The Alternative Premier League

tommyhaas

I've been posting this in the Liverpool thread the past few weeks, but its as relevant for any of the sides chasing a top 4 spot (And yes, Newcastle are included) or the title. Its from a blog of the same title by a sports scientist (or something)

My motivation for starting this blog is based on my feeling that league tables don’t tell the full picture during a season as they don’t take into account the teams played so far neither do they illustrate whether the fixtures were home or away. Over the course of this season, I will be updating this blog to apply a simple model that I use to evaluate league standings based on the matches played. I first started looking at league tables in this way about four years ago, by adapting the old theory that in order to win the league, a team needs to win all of their home games and draw all of their away games.

There are a few assumptions I have to make to keep this model as simple as possible. Firstly, I assume that 90 points should be a target points total for a title challenge. You can't control league position entirely yourself, so focussing on points makes more sense to me. If a team earns 90 points this season and doesn’t win the league, it would just mean another team has done exceptionally well.

In order to achieve 90 points, I make the assumption that a team needs to win every home game (57 points) and their 7 'easiest'* away games (21 points), drawing the remaining 12 away games (12 points). This means that a draw for Liverpool away at Old Trafford is a par** result for a title winning team, with a win there being 2-over-par, but a draw at home to Norwich is 2-under-par. During the season (as with during a round of golf to use that analogy further) it is unlikely for you to play to par in every match (hole), but any dropped points need to be made up elsewhere.

* I assume the 7 'easiest' away games are against the teams that finished between 17th and 14th last season and the three promoted teams as I need to determine my par scores before the season starts (I appreciate this isn't ideal, but I have to start somewhere).
** If a team is on par after 38 games, they will hit 90 points, if they are over par after 38 games, they will have more than 90 points (unlike golf, in this model, over par is good, under par is bad).

The Alternative table thus far



While it doesn't take into account factors such as the positive/negative affects of form or the pressure of being top, I still think it gives a good indication of the performance of sides thus far, and their chances from here on in

http://rawkprof.blogspot.com/2011/08/part-1-introduction.html

everdead.ie

tommyhaas wrote: »

I've been posting this in the Liverpool thread the past few weeks, but its as relevant for any of the sides chasing a top 4 spot (And yes, Newcastle are included) or the title. Its from a blog of the same title by a sports scientist (or something)



The Alternative table thus far



While it doesn't take into account factors such as the positive/negative affects of form or the pressure of being top, I still think it gives a good indication of the performance of sides thus far, and their chances from here on in

http://rawkprof.blogspot.com/2011/08/part-1-introduction.html

So currently everyone but Arsenal are on track to score over 71 points and City and United look like breaking the 90 point barrier?

tommyhaas

everdead.ie wrote: »

So currently everyone but Arsenal are on track to score over 71 points and City and United look like breaking the 90 point barrier?

If they continue to achieve the required results, which they have so far, yes. However, as they all have to play one another at least once more, its extremely unlikely

The table isn't suggesting that each bar Arsenal will finish with over 71, it just demonstrates what is required to do so

Dotrel

Well it puts Liverpool in for a 3rd place finish so you already know it doesn't work.

tommyhaas

Dotrel wrote: »

Well it puts Liverpool in for a 3rd place finish so you already know it doesn't work.

No it doesn't. It demonstrates that if Liverpool were to go from now until the end of the season achieving par (which is very unlikely), and everyone else was to do likewise (even more unlikely), then Liverpool would finish third. That's not a suggestion, its fact

Its equally applicable to any other side

Dotrel

tommyhaas wrote: »

No it doesn't. It demonstrates that if Liverpool were to go from now until the end of the season achieving par (which is very unlikely), and everyone else was to do likewise (even more unlikely), then Liverpool would finish third. That's not a suggestion, its fact

Its equally applicable to any other side

Fair enough. But I think considering the author wasn't happy with the information the standard league table presents he hasn't really come up with much of an alternative.

It's basically a graph that tells us how teams are doing compared to some hypothetical satisfactory result vs how a team performed last season.

If you were to simplify it and used it in Irelands world cup group that graph might tell us that Ireland (with 2x draws) got "-2" in results vs Slovakia, when in reality when the table was finished we'd actually achieved "-4". So basically it has told you nothing.

tommyhaas

Dotrel wrote: »

Fair enough. But I think considering the author wasn't happy with the information the standard league table presents he hasn't really come up with much of an alternative.

