HOW TO WIN THE PREMIER LEAGUE: a historic analysis of what winners do

HOW TO WIN THE PREMIER LEAGUE

By Tjekem

How does a team win the Premier League?  Do you need the top goalscorer? The best offence? The best defence? What about goal difference?

The only thing that we know for sure; is that you need the most points! But how many?

For the purpose of this piece, I will look at the league since 1992. There seems to be two distinct eras; one from 1992 to 2002 and another from 2003 [when the Oligarchs and the oil money came in] to present [2014]. I suspect that there will be a third when we can see trends brought about by FFP.

DO YOU NEED THE TOP GOALSCORER?

NO!
Only 9 of the last 22 winners had the top goal scorer.

  1. 19994/95 – Alan Shearer for Blackburn with 34 goals. Blackburn Rovers won.
  2. 1998/99 – Dwight Yorke of Man U was joint top goal scorer with Jimmy Floyd Hasselbaink of Leeds and Michael Owen of Liverpool with 18 goals. Man U won.
  3. 2001/02 – Thierry Henry for Arsenal with 24 goals. Arsenal won.
  4. 2002/03 – Ruud van Nistelrooy of Man U with 25 goals. Man U won
  5. 2003/04 – Thierry Henry for Arsenal with 30 goals. The Arsenal won.
  6. 2007/08 – Cristiano Ronaldo of Man U with 31 goals. Man U won.
  7. 2009/10 – Didier Drogba for Chelsea with 29 goals. Chelsea won.
  8. 2010/11 – Dimitar Berbatov of Man U was joint top scorer with Carlos Tévez of Man City. Man U won.
  9. 2012/13 – Robin van Persie of Man U [had to shower after typing that] with 26 goals. Man U won.

As you can see, you definitely don’t need the top goalscorer; but what about the top offence?

DO YOU NEED THE TOP OFFENCE?

YES!
Sixteen [16] of the last twenty two [22] seasons, the top offence won the league. Remember that second era? Well, nine [9] of the last eleven [11] winners had the top offence.

How does defence come into the equation?

DO YOU NEED THE TOP DEFENCE?

NO!
Strange, right? Only nine [9] of the last twenty two [22] winners had the top defence. We know that you need defence; so what is happening here?

How about combining the two?

DO YOU NEED THE TOP GOAL DIFFERENCE?

YES!
Sixteen [16] of the last twenty two [22] winners had the best goal difference. Remember that second era again? Well, ten [10] of the last eleven [11] winners had the top goal difference.

What about all three? Defence, offence and goal difference?

IS THERE A GOAL DIFFERENCE TO AIM FOR?

YES!

The last time a goal difference less than +40 won the league was Arsenal in 1997/98 with +38. So, a goal difference greater that +40 appear to be the number; if you want a chance to win the league.

DO YOU NEED THE TOP OFFENCE, DEFENCE AND GOAL DIFFERENCE?

NO!

Only 6 of the last 22 winners have had the top defence, offence and goal difference.

Man U [2000/1, 2007/8, AND 2012/13], Arsenal Invincibles [2003/4], Chelsea [2005/6] and Man City [2011/12].

HAS ANYONE EVER WON THE LEAGUE WITHOUT BEING TOPS IN ONE OF THESE CATEGORIES [OFFENCE, DEFENCE OR GOAL DIFFERENCE]?

NO!

HOW MANY POINTS SHOULD YOU SHOOT FOR?

The average points since 1992 is 86. Remember that second era? Well, the average number of points for the winner since 2003 has been 88.

I once read somewhere that Arsène Wenger and Alex Ferguson broke the season down in 10 game chunks and tried to average a minimum of 2.3 points per game; then they made the dash for the last 8 games.

