How much does each point cost? Here comes the miracle man

By Walter Broeckx with many thanks to Mike one of our readers

I might sound a bit pretentious to say that our articles are really worth reading. As a regular writer myself I think they do of course. But in my country praising yourself is not really something you do. So apart from my articles the articles on Untold Arsenal are top quality. That’s more like it.

But if there is one thing that sometimes surpasses the articles it is that we have some great and intelligent comments most of the time.

And this week I found another comment that was so fantastic that I took the liberty to contact the writer (Mike) and asked him if I could use his comment. He tried to put a table on line in which a comparison was made over 10 seasons in the PL.

Comparing the spending of the teams in the PL between 2003-2013. Comparing how many points they won in total and then seeing how many points were made on average.

Then this table calculated the net £’s spent per point won. And finally it calculated how many points were gained per £m spent.  As our comment don’t allow tables it all turned a bit unreadable so I asked Mike the table and wrote it down in an article and will comment on it below this amazing table.

puntenwinst 10 seizoenen

In the first column of numbers you can see at how much money each club has spent on purchases over this 10 year period. Chelsea and Man City are far ahead of the rest but that isn’t really news I think. In third place we see that Liverpool has spent a few pounds. And then we see that our neighbours from down the road have spend the fourth highest amount of all the clubs. And all that for what was it…one year in the CL.

Man Utd is in 5th place and then we find Arsenal. So we did spend a few pounds in that period but not even close to the mega spenders that are Chelsea and City.

The most important things to look at is the nett spend over this 10 year period. And then we see that Chelsea and Man City bought their way to success. The third highest spender however was Liverpool. In 4th place we find Manchester United.

Arsenal are third from bottom. Only two teams spent less than us over that period. Newcastle and Crystal Palace.

If we take that net spend and then compare this to the points won in total and on average we get a very interesting last column.

In this column you can see how many points were gained per £million spent. And now it is getting extraordinary.

And now this shows what miracles Wenger has done compared to all the other teams. Because if you calculate it like that you can see that per £1m spend we managed to get 44 points. The second team in that column is Newcastle with 24 points. Not bad at all compared to the spending. And in third place we find a team that also seemed to have overachieved in the 10 years covered in this period: Everton.

If we compare the Arsenal numbers with the numbers of Manchester United it shows that Wenger has almost got 9 times as many points per £1m spent compared to Ferguson.  If you want to compare Wenger and Ferguson it is easy to look at the titles won. But this table shows first of all with how small was the sum Wenger had to work and also how he in fact outnumbered Ferguson completely. Would Ferguson have won with the budget Wenger had at his disposal? I’m rather sure he wouldn’t have done it. I think he would have walked away from it.

But Wenger didn’t walk away to other clubs when he had the chance. No he stayed to fight with both arms tied behind his back, and he stood up to the media. And most painfully he had to stand up against a part of the club fans who don’t look further than their own little noses.

And if we compare it with the mega spending power of Chelsea and City the work of Wenger becomes even more impressive. Because if we calculate it like that the number of points he got from each spend £1m is 44 times higher than that of those super clubs.

I have said it many times before and have written it many times before. What Wenger has done was a miracle. Keeping a team in the Pl after building a new stadium itself is already a miracle. Keeping a team with no budget in the top 4 is also a miracle.

Doing both things at the same time is even more than a miracle.  My knowledge of the English language doesn’t go that far to find a proper word to describe it.

What I do know is that every Arsenal supporter is forever in debt to Wenger for what he has achieved in those 10 years. And that makes all the abuse aimed at this man all the sadder. Yes he will have made mistakes in that period. Because after all he is human.

But what he did was building a club. A whole new club. In a period just 100 years after Arsenal were within half an hour of going bankrupt we then came back with Henry Norris saving us. I think it is fair to say that what Wenger has done surpasses even this. And it is something that in 20 years time will be looked upon as the most important period in our clubs history.

It took an amazing person to do what he has done. Taking a club to the next level with no money, with your arms behind the back is what no other manager has ever done as far as I know.  Without this man we might be playing Leeds right now…..

The books

Untold Arsenal is here


88 Replies to “How much does each point cost? Here comes the miracle man”

  1. Because I am the guy who much of the time actually presses the “Publish” button to make the story go live, I have the opportunity to add the first comment.

    I try not to do this – I have every opportunity to write articles and see them published, so I don’t like taking up space on commentaries.

    But I am so grateful to Mike for presenting this research and to Walter for working with Mike to make it readable on the site.

    This is stunning. I have never seen these figures before. If I had something in mind when I first set up Untold I guess it was this sort of analysis.

    Mike, we owe you. Big time.

  2. What an eye opener this is. Incredible is the only word to describe it. Our figure is so far removed from all the others it is hard to comprehend what an amazing achievement it has been to keep us in the top four for so long. My respect for Mr Wenger has gone up to an even higher level, if that is possible.