It's basically a graph that tells us how teams are doing compared to some hypothetical satisfactory result vs how a team performed last season.

If you were to simplify it and used it in Irelands world cup group that graph might tell us that Ireland got two "par" results vs Slovakia, when in reality when the table was finished we'd actually achieved "-4". So basically it has told you nothing.

It demonstrates how sides are doing relative to how they should be doing on the basis of the games they've played so far. A good example is Bolton, I think its fair to say that after the opening couple of fixtures in which they faced United, Liverpool, City, Arsenal and Chelsea, that the league table didn't accurately reflect their position, in that they would not have been expected to get much, if anything out of those games, however with a run of easier games on the horizon, its quiet likely that their actual league position will rise steadily

I think its fair to say that as the par system demonstrates, that three points at Old Trafford is a far better result then 3 points at home to Swansea

Dotrel

Also, how does it factor in the fact that while top 13 rated teams will indeed be aiming for this hypothetical "26w +12d =90pts" target but teams rated 14-20 only have 6 away "wins" to go for because they themselves are one of the 7 "easy away wins".

Basically the whole thing seems very flawed.

tommyhaas

Dotrel wrote: »

Also, how does it factor in the fact that while top 13 rated teams will indeed be aiming for this hypothetical "26w +12d =90pts" target but teams rated 14-20 only have 6 away "wins" to go for because they themselves are one of the 7 "easy away wins".

Basically the whole thing seems very flawed.

Its based around sides going for the CL or title. Obviously different criteria would need to be applied to sides who finished between 14th and 17th, and those who are newly promoted

Personally, I think its a good indicator of performance so far, and will become more accurate as the season progresses

Mr. Guappa

tommyhaas wrote: »

Its based around sides going for the CL or title. Obviously different criteria would need to be applied to sides who finished between 14th and 17th, and those who are newly promoted

Personally, I think its a good indicator of performance so far, and will become more accurate as the season progresses

So is the actual Premier League table.

Dotrel

tommyhaas wrote: »

It demonstrates how sides are doing relative to how they should be doing on the basis of the games they've played so far. A good example is Bolton, I think its fair to say that after the opening couple of fixtures in which they faced United, Liverpool, City, Arsenal and Chelsea, that the league table didn't accurately reflect their position, in that they would not have been expected to get much, if anything out of those games, however with a run of easier games on the horizon, its quiet likely that their actual league position will rise steadily

I think its fair to say that as the par system demonstrates, that three points at Old Trafford is a far better result then 3 points at home to Swansea

Yeah but it falls apart because it over simplifies what is a hard game and what is not.

An away win at Sunderland would probably be easier than a home win vs Man United , yet this guys system would claim the opposite.

tommyhaas

Mr. Guappa wrote: »

So is the actual Premier League table.

See Bolton's current position relative those around them, given the sides each have faced so far

Dotrel wrote: »

An away win at Sunderland would probably be easier than a home win vs Man United , yet this guys system would claim the opposite.

I actually agree with that point, however if you see his blog, he talks extensively about home those who have won the league over the past few years have always done so on the back of outstanding home records, particularly in the last 6 years, hence the 3pt par at home is based on history as oppose to speculation

Mr. Guappa

tommyhaas wrote: »

See Bolton's current position relative those around them, given the sides each have faced so far

Taking the Bolton example, they have played Liverpool and City away (Par = 1 point) and also QPR, Wigan and Swansea away (Par = 3 points). Given that Par is 3 points for all home games then this system takes no account of the strength of Bolton's opponents at the Reebok. Therefore, the actual league table far more accurately reflects how poor Bolton have been this season.

tommyhaas

Mr. Guappa wrote: »

Taking the Bolton example, they have played Liverpool and City away (Par = 1 point) and also QPR, Wigan and Swansea away (Par = 3 points). Given that Par is 3 points for all home games then this system takes no account of the strength of Bolton's opponents at the Reebok. Therefore, the actual league table far more accurately reflects how poor Bolton have been this season.

