Points Tally Compared to Last Season

[Cocacolajojo1]

But the degree of competitiveness measured in draws could surely be lying if a large number of teams, say the bottom ten teams, drew (?) against each alot, no? I mean that calculation could surely be avoided by just comparing the Point-tally of the top teams to the last seasons?

[BlueinBosnia]

But the degree of competitiveness measured in draws could surely be lying if a large number of teams, say the bottom ten teams, drew (?) against each alot, no? I mean that calculation could surely be avoided by just comparing the Point-tally of the top teams to the last seasons?

Exactly the opposite, actually. A league is about everyone playing everyone, not the top teams playing everyone. If you want to compare just the top teams, then position is what you should be looking at (or points earned as a proportion of those in the league as a whole), and not straightforward points tallies.

Surprisingly, your idea that the bottom teams draw more is also wrong. This season, it's been the current middle-4 teams, with a distribution slightly skewed toward the lower half of the table (you'd probably be statistically most likely to have drawn the most games in 11th place), whereas last season, it was teams in 5th-8th positions, with a stronger skewness toward the top (probably 7th or possibly 8th position).

[Rag_hater]

I think the amount the Sheikh has spent a better points tally and the records we have broken was the least to expect.I wish people would drop this shit of where we have come from,at this moment in time we are rich,richer than anyone.

[Nigels Tackle]

But the degree of competitiveness measured in draws could surely be lying if a large number of teams, say the bottom ten teams, drew (?) against each alot, no? I mean that calculation could surely be avoided by just comparing the Point-tally of the top teams to the last seasons?

Exactly the opposite, actually. A league is about everyone playing everyone, not the top teams playing everyone. If you want to compare just the top teams, then position is what you should be looking at (or points earned as a proportion of those in the league as a whole), and not straightforward points tallies.

Surprisingly, your idea that the bottom teams draw more is also wrong. This season, it's been the current middle-4 teams, with a distribution slightly skewed toward the lower half of the table (you'd probably be statistically most likely to have drawn the most games in 11th place), whereas last season, it was teams in 5th-8th positions, with a stronger skewness toward the top (probably 7th or possibly 8th position).

couple of things...

have you allowed for the fact the 2012 is a leap year in your calcs?
presumably you've adjusted our points tally for the 'savic factor'?

[Mingchester Mingy]

Because every time a draw happens, somewhere in the league a point is lost.

This season, there are 4.4% more points 'floating around' the league table as a whole.

Therefore, for a team's standings to be accurately compared on points accrued, and not position, this has to be taken into account.

To do this, one must multiply this season's points by 0.956, or last season's by 1.044.

Obviously, this number will change from week-to-week, as the number of draws divided by the number of games played will change.

Nerd











But thanks. I will use it in the pub to make myself look clever.

LMAO....Me too mate, Me too !

[BlueinBosnia]

But the degree of competitiveness measured in draws could surely be lying if a large number of teams, say the bottom ten teams, drew (?) against each alot, no? I mean that calculation could surely be avoided by just comparing the Point-tally of the top teams to the last seasons?

Exactly the opposite, actually. A league is about everyone playing everyone, not the top teams playing everyone. If you want to compare just the top teams, then position is what you should be looking at (or points earned as a proportion of those in the league as a whole), and not straightforward points tallies.

Surprisingly, your idea that the bottom teams draw more is also wrong. This season, it's been the current middle-4 teams, with a distribution slightly skewed toward the lower half of the table (you'd probably be statistically most likely to have drawn the most games in 11th place), whereas last season, it was teams in 5th-8th positions, with a stronger skewness toward the top (probably 7th or possibly 8th position).

couple of things...

have you allowed for the fact the 2012 is a leap year in your calcs?
presumably you've adjusted our points tally for the 'savic factor'?

It's quite simple maths (nothing more challenging than you're taught in school at the age of 13/Year 9, just applied in a different scenario). Sorry I mastered basic literacy at a relatively young age and went on to do other things. I guess sarcasm would've gotten me further in life...

[Cocacolajojo1]

Exactly the opposite, actually. A league is about everyone playing everyone, not the top teams playing everyone. If you want to compare just the top teams, then position is what you should be looking at (or points earned as a proportion of those in the league as a whole), and not straightforward points tallies.

Surprisingly, your idea that the bottom teams draw more is also wrong. This season, it's been the current middle-4 teams, with a distribution slightly skewed toward the lower half of the table (you'd probably be statistically most likely to have drawn the most games in 11th place), whereas last season, it was teams in 5th-8th positions, with a stronger skewness toward the top (probably 7th or possibly 8th position).

Interesting. Ill reply in a few days when i get ahold of a,proper computer

[Ted Hughes]

Interesting concept & not without its merit.