Let us look at the winners of the PL in the modern era

Year Team W D L Pts Pts/Game
2013/2014 Man City 27 5 6 86 2.26
2012/2013 Man U 28 5 5 89 2.34
2011/2012 Man City 28 5 5 89 2.34
2010/2011 Man U 23 11 4 80 2.11
2009/2010 Chelsea 27 5 6 86 2.26
2008/2009 Man U 28 6 4 90 2.37
2007/2008 Man U 27 6 5 87 2.29
2006/2007 Man U 28 5 5 89 2.34
2005/2006 Chelsea 29 4 5 91 2.39
2004/2005 Chelsea 29 8 1 95 2.50
2003/2004 Arsenal 26 12 0 90 2.37
AVERAGE 27 7 4 88 2.32

What does the above table tells us? Well, first you can’t have more than 6 losses and secondly, your combination of draws and losses can’t total more than 15.

DO YOU HAVE TO BEAT THE TOP 4?

NO!

Let us look at the past Champions of the modern era vs. the top 4/bottom 16

VS. Top 4

Year Team W D L Pts Pts/Game
2011/2012 Man City 5 0 1 15 2.50
2009/2010 Chelsea 5 0 1 15 2.50
2005/2006 Chelsea 5 0 1 15 2.50
2004/2005 Chelsea 4 2 0 14 2.33
2003/2004 Arsenal 4 2 0 14 2.33
2007/2008 Man U 4 1 1 13 2.17
2012/2013 Man U 3 1 2 10 1.67
2010/2011 Man U 3 1 2 10 1.67
2006/2007 Man U 2 2 2 8 1.33
2013/2014 Man City 2 1 3 7 1.17
2008/2009 Man U 1 2 3 5 0.83
AVERAGE 3.45 1.09 1.45 11.45 1.91

DO YOU HAVE TO BEAT THE TOP 4?

NO!

VS. Bottom 16

Year Team W D L Pts Pts/Game
2008/2009 Man U 27 4 1 85 2.66
2006/2007 Man U 26 3 3 81 2.53
2004/2005 Chelsea 25 6 1 81 2.53
2013/2014 Man City 25 4 3 79 2.47
2012/2013 Man U 25 4 3 79 2.47
2005/2006 Chelsea 24 4 4 76 2.38
2003/2004 Arsenal 22 10 0 76 2.38
2011/2012 Man City 23 5 4 74 2.31
2007/2008 Man U 23 5 4 74 2.31
2009/2010 Chelsea 22 5 5 71 2.22
2010/2011 Man U 20 10 2 70 2.19
AVERAGE 23.82 5.45 2.73 76.91 2.40

There are only 18 points available against the top 4, while there are 96 points available against the bottom 16. If a team can dominate the bottom 16 they can still win the league!

There are three [3] champions that did this;
1. Man U [2006/2007] accumulated 8 points out of a possible 18, but crushed the bottom 16 for 81 points out of a possible 96 points.

2. Man U [2008/2009] accumulated 5 points out of a possible 18, but crushed the bottom 16 for 85 points out of a possible 96.

3. The Champions of last season, Man City did this too! They had 7 points against the top 4, but 79 points against the bottom 16.

P.S. Chelsea finished with 16 points out of 18 points against the top 4 and did not win the league!

—-

This examination is continued in the next article which looks at beating the top six, and being top of the league at Christmas

Let us look at the winners of the PL in the modern era

 

Year Team W D L Pts Points/Game
2013/14 Man City 27 5 6 86 2.26
2012/13 Man U 28 5 5 89 2.34
2011/12 Man City 28 5 5 89 2.34
2010/11 Man U 23 11 4 80 2.11
2009/10 Chelsea 27 5 6 86 2.26
2008/09 Man U 28 6 4 90 2.37
2007/08 Man U 27 6 5 87 2.29
2006/07 Man U 28 5 5 89 2.34
2005/06 Chelsea 29 4 5 91 2.39
2004/05 Chelsea 29 8 1 95 2.50
2003/04 Arsenal 26 12 0 90 2.37
AVERAGE 27 7 4 88 2.32

 

What does the above table tells us? Well, first you can’t have more than 6 losses and secondly, your combination of draws and losses can’t total more than 15.

DO YOU HAVE TO BEAT THE TOP 4?

NO!