I was using Bolton as an extreme example. Obviously as they are not realistically chasing CL football, hence par results would differ for them

Look at Bolton relative to Wigan, only one point separates them. Wigan have a run coming up in which 4 of their 5 games are Utd, Liverpool, Chelsea and Arsenal. Obviously you would expect that within this period, Bolton will accumulate far more points then Wigan and the gap will widen

Its similar in ways to the end of season run in, in the that the favourite to come out on top are not always the team with the most points prior to the run in

K-9

Mr. Guappa wrote: »

Taking the Bolton example, they have played Liverpool and City away (Par = 1 point) and also QPR, Wigan and Swansea away (Par = 3 points). Given that Par is 3 points for all home games then this system takes no account of the strength of Bolton's opponents at the Reebok. Therefore, the actual league table far more accurately reflects how poor Bolton have been this season.

In fairness it isn't really aimed at teams that aren't aiming to win or finish top 4. You'd need a different criteria, say 42 points for Bolton, a point away to the top 6 would be ahead of expectations and maybe a draw at home to them, the norm.

There is a certain logic to it, it's why supporters look at the season run ins and say this team has an easier run in than the other one and certain games are six pointers.

PARlance

tommyhaas wrote: »

...1. Its from a blog of the same title by a sports scientist (or something)
2.
While it doesn't take into account factors such as the positive/negative affects of form or the pressure of being top, I still think it gives a good indication of the performance of sides thus far, and their chances from here on in

tommyhaas wrote: »

3. See Bolton's current position relative those around them, given the sides each have faced so far


4.
I actually agree with that point, however if you see his blog, he talks extensively about home those who have won the league over the past few years have always done so on the back of outstanding home records, particularly in the last 6 years, hence the 3pt par at home is based on history as oppose to speculation

1. Are you this "sports scientist (or something)"?
You have produced an identical graph to the one used on that blog, only difference being that this one has been updated to GW10 whereas blog remains last updated to GW8.
When is the sports scientist going to post this updated graph on the blog?
Or how did you get the updated one?

2. It doesn't take into account much factors to be honest
To highlight one area for example
The 14-17th(from previous yr) placed teams and 3 newly promoted teams are very seldom the worst 7 teams.
The last 10-15 league games of a season are all about your actual position and the position of the team you're playing.
The easiest 7 away games are more likely to come at the end of the season against mid table teams that have little to play for.

3. I don't think this model is designed for Bolton either? Surely you/sports scientist would have to assign Bolton a different par, thus creating a different model?

4. Why not just base the model on home games won?

5. And this is an additional one... The blogger/sports scientist/maybe you is clearly a Liverpool supporter....Has it gotten this bad????? Really

tommyhaas

ajcurry123 wrote: »

1. Are you this "sports scientist (or something)"?
You have produced an identical graph to the one used on that blog, only difference being that this one has been updated to GW10 whereas blog remains last updated to GW8.
When is the sports scientist going to post this updated graph on the blog?
Or how did you get the updated one?

What the fook?

No, it certainly is not me. The author posts on RAWK, and posted the update there, of which I copied. I would post the link to the thread, however I'm sure a detective like yourself would get far more out of finding it for yourself

ajcurry123 wrote: »

2. It doesn't take into account much factors to be honest
To highlight one area for example
The 14-17th(from previous yr) placed teams and 3 newly promoted teams are very seldom the worst 7 teams.
The last 10-15 league games of a season are all about your actual position and the position of the team you're playing.
The easiest 7 away games are more likely to come at the end of the season against mid table teams that have little to play for.

Fair point about the midtable teams having little to play for at the end of the season, this certainly is not factored in

ajcurry123 wrote: »

3. I don't think this model is designed for Bolton either? Surely you/sports scientist would have to assign Bolton a different par, thus creating a different model?

Firstly, I'm not a sports scientist. I'm not even entirely sure what a sports scientist is, however the author stated he was on RAWK. Again, I would post the link to the thread, however I'm sure after you finish playing Cluedo, finding it'll give you something to do

Regarding Bolton, I was just using them as an extreme example of a side who had a tough run of games so as to demonstrate a point. Had you bothered reading the thread there detective, you may have noticed I said stated previously that in order for it to be relevant to Bolton, a different par level would have to be drawn up

ajcurry123 wrote: »

4. Why not just base the model on home games won?

5. And this is an additional one... The blogger/sports scientist/maybe you is clearly a Liverpool supporter....Has it gotten this bad????? Really

What would be the relevance on basing it on home games only?