[Cocacolajojo1]

Exactly the opposite, actually. A league is about everyone playing everyone, not the top teams playing everyone. If you want to compare just the top teams, then position is what you should be looking at (or points earned as a proportion of those in the league as a whole), and not straightforward points tallies.

Surprisingly, your idea that the bottom teams draw more is also wrong. This season, it's been the current middle-4 teams, with a distribution slightly skewed toward the lower half of the table (you'd probably be statistically most likely to have drawn the most games in 11th place), whereas last season, it was teams in 5th-8th positions, with a stronger skewness toward the top (probably 7th or possibly 8th position).

Interesting. Ill reply in a few days when i get ahold of a,proper computer

Well it took me some time to answer this, as I was thrown of the flight from glasgow just a couple of days later and forgot about until a couple of weeks ago. I've been mulling it over since and have actually tested your idea.

If i understand it correctly, as it says on the first page of this thread, your idea is that the more even a league is - evenness being measured in the amount of points lost through draws - the harder it is to get a high points tally. Or rather, when comparing the points tally of say City in different seasons, one has to account for the evennness of different seasons. For example, 91 points accrued during a an easy season could actually be less impressive than 75 points during a difficult season. While I agree with your general point, as I think anyone would, I think your idea of how to measure it is somewhat crude.

Well I tested some scenarios on a hypothetical league containing 10 teams to make it simple. For an easier illustration I've calculated the average points tally per team rather than points lost. I think it measures the same thing but the other way around.

Scenario 1: All matches are draws. Each teams get 18 points. Fair enough. The average tally is 18 points per team.

Scenario 2: Each team wins 6 games and draw 6 and then lose 6. Each team gets 24 points. The league is just as competitive as in scenario one but there are less points lost, or rather, the average point tally per team is much is a third higher, 24 points per team.

Scenario 3: All teams win their home games but lose their away games. Each team gets 27 points. The league is still as tight as in the above scenarios but the average tally is now 50 % higher than in scenario 1. See what I'm getting at? The amount of points accrued or lost is a crude instrument that does not account for these variations. It measures something, it even measures evenness to a certain extent, but is highly inaccurate without doing some sort of stratification of the league and counting points gained or lost between different layers of teams.

Scenario 4: This crudeness is illustrate perfectly by a league that's similar to the spanish league. One team wins all its matches while the other teams lose to their games against the first team and then draws the rest of their games. The winning team ends up with 54 points and the rest with 16 points each. The average point's tally is 19,8 which is slightly above the tally in scenario 1 but less than scenario 2 and 3. Sure, the league is still somewhat even in scenario 4 but it is not as even as in scenario 2 and 3, which the average points tally would suggest.

Scenario 5. The top five teams win all their games against the bottom five and draw against each other. The bottom five all draw against each other. The top five all end up with 38 points each while the bottom five end up with 8 each. Average points tally = 23. The league in this scenario is slightly more even than scenario 2 and more even than scenario 3 measured in average points yield but it is in fact less even. The league is divided in two portions. It is however more even than the league in scenario 4, at least when looking at the competition for first place. Yet, the points average suggests otherwise which in turn suggests the points tally, or as you would say; lost points, is somewhat flawed.

I don't have a solution. Sorry. I think a way to go would be, as mentioned above, to split a league into different layers and then measure the amount of points that travel between the layers through draws and wins. Perhaps I've calculated wrong and your idea still works. Perhaps this was just something you thought of in the spur of the moment and then never though about again. Whatever it was, you've managed to fuck with my brain and made me waste time when calculating this.

[john68]

I know it was a long time since I was in year 9 but when I was a lad;

ball + head + kick + fucking hard + momentum = goal x as many times as possible...divided by tackle or save + clearance x more times than opposition + run like fuck = goals for - goals against = points x more than nearest rivals = LEAGUE CHAMPIONS.

Much simpler than that shite you just dug up on yer trowel Andrew.

Uni cunts....ruining football.

[Socrates]

I know it was a long time since I was in year 9 but when I was a lad;

ball + head + kick + fucking hard + momentum = goal x as many times as possible...divided by tackle or save + clearance x more times than opposition + run like fuck = goals for - goals against = points x more than nearest rivals = LEAGUE CHAMPIONS.

Much simpler than that shite you just dug up on yer trowel Andrew.

Uni cunts....ruining football.

JOHN? That goatfeeding fucker has hijacked your account!

[Cocacolajojo1]

I know it was a long time since I was in year 9 but when I was a lad;

ball + head + kick + fucking hard + momentum = goal x as many times as possible...divided by tackle or save + clearance x more times than opposition + run like fuck = goals for - goals against = points x more than nearest rivals = LEAGUE CHAMPIONS.

Much simpler than that shite you just dug up on yer trowel Andrew.

Uni cunts....ruining football.

You didn't add rain to your equation. Tsk tsk tsk.