Posted 29 October, 2014 Share Posted 29 October, 2014 But they enjoy it Bud. Indeed. Ironic posters spouting get a life. Absolute bs though if you ask me but nonetheless valid discussion point for a forum I am with Danny Baker's view: Why is everyone giving so much credence to the game's statistics now? Surely now we can finally see 'stats" are train spotting bull****. Link to comment Share on other sites More sharing options...
lifeintheslowlane Posted 29 October, 2014 Share Posted 29 October, 2014 I don't know if this has been said before but 93% of all statistics are made up. Link to comment Share on other sites More sharing options...
Recommended Posts
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 accountSign in
Already have an account? Sign in here.
Sign In Now