How is the team that the author supports in any way relevant to the model? If you bothered your arse looking there detective, you'd notice that the author drew up the guideline at the start of the season, hence he could not have predicted that 10 games in, it would look favourably on Liverpool to an extent

K-9

ajcurry123 wrote: »

2. It doesn't take into account much factors to be honest
To highlight one area for example
The 14-17th(from previous yr) placed teams and 3 newly promoted teams are very seldom the worst 7 teams.
The last 10-15 league games of a season are all about your actual position and the position of the team you're playing.
The easiest 7 away games are more likely to come at the end of the season against mid table teams that have little to play for.

Yeah, but the top 6/7 don't change much and everybody would agree Stoke is also a hard place to go, a draw a good result.

3. I don't think this model is designed for Bolton either? Surely you/sports scientist would have to assign Bolton a different par, thus creating a different model?

Yeah, not getting relegated would be their main target.

4. Why not just base the model on home games won?

5. And this is an additional one... The blogger/sports scientist/maybe you is clearly a Liverpool supporter....Has it gotten this bad????? Really

It takes account of Liverpool playing 6 of their 7 toughest Away games in their first 20, City, Arsenal, Chelsea, Spurs, Stoke and Everton with only United Away after that. Seems a fair enough way to rate Away games.

Also if your dropping points to Norwich and Sunderland at Home, they need to be making that up Away from home, the graph shows they have.

As you point out, it doesn't take account eg. of Newcastle having a great season so far or the promoted teams all having pretty decent starts to the season.

PARlance

Forgive me for calling you a sports scientist...I would have expected him to post the update to his blog, the link that you gave in this thread!

I'll stay well clear of RAWK if you don't mind!

Re Bolton, cup of tea and smoke got in the way of posting sooner. I didn't see you post, but fair play for admitting another flaw in it as a point of reference v prem league table

Re Home games, the sports scientist claims, as you've posted yourself,
"he talks extensively about home those who have won the league over the past few years have always done so on the back of outstanding home records, particularly in the last 6 years, hence the 3pt par at home is based on history as oppose to speculation"
Then why not based a prediction system purely based on home games won, it would make more sense then the current one the sports scientist is using

Re Supporting Liverpool, that was a pop at Liverpool from a United supporter...nothing more, nothing less

tommyhaas

ajcurry123 wrote: »

Forgive me for calling you a sports scientist...I would have expected him to post the update to his blog, the link that you gave in this thread!

I'll stay well clear of RAWK if you don't mind!

Re Bolton, cup of tea and smoke got in the way of posting sooner. I didn't see you post, but fair play for admitting another flaw in it as a point of reference v prem league table

Re Home games, the sports scientist claims, as you've posted yourself,
"he talks extensively about home those who have won the league over the past few years have always done so on the back of outstanding home records, particularly in the last 6 years, hence the 3pt par at home is based on history as oppose to speculation"
Then why not based a prediction system purely based on home games won, it would make more sense then the current one the sports scientist is using

Re Supporting Liverpool, that was a pop at Liverpool from a United supporter...nothing more, nothing less

Well I think it has to be based around the full season, in that points dropped at home can be made up for away. I still maintain that its far easier to win at Stoke then at home to City, however on the evidence of past league winners I can see the merit in the stance he has taken regarding home points

The table is not without its flaws, and personally I would have assigned a point home and away (from Liverpools point of view) against Utd, Chelsea, City, Arsenal, Spurs and Everton (as its a derby) as par, however I think there is merit to this system

PARlance

tommyhaas wrote: »

Well I think it has to be based around the full season, in that points dropped at home can be made up for away. I still maintain that its far easier to win at Stoke then at home to City, however on the evidence of past league winners I can see the merit in the stance he has taken regarding home points

The table is not without its flaws, and personally I would have assigned a point home and away (from Liverpools point of view) against Utd, Chelsea, City, Arsenal, Spurs and Everton (as its a derby) as par, however I think there is merit to this system

All I am saying is that in highlighting the importance of the home form of the last 6 winners of the league, he is, putting forward a more accurate, less complicated model.