 

Let us look at the past Champions of the modern era vs. the top 4/bottom 16

VS. Top 4

 

Year Team W D L Pts Pts/Game
2011/12 Man City 5 0 1 15 2.50
2009/10 Chelsea 5 0 1 15 2.50
2005/06 Chelsea 5 0 1 15 2.50
2004/05 Chelsea 4 2 0 14 2.33
2003/04 Arsenal 4 2 0 14 2.33
2007/08 Man U 4 1 1 13 2.17
2012/13 Man U 3 1 2 10 1.67
2010/11 Man U 3 1 2 10 1.67
2006/07 Man U 2 2 2 8 1.33
2013/14 Man City 2 1 3 7 1.17
2008/09 Man U 1 2 3 5 0.83
AVERAGE 3.45 1.09 1.45 11.45 1.91

VS. Bottom 16

 

Year Team W D L Pts Pts/Game
2008/09 Man U 27 4 1 85 2.66
2006/07 Man U 26 3 3 81 2.53
2004/05 Chelsea 25 6 1 81 2.53
2013/14 Man City 25 4 3 79 2.47
2012/13 Man U 25 4 3 79 2.47
2005/06 Chelsea 24 4 4 76 2.38
2003/04 Arsenal 22 10 0 76 2.38
2011/12 Man City 23 5 4 74 2.31
2007/08 Man U 23 5 4 74 2.31
2009/10 Chelsea 22 5 5 71 2.22
2010/11 Man U 20 10 2 70 2.19
AVERAGE 23.82 5.45 2.73 76.91 2.40

There are only 18 points available against the top 4, while there are 96 points available against the bottom 16. If a team can dominate the bottom 16 they can still win the league!

 

There are three [3] champions that did this;
1. Man U [2006/07] accumulated 8 points out of a possible 18, but crushed the bottom 16 for 81 points out of a possible 96 points.

2. Man U [2008/09] accumulated 5 points out of a possible 18, but crushed the bottom 16 for 85 points out of a possible 96.

3. The Champions of last season, Man City did this too! They had 7 points against the top 4, but 79 points against the bottom 16.

 

P.S. Chelsea finished with 16 points out of 18 points against the top 4 and did not win the league!

 

 

The books

The complete Arsenal Anniversary series is to be found on the Arsenal History Society site.

34 Replies to “HOW TO WIN THE PREMIER LEAGUE: a historic analysis of what winners do”

  1. Tjekem, (Name corrected)

    ‘…are favorable to us’ not ‘…is favorable to is’

  2. Very interesting and insightful.

    The one thing I do know is:

    ..if we do well in the top 4 ‘Mini’ League but slip up against the rest the media will bash us for being ‘bullied’ by the Stokes, Birminghams, Blackburns of this world.

    We’ve all seen the headlines:

    ‘It’s no good beating the big boys if you can’t deal with Stoke away on cold, wet, Tuesday evening’

    Then when, as last year, we underachieve in the top 4 ‘mini’ League, but dominate against the rest we get slaughtered for that.

    Seems to me there are just 2 absolutely crucial requirements.

    1-Dominate the teams outside the top 4

    2-Score a lot of goals, the most would be preferable.

    As long as you keep a reasonably tight defence, achieving those 2 things will more often than not be enough to win the League.

    When all is said and done it’s hardly rocket science, but much easier said than done.

  3. By the way, I think it was Bootoomee who earlier this Summer put up a staunch and relentless argument that far too much was being made of our defeats to City, Liverpool and Chelsea.

    It seems your statistics irrefutably back him up.

    Isn’t it funny how the non existent AAAA have adopted those defeats as there new weapon of choice to discredit Wenger, even though it seems, in the great scheme of things, they have in actual fact very little bearing on how you prevail over an entire season.

    Still, when needs must I suppose.

  4. @Jambug,

    I wonder if the big defeats that we suffered last season affect the fans more than it affects the players? After all, it’s still only 3 points lost.

    Professional sports people are renowned for forgetting ‘failures’ and moving on to the next event.

