Topic: The Alternative Premier League Table 2012-13

[Prof]

The Alternative Premier League Table 2012-13

This is a continuation of the original APLT thread which ran last season and can be found here: http://www.redandwhitekop.com/forum/index.php?topic=278916.   The blog I used to write occasional summaries (until I lost all faith in January) can be found here: rawkprof.blogspot.co.uk

For those of you who are new to the APLT, or those who need a reminder of the model, the APLT makes an assumption that in order to win the league title, a team needs to win 90 points of the season.  This can be achieved by winning all home matches, the seven ‘easiest’ away matches (the three promoted teams and the 14th-17th ranked teams from the previous season) and draw the 12 remaining away matches.  I refer to these as the ‘par results’.  As in golf, par will be achieved more often than not, but sometimes points are dropped or gained in relation to par.

The fixture list below for the top seven teams indicates the ‘hardest’ matches which are all par 1s, with the remaining fixtures all par 3s.



As with last season, Liverpool have a difficult opening run of fixtures, with all of their first seven away matches being par 1s.  For Liverpool, 28 points from the first 14 matches will be ‘title winning form’ assuming the APLT model is correct.

As a contrast; Chelsea, Man City, Spurs and Newcastle all have four par 1s, with Man Utd and Arsenal both having five par 1s in the opening 14 matches.  Therefore, Liverpool could be between 4 and 6 points worse off to their rivals at this point, but with ‘more-winnable’ matches remaining in their fixture list, this gap could be closed.

This means that the APLT will place Liverpool higher in the table than in the real league table early in the season to reflect their more difficult start to the season.

If you have any doubts about the reliability of the model, the tables below show the results of the 26 par 3s and 12 par 1s for each of the six originally featured teams.


 
As I illustrated in depth last season, the predictability of results in the par 1s is actually quite low (title winning teams over achieve as a rule to offset dropped points in the par 3 matches), but title winning teams must win most if not all of their par 3 matches.  Last season, Man City won 24 out of their 26 par 3s (they drew at home to Sunderland and lost away at Swansea – not exactly the most predictable banana skins if the pars are assigned before the season starts).

Over the season, I will plot the results in relation to par for all the teams featured on a graph (an example of last season’s completed graph is below).  If a team plays to par, the line on the graph will be horizontal, whereas dropped points will lead to a negative gradient and gained points and positive gradient.


 
The green depreciating line reflects a drop of 0.5 points per game (dropping to 19 points below par by game 38) to illustrate a 71 point season, a reasonable estimation of the points needed to achieve a top four finish.  Any team with ambitions to finish top four needs to be above this green line by the end of the season.

I’ll do my best to keep this up-to-date as best as possible as the season progresses and I hope it illustrates the impact of results on our season.  If we don’t win away at West Brom in match 1, it isn’t a complete disaster.

[Darkshrimp]

good to see this is back It was the source of hope when we seemed hopeless last season..... (Until it really became hopefully).

[Fitzy.]

I love this stuff. Will certainly be good to refer to after our tough start - see where we stand even if results are not spectacular.

[gorgepir]

Still 71 point line or maybe it should be 70?

[El Torres gol]

I gave up when in the summary you said win all home games then the next seven easiest away then the other 12 away?? Did you mean draw the other 12 away games?
Let's face it any team who wins all its games in a season will defo win the league

[Prof]

Oops. Thanks.  You are right, I've amended the OP.  Hopefully it makes sense now.

I appreciate this thread isn't for everyone, it's just my way of seeing how close we are to where we want to be.

[Prof]

Still 71 point line or maybe it should be 70?

It's in there as a guide.  The half-point-per-game depreciation is convenient, hence 71 points is where we end up.  In reality, the final total needed isn't definitely known until the final matches have been played, so if they target point total is slightly out, it isn't that crucial.  If you get 71 points and don't get top four, you can consider yourself unfortunate.  Similarly, if you get 90 and don't win the league.

[DanA]

One of the best thread for last season I reckon. Lets hope we "toe" the line this time.

[stewil007]

One of the best thread for last season I reckon. Lets hope we "toe" the line this time.

which line tho?  thats the question

[Istillstilltrustrafa]

  loved this last year, great to see it up and running again... thanks prof 

[flashman]

Great thread, thanks prof

[kennedy81]

Oops. Thanks.  You are right, I've amended the OP.  Hopefully it makes sense now.