Eg If a team is averaging 2.5+ points per home game after 5, 10, 15 home games then there are likely to be on course for the title or top 4.
No real need to bring anything else into it if looking for a different model!!!

Go on, you are the Sports Scientist aren't you?

tommyhaas

ajcurry123 wrote: »

Go on, you are the Sports Scientist aren't you?

I think a look through my posts on here will demonstrate a number of things. Firstly, that I clearly don't posses the ability to to make that kind of graph on a computer. Secondly, that I know absolutely nothing about sports injuries, which I'd imagine a sports scientist would, and thirdly, and probably most tellingly, I've posted on numerous occasions what it is that I do for living, and it certainly is not sports science

K-9

ajcurry123 wrote: »

All I am saying is that in highlighting the importance of the home form of the last 6 winners of the league, he is, putting forward a more accurate, less complicated model.

Eg If a team is averaging 2.5+ points per home game after 5, 10, 15 home games then there are likely to be on course for the title or top 4.
No real need to bring anything else into it if looking for a different model!!!

Go on, you are the Sports Scientist aren't you?

Yeah, but it's kind of pointless ignoring Away games! They do have relevance.

If you look at United last season, they drew 10 Away games and people thought that was a poor stat, they actually did okay Away, comparable to City and Chelsea, the Home form won it!

Liverpool's dreadful Away form cost them 5th and maybe 4th.

Concentrating on Home form is ignoring his point. You can drop points at home, as long as the team makes it up Away from home, all is good. Arsenal compared to Spurs and Liverpool last season eg.

amacachi

One thing to note with Arsenal is that 40% of their games so far has been against other expected "Top 6" teams whereas over the season it's 26% so one may say they should have an easier run of games coming up and their current position doesn't reflect that.

Morzadec

It's far from a flawless system, and I don't think anyone is claiming it is.

But it provides an interesting alternative look at the table which maybe gives a bit more information that the real PL table.

He has to set some sort of parameters and it's hard to argue with those that he has chosen (even though I would agree that, for example a win at West Brom being considered 2 points better than a win at Norwich is perhaps flawed. I'm not sure how this could be remedied, perhaps making par in some matches 2 points or even 1.5, even though these are clearly points that are literally unattainable.)

Anyway, it's not a perfect system and shouldn't be taken as the definite truth about how teams are really doing, but it provides an interesting alternative perspective that imo gives more information than the actual standings.

K-9

amacachi wrote: »

One thing to note with Arsenal is that 40% of their games so far has been against other expected "Top 6" teams whereas over the season it's 26% so one may say they should have an easier run of games coming up and their current position doesn't reflect that.

It's actually one of the flaws in the system, Liverpool did alright against the top 5 last season, matters nothing if they're dropping points in games against the bottom 3/4, which was their undoing.

Then again his system does actually take account of that!

Morzadec wrote: »

It's far from a flawless system, and I don't think anyone is claiming it is.

But it provides an interesting alternative look at the table which maybe gives a bit more information that the real PL table.

He has to set some sort of parameters and it's hard to argue with those that he has chosen (even though I would agree that, for example a win at West Brom being considered 2 points better than a win at Norwich is perhaps flawed. I'm not sure how this could be remedied, perhaps making par in some matches 2 points or even 1.5, even though these are clearly points that are literally unattainable.)

Anyway, it's not a perfect system and shouldn't be taken as the definite truth about how teams are really doing, but it provides an interesting alternative perspective that imo gives more information than the actual standings.

Yeah, I'd say games against the promoted teams in the first half of the season are harder to earn, the feel good factor. Plus Away games at the bottom 5/6 teams fighting relegation at the end of the season can be potential banana skins.

There are pitfalls in any system like this, it seems a decent way of trying to average out a season.

Looking at a teams run in and going United have an easier run in is the exact same thing, except in reverse!

everdead.ie

amacachi wrote: »

One thing to note with Arsenal is that 40% of their games so far has been against other expected "Top 6" teams whereas over the season it's 26% so one may say they should have an easier run of games coming up and their current position doesn't reflect that.