  5. Chelsea won the top 4 ‘mini league’ with 16 out of a possible 18 points, but did not win the title because they dropped too many points against the bottom 16 teams.

  6. Tjekem

    As you say, professional sports people do tend to have the ability to put such defeats behind them.

    As for fans. Well I think that is far more complicated.

    Most of us, no matter how ‘pro’ or ‘anti’ Wenger you are, would be gutted after a defeat like that.

    But I think the biggest differences are these.

    -If you are ‘pro’ Wenger you will be ‘upset’.

    -If you are ‘anti’ Wenger you will be ‘angry’.

    As one firmly in the pro Wenger camp I just felt upset for him and the lads. OK it’s not good, but you put it in to perspective, 1 nil, 6 nil, it’s still only 3 points at the end of the day, you hope the team learns from it and you move on.

    As an anti Wenger type it must be manor from heaven. The media love it. As we can see they STILL haven’t let it go now and the AAAA’s are still hanging on to it as if those 3 defeats are THE most important thing to take out of last season.

    So how it effects the fans rather depends on where you stand in the first place.

    And don’t forget the effects on our fans are often magnified by the constant tirade of negativity and abuse that erupts from the media every time we suffer such a defeat.

  7. @Tjekem

    I haven’t gone through the numbers the way you have, but parts of what you have match my feelings for the numbers. It is a small dataset, which is a problem. Have you tried using medians or other robust measures? Did you try to determine the splitting year (2002/2003) by statistical methods?

  8. “Professional sports people are renowned for forgetting ‘failures’ and moving on to the next event.”

    Two weeks after beating Arsenal in the league and gaining three points Livepool were knocked out of the cup by the eventual winners, the Arsenal!

  9. @Gord,

    I looked at the PL since 1992 and determined that there were two distinct periods, 1992 to 2002 and 2003 to present. The second period [2003 to present] represents the introduction of huge outside money [Russian and oil] into the League. It is my belief that this ‘new money’ allowed teams to have deeper squads. Hence the average points total for winning increased from 86 to 88. Also offensive output increased because of the increase in the quality of squad depth.
    The consistency in the numbers are more pronounced in the 2003 to present [modern era] set of numbers. This will show up in Part 2.

    My first thoughts were to use statistical methods, but once I started, I realised that just compiling the tables and looking at them fulfilled the objective of the analysis. [I suppose I could go back and look at it in more detail, but I believe that the conclusions would be the same].

    Yes, the data sets are small, but I wanted to illustrate that there are many ways to win the League!

  10. Jambug

    “Then when, as last year, we underachieve in the top 4 ‘mini’ League, but dominate against the rest we get slaughtered for that”

    Arsenal only got 2 points out of possible 12 against Man U and Everton last season if my memory serves me.
    Hardly dominating.

    Also if I read this article correctly ,only three times out of ten the eventual winner had a loosing record against the top four. That’s 70% at a time when a team has a good record against other top four teams , they win the league yet the author opines “NO” it’s not necessary but in other cases like eg; top goal difference or top offense, that percentage is only slightly higher at 72% and he unequivocally says “YES” , it’s required.

    I guess it’s all a matter of what you want to believe in. I’ll side with Wenger and countless Arsenal players who have gone on the record and said that improving the record against bigger teams was a priority this season.

  11. finsbury

    Indeed they where 🙂

    As well as in the PL some weeks later.

    If I’m not mistaken the Vampire failed to score one single goal against us.

    Over all we had the better of Liverpool last season though you wouldn’t believe it.

  12. @Gourd,
    Here are three examples to highlight these two distinct eras.
    1. 16 of the last 22 Champions had the top offence, but 9 of the last 11 Champions [modern era] had the top offence.
    2. 16 of the last 22 Champions had the top goal difference, but 10 the last 11 Champions [modern era] had the top goal difference
    3. The average points of the Champion was 86, since 2003 it is 88

    There are more differences between the two eras which will show up in Part 2.