I appreciate this thread isn't for everyone, it's just my way of seeing how close we are to where we want to be.

surely a look at the actual Prem League table will tell you that.
this whole model is flawed is you assume a home win against our nearest rivals is the same as a home win against QPR.

[Thush]

Thanks, Prof. Glad to see this is back.

[Prof]

surely a look at the actual Prem League table will tell you that.
this whole model is flawed is you assume a home win against our nearest rivals is the same as a home win against QPR.

If that's how you want to look at it, I have no problem with that.

To me though, a home win against a rival costs them one point against their par.  If we draw against QPR or a direct rival, we are -2 for either and our rival scores par.  If we lose to a rival at home, then they gain 2 points more than their par and we lose 3.  That's a 5 point swing.

Like I've said before, I know some people don't like this approach, this is here for those that do and to provide a slightly different take on things.

[Il Capitano]

Looking forward to seeing how this unravels over the course of the season. Hopefully it won't spiral out of control downwards like it did last year.

[JasonF]

Enjoyed this thread last year, well until our form tailed off and we didn't look like getting close to either line anyway. Hopefully it'll work out better this time around

[Cicero]

Fucking hell what is this?

[hedger]

What is the point? Seriously?

[LiverpoolKopKings]

What is the point? Seriously?

The point is for those of us that like to analyse the season with the aid of statistics to have a thread were to. Thanks to the dedicated hard work of prof we can.

[Crimson_Tank]

The point is for those of us that like to analyse the season with the aid of statistics to have a thread were to. Thanks to the dedicated hard work of prof we can.

Is correct.

[Hymer Red]

Good to see this thread back Prof thanks

[Matt8Pie]

Do you have a girlfriend?

But seriously, I saw this last season as just another excuse for poor performances and results. It's easy to say "well we're 10th because we've played x big teams and United have play x small teams and are in first so when we play x small teams and United play x big teams the points should be..." when in actual fact, the league table tells the real story regardless of whether a big team or little team has been played and you can't predict the outcome at all.

I'm not saying I'm against you doing it, I just don't see what you're trying to gain with it because as I said, it just gives excuses and helps contribute to the bitterness on this forum, in my opinion.

[Prof]

Do you have a girlfriend?

  No

A wife and three children under three though.  Hence the fact my posts are often rushed and after midnight.

I'm not saying I'm against you doing it, I just don't see what you're trying to gain with it because as I said, it just gives excuses and helps contribute to the bitterness on this forum, in my opinion.

My intention is actually to do the opposite.  At no point last season were we level with Man Utd on the graph, so it isn't a case of making excuses for poor results.

I'm trying to provide a rational and objective measure of the results we achieve throughout the season, against benchmarks stated before the season starts.  I'm not making knee-jerk responses to results that I think are bad.  How many times do people write in post-match threads 'we should be beating x away from home' when in fact a draw in that fixture is a good result.  You don't need to win away at Fulham to win the league, a draw there is a good result.  What you must do though, is win at home consistently and some away matches.

I saw this last season as just another excuse for poor performances and results.

None of the par results can be considered poor results.  26 of them require a win and 12 require draws in the most challenging away games.  At no point is this about making excuses.

Last season, Man City and Man Utd had 6 and 7 below par results respectively.  We had 24.  We haven't won the leage in 22 years, not because we can't beat the big teams at home (because we have numerous times), we haven't won the league because we have lacked consistency.

[owens123]

What's with all the moaning bastards in here? Prof puts a lot of time and effort into this. I and a lot of others finding it very interesting and enjoy the discussion. If you don't see the point, just don't look through the thread, simple as. There's no need to start bashing on someone who's putting a lot of time into something like this. It's not even just moaning, some people are being completely rude and disrespectful.

[Prof]

Thanks to those of you who have posted supportive comments.  I only post this stuff as people have expressed interest in it.

I don't know how many times I've acknowledged that it isn't everyone's cup of tea.  I'm happy for people to question the model and make suggestions.  Last season, a number of people challenged it, and it made for healthy debate.  I think I've managed to provide some quite robust arguments for some of the biggest assumptions, and noted where the biggest weaknesses are.

This season, I'd like to see us win at least 20 of our 26 par 3s.  I think this is realistic and will give us a platform to build on over the next few seasons.  Getting the 12 points from the par 1s (either through draws or wins) would then give us a points total that should be enough for top four.