Similar situation with spurs 9 played 4/5 vrs top 6/7(Newcastle).
Now there is a good run of games where we don't have to face any of that top 7 teams.

tommyhaas

Morzadec wrote: »

It's far from a flawless system, and I don't think anyone is claiming it is.

But it provides an interesting alternative look at the table which maybe gives a bit more information that the real PL table.

He has to set some sort of parameters and it's hard to argue with those that he has chosen (even though I would agree that, for example a win at West Brom being considered 2 points better than a win at Norwich is perhaps flawed. I'm not sure how this could be remedied, perhaps making par in some matches 2 points or even 1.5, even though these are clearly points that are literally unattainable.)

Anyway, it's not a perfect system and shouldn't be taken as the definite truth about how teams are really doing, but it provides an interesting alternative perspective that imo gives more information than the actual standings.

This is the thread on RAWK about it

http://www.redandwhitekop.com/forum/index.php?topic=278916.160

On pg 5/6 of the thread, the author 'Prof' explains why he stayed away from par 2 games. His primary reason was that he wanted it to be possible to achieve a par in each match, which obviously wouldn't be with a par 2

thebullkf

tommyhaas wrote: »

It demonstrates how sides are doing relative to how they should be doing on the basis of the games they've played so far. A good example is Bolton, I think its fair to say that after the opening couple of fixtures in which they faced United, Liverpool, City, Arsenal and Chelsea, that the league table didn't accurately reflect their position, in that they would not have been expected to get much, if anything out of those games, however with a run of easier games on the horizon, its quiet likely that their actual league position will rise steadily

I think its fair to say that as the par system demonstrates, that three points at Old Trafford is a far better result then 3 points at home to Swansea

while i quite like the system, 3 points is 3 points, the win at united is morale wise worth more... unless you're in direct competition then its worth more on GD. eg 6pointer .etc

tommyhaas

Th latest update

Part 6 – The Good, The Bad and The Level Par

Is beating Arsenal away, Everton away and West Brom away the stuff of champions? Maybe. But what certainly aren’t results befitting champions elect are home draws to newly promoted clubs.

I’ve only looked back six years, but in all of those seasons, only one home match has not been won by the eventual champions to a newly promoted club. That was Man Utd’s home draw with West Brom last season.

Liverpool have had some good results in the league this season. Despite the losses away at Stoke and Spurs in consecutive games, their away form has been better than the model would suggest a title winning team should have. However, their away form has been overshadowed by some very bad results at Anfield, with draws to Sunderland, Norwich and Swansea. While some credit needs to go to those teams for the way they played, title winning teams win those games.

Consistency is the key to any successful season, and having the occasional great result like beating Man Utd 6-1 at Old Trafford will not define the season for Man City. What will define their season will be whether they can maintain their consistency in winning the most winnable matches.

The table below summarises the number of results so far that have matched, fallen below, or exceeded the par result for each team. What it illustrates is that Liverpool have been incredibly inconsistent so far, with the most above par and the most below par results. While it is good to have offset some of the damage caused by the below-par results, it is not reasonable to expect them to continue to pick up points in the harder to win matches. A win at Stamford Bridge next time out would pull them back to 2-below-par, but that certainly won’t be the shortest priced result in any bookmaker’s shop. They need to learn very quickly how to win consistently at home, or they might soon be looking up at the 71 point line.

The Alternative Premier League Table graph shows that Liverpool is the only team featured not to have had a period of consistent title winning form. Arsenal have recovered tremendously with a six match, above par run, and Spurs have stormed back into contention for the top honours since their opening two losses

http://rawkprof.blogspot.com/2011/11/part-6-good-bad-and-level-par.html

tommyhaas

Latest update

Beginning to take shape. Lack of consistency is killing Liverpool, while Arsenal and Spurs curves both represent the difficulty and time in climbing back up towards par after a poor start. City have a par 1 game at Liverpool this weekend, so an opportunity to move further over, while Liverpool need a win to stay at two under. Spurs and Newcastle are the only other two to have an opportunity to move up this weekend, as all play par 1's, Spurs at West Brom and Newcastle at Utd. A win for Spurs would put them on par, with an excellent opportunity of maintaining that position with a decent run of fixtures coming up to Christmas

Fixture par ratings, coloured fixtures being par 1, while all other are par 3