  13. Tom

    Actually Tom, I apologise. I don’t particularly agree with your last post, or in fact a lot of your posts, but I we’ve had a few decent battles.

    I was mixing you up with the odious Rupert.

    Sorry, my mistake.

  14. @Tom,

    All I am trying to show is that there are many, many ways to win the League. The notion that you MUST finish top of the ‘mini league’ is false, as evidenced that 3 teams won the League while losing the ‘mini league.’ The percentages are immaterial, because it happened!

    How do you explain Chelsea winning the ‘mini league’ with 16 out of 18 points and not winning the League? Yet Man City lost the ‘mini league’ with 7 out of 18 points, but still won the title!

  15. I have nothing but respect for you and if our debates got heated at times that’s because we are both passionate about Arsenal, so no harm no foul in my book , but out of curiosity what part of my post do you disagree with?

  16. Tjekem
    First off, I did enjoy your article and your data is interesting.

    Consistency is the name of the game and the team with the least ups and downs during the season usually wins it.

    As for Chelsea having a superior record against other big teams and not winning the league, well, Mourinho does something no other manager is willing to do especially when faced with a quality opponent.

    What other manager of a big club would go to Anfield and refused to engage in any type of offensive play for most of the game? I can’t think of any , can you?
    His number one objective in important games is to always stop the opposition from executing what they do best. It usually makes for a lousy spectacle but he doesn’t care.

    His poor record against so called “smaller ” teams was clearly the fault of referees and Chelsea players he didn’t approve of, didn’t you hear? 🙂

  17. @ tom 4:12 pm

    The author asks if it is absolutely necessary to win the top-4 mini league, in order to win the League, and the answer is NO! This is not his opinion, it’s a FACT!

    Also, context is crucial: Wenger and “countless” Arsenal players commented about the “bigger” teams, because our record against the “smaller” teams is actually great. Again, their comments were more to do with the way we lost those games ie. scoreline, early goals, concentration, etc.

    Like you said, I guess it’s all a matter of what you want to believe in.

  18. @Tom,

    So true when his teams looses! It’s never his fault! Genius that he is……but, when his teams win? It’s all about him.He won, not the team.

  19. Brilliant article. The league is won through consistency, as Bootoomee would say 🙂 Lose the odd game here or there, even to your immediate rivals, but maintain good form against the rest and that will be enough to see you over the line. Chelsea had the best tally out of the top four mini league but that wasn’t good enough even for second place!

  20. What good is thrashing Arsenal 6 nil today and then lose your next two home matches to the likes of Palace and Sunderland 🙂 The result is you finish 3rd, and without a trophy 🙂 🙂

  21. @AL,

    Thanks!
    What we can say is:
    1. Maximum losses for the champion is 6
    2. The combination of losses and draws CANNOT be greater that 15

    Look for some more nuggets in Part 2.

  22. Tom

    You question the authors following statement:

    ‘To win the PL do you have to win the top 4 mini league? No’.

    That is all he says, and that is a fact.

    He doesn’t say you can loose every game. The author is just making the point that it is in no way essential to ‘win’ the top 4 mini league. In fact 5 times teams have won the PL accruing an average of less than 2pts per game in that mini league.

    Compare that to the PL winners record against the rest in which they have never accrued less than 2.17pts per game

    It is obviously preferable to have a good record against both sets but I agree with the conclusions drawn by the author that by and large, a good record against the ‘rest’/poor record against the top 4, is infinitely preferable to the reverse.

    And by saying the following you actually support his argument.

    “Arsenal only got 2 points out of possible 12 against Man U and Everton last season if my memory serves me.
    Hardly dominating.”

    This is why:

    What you draw attention to is that in fact, against the other 6 of the top 7 we actually had a diabolical record, accruing just 11 points from a possible 36 and yet we STILL finished just 7 points behind the Champions.

    The 11 points we accrued from the 36 available is only about 30% of the available points.

    And that’s the point, just a slight improvement on that, to a still poor 18 points, or 50%, would of been enough.