[DaveCharlie]

Cheers again from me, Prof - love this thread, always look forward to the updates.

[DanA]

Found it really enlightening. You can really pin point where Lucas got injured, where Roberto Di Matteo too over for Chelsea and when Harry's head was turned. Far more than the League table would. You can also see (Contrary tot the league table) that it was always really close between City & United for the title and that United bottled it.

[TommyT1995]

Looks really enlightening, I saw this thread last season but didn't bother having a look but will definitely keep up with it this season.
Btw for the people who that it's pointless and what not; there is the option of not reading the thread...
Anyway look forward to it, only about a week to go before we start.

[Matt8Pie]

No

A wife and three children under three though.  Hence the fact my posts are often rushed and after midnight.

My intention is actually to do the opposite.  At no point last season were we level with Man Utd on the graph, so it isn't a case of making excuses for poor results.

I'm trying to provide a rational and objective measure of the results we achieve throughout the season, against benchmarks stated before the season starts.  I'm not making knee-jerk responses to results that I think are bad.  How many times do people write in post-match threads 'we should be beating x away from home' when in fact a draw in that fixture is a good result.  You don't need to win away at Fulham to win the league, a draw there is a good result.  What you must do though, is win at home consistently and some away matches.

None of the par results can be considered poor results.  26 of them require a win and 12 require draws in the most challenging away games.  At no point is this about making excuses.

Last season, Man City and Man Utd had 6 and 7 below par results respectively.  We had 24.  We haven't won the leage in 22 years, not because we can't beat the big teams at home (because we have numerous times), we haven't won the league because we have lacked consistency.

Okay that makes more sense to me now. So you're doing your bit to try a restore a bit of peace and perspective on here then?

Good luck!

[Red number seven]

According to this we were looking good for top 4 until about game 20 when we started a serious decline in relative league form.

That was January when we lost 3-1 to Bolton, just after we'd beat City in the league cup - feels about right.

The absence of both Lucas and Suarez (initially did ok for amonth without Lucas), the effects of the negativity around the club/ Kenny/ Suarez plus the distraction of the cups was probably when the league season went tits up. This should give us hope that, given those things are hopefully not likely to recur (obviously you never know with injuries), we can chalenge for 70 plus points and a return to the top four this season.

[Niru Red4ever]

I stopped checking the thread when our season went haywire. Now, looking at it really shows how things unraveled since Jan.

Also, Man Utd and Man City were really close throughout the season - somehow during mid season I had that idea that City was streets ahead. Anyways, Utd did bottle it up at the end, which was apparent without this table.

Chelsea also suffered under Di Matteo; but then they had other distractions. Will be interesting to see how he copes with the league this time around.

[AnyGivenSunday]

I've not followed this thread in previous years (I think I've popped in mid-way through previous seasons but couldn't be bothered to read back to work out what was going on ), but I intend to keep an eye on it this season.  Cheers Prof

[gorgepir]

Just wanted to thank you with the work you do on this table. It is really the "correct" form table (under the assumption of course).

[Tony Clueless]

This is miles better than the original league table.  It gives us a chance 

[Blinis]

Nice to see this Alternative League Table again this year. Thanks Prof!

[moxey]

What happens if City get 91 points?

[Prof]

What happens if City get 91 points?

Then they finish on +1.  There's nothing to stop three teams finishing above the 90 point line, just on past evidence, no one has achieved 90 points and not won the league, so it's unlikely.  The 90 point line just provides a reference point.

To go back to the original golf analogy, in stroke play golf, scores are tracked against par to indicate how well someone is doing.  If you have a 12 handicap, you adjust the par for each hole, so can see how well you are doing against that target score.  The 90 point line is essentially a par performance for a team that wins the league.  They might make 'birdies' or 'bogies' on the way, but finishing on par or better is likely to win the league.

When Chelsea won the league with 95 points, they would essentially be 5 over the 90 point line by the end.

[Prof]

This is miles better than the original league table.  It gives us a chance 

 

You do realise that by the 38th game, both tables are identical?

[DyingAtheist]

Used to pop in here for a look last season, always gave me a bit of hope, so thanks for putting this up again!

To the people complaining, why bother? If some of us enjoy it, and are thankful for all of Prof's hard work, then why does the thread require your negativity?
(It doesn't)

Cheers Prof.

[mig]

Lost track of this last season. Excellent thread. Bookmarked!