    If you had that kind of record against the rest you would be in all kinds of shit because 50% out of the possible 78 points available is just 36 points. Mid table form at best.

    It’s all a bit messy and boring but to sum up, basically you can have a very average return against the top teams and still win the League.

    You cannot have anything but a very good return against the rest to have any chance of winning the League.

    phew, that was painful 🙂

    Sorry if any of the math is slightly out but it’s been a long day, but I hope you see my point.

  23. 50% of 78 points is obviously 39 points. Still mid table form.

    Sorry. told you it had been a long day.

  24. The AAA and any regurgitating zombies have been programmed to repeat chitter chatter about Arsenal’s games against the other top four teams last year.

    Unbeaten at home. City and Gazprom (thanks to Deano) happy to get away with a draw. Liverpool beaten twice.
    I believe such gibberish can be referred to as memes, or traditionally as simple propaganda.

    Arsenal could improve their Away form in the PL against their main rivals.
    They had the confidence to see out the gameplan in Dortmund (also edited out of the narrative). I expect that with greater momentum at the club and increased confidence within the squad that the team will so better in those away games this year. Hopefully they will also retain their consistency against the rest too!

  25. Yes finsbury, Arsenal beat Liverpool twice last season, but with the way the 5-1 score or whatever it was gets repeated you’d think we had gone a whole decade without beating them. And like you rightly point out, only the refs saved the oil clubs from certain defeat at the Ems. You could add utd to that list after vidic was not penalised for a blatant shove on Giroud.

    Out of all of our so-called heavy defeats, the only one which wasn’t ref-assisted, that we can conclude the outcome would not have been much different, is the Liverpool one (even then the ref assisted them by granting them a non foul after only one minute which they scored from). The outcome of the other two was heavily influenced by the refs. Had it been a total implosion Brazil-style it would be concerning. That’s what makes me not read too much into those heavy defeats.

  26. jambug & AL,

    Thanks for pointing out the fact that I wrote a few articles and had a lot of ‘battles’ here against the notion that our 3 big away losses to our top4 rivals cost us the league. I was nodding vigorously as I read Tjekem’s excellent data driven article. And it felt good to see your references.

    Other than bragging rights for fans and net point gains, there is really no need to be fixated with games against rivals. It is for this reason that I find the cliched term “must-win-game” annoying. Any team aiming for the league title should endeavour to win EVERY game. You could lose to all your top 4 rivals and still win the league if you do enough against the remaining 16 AND they fail considerably to do the same.

    In 2006/7 season, Arsenal won the top4 mini league while Man United came last. Man United won the league while Arsenal came 4th. Need I say more?

    Tony wrote a few days ago about how close we came to winning the league last season but made a classic mistake on the matter; and that was tinkering results with the other top 4 teams. This was unnecessary for Arsenal to have won last season. We lost the league by 7 points but without changing any of the results of the other top4 teams, we could have made up 8 points (because of our poor GD) by:

    – beating Swansea at home (+2)
    – beating Everton at home (+2)
    – beating Man United at home (+2) (this is particularly painful)
    – beating West Brom away (+2)
    – Drawing Stoke away (+1)

    Last season was good for us against the bottom 16 teams but Man city still got 5 points in those fixtures than we did. My hope is that we go all out and get the maximum against them this season then against the rivals, we do the best we can.

  27. Conservative or whatever, and I think its the approach of the “Ordinary One”.

    You don’t need or aim for win against top 4… not even top 6.

    You play for a draw and look for wins given chances in the game depending on form or errors. You should get a few. One or two at least.

    You win every single game under for the rest. May be you draw a few but its always about winning.

    Logically, a top team can do that without any trouble but we know there are things called luck and form but game plan goes with logic… If you do the above perfectly, you got AN INVINCIBLE!!!

    and if other big teams can’t do that very well, you get the league…

    Well sometimes big teams lose to weaker teams… like Chelsea last season and Arsenal, far more often than we like 3 or 4 years ago.

Leave a Reply

Your email address will not be published. Required fields are marked *