  3. What a super article Walter, and as for “Mike”, I had seen the comment and the figures he had tried to post, thinking I am sure you guys would come up with a masterpiece in this article.

    This puts such clarity on the magnitude of the work done by AFC and especially AW over the past years!

    No one with any common sense can dispute this – no matter what spin they look to put on it!

  4. Over this period of time, not only did we get more points for our PDSm spent in comparison to the rest in the PL , but we have built a monument of a stadium along with all other facilities which have been well-upgraded!!!


  5. Exceptional work by Mike. Great job by Untold Arsenal in publishing such great work.

    I hope that this article spreads like wildfire and wash the AAA brains clean (along with the lazy journalists throughout England.)

  6. thanks Walter and Mike
    Wenger is a genius and the only manager who cares about the club he works for

  7. Beware the man who reads only one book, or who looks at only one table. Wenger certainly deserves huge credit for making bricks without straw. On the other hand the business model the club has followed is just one model. You can have a sustainable model with high debt- as long as you can service that debt. The Glazers at Man Utd have successfully operated a high debt based model.
    If more had been invested in the Arsenal squad more would have been achieved, albeit with a lower ‘points to spend’ ratio. So the question is, was a messianic commitment to a minimum debt model the optimum model?

  8. As I said when I first saw it in the other thread. Astonishing, truly astonishing.

    Again, well done Mike, and of course Walter.

    I cant wait to see who’s the first troll to raise his head above the parapit and attempt to discredit and dispute the indisputable.

    Stan? Cook?

  9. Brilliant by Walter and Mike. An outstanding piece of research, the type of research our lazy journalists are incapable and unwilling to produce. Send this to the media and ask them to comment. If anyone has membership to that French sounding blog, where the chief AAA leader produces his daily anti-Wenger rant, please post it, retreat and wait for the fireworks.

  10. By the way. This table doesn’t even take into account the magnificent stadium and training facility, which again was achieved with the help of our much maligned manager.

  11. Proudkev.

    I’m waiting for:

    We had/have the money.

    The board have trousered all the money.

    We won nothing.

    We are a football team not a business.

    The same old disingenuous shite we always get when stats like this are shown.

  12. Thank you for such painstaking research and an illuminating article, which provides striking confirmation of what many of us have always believed.

    Arsene Wenger is in a class of his own.

  13. Jambug. It does get a tad too predictable. The thing I never understand is how you can claim to support a great football club like ours yet use every opportunity to moan about it. Very strange the way some people are wired.

  14. Have just been arguing with some AAAs at justarsenal who were claiming that wenger was a nobody before joining Arsenal. I counter argued that Arsenal was not a big club before wenger came pointing out that it only dominated the league some 60 years before wenger cane in. I further argued that after 1936 and prior to wenger Arsenal was flirting with success. It was during the wenger period that Arsenal grew into a truly big club we know today.

    By the way those that occasionally visit justarsenal site taketime to read comments by The Analyzer ( that my user name there) . If you agree with the comments just click like to wind up the AAAs. The whole of this week I will be fighting the AAAs on justarsenal.

  15. Two doubles, some of the most breathtaking football ever seen, the only team to go unbeaten, a new training facility, a 60,000 seater stadium and all for less money spent per point then any other team. And yet Wenger cops flak from all angles. Complete madness but it proves how many idiots there are. At least we all know what we are talking about!

  16. As I was driving to Mums yesterday morning I thought, as we had won, Spurs had lost, Liverpool drawn, and there was all the fall out from OT, I thought it would be safe, well comparatively safe, to listen to Talkshite. Well it was, comparatively, but they still came out with this little gem of a dig

    “…….he(cant remember who HE was) has every chance of emulating the great managers such as Fergie, Jose, Clough, Wenger. I only include Wenger because he seems to of mastered the art of entering every competition without a chance of winning anything, yet still keeping his job.”

    That is not verbatim but pretty close.

    What’s more Perry groves was sitting right next to him and said not a dickybird in Wengers defence, yet, correct me if I’m wrong, I believe he actually works on

    What a shit hole of a station that is and Groves is on there endorsing there abuse of his employees and the Club he’s supposed to love.

    Shame on you Groves and any other Gooner, current or ex, that goes near that place.

  17. Now we see a little more why they target AW, because he is that good and Arsenal would be flying high above the rest were he and Arsenal left alone.

  18. Thanks to Tony for publishing this article. Thanks to Walter for writing it and huge thanks to Mike for creating the excellent table.

    I have always been a proud and staunch AKB and this is why. It is no joke that the the European body of Football statisticians voted Arsene Wenger as the best manager over the last decade. I guess they possess this kind of data as a result of their profession and don’t base their value of managers on how many trinkets they acquire by moving from rich clubs to rich clubs.

    As my hero and role model, it is impossible for me to respect or love Arsene more than I already do but on occasions like this, I wish I could.

  19. Excellent article, highlights Wenger’s brilliance. Now that he has a decent amount to spend we’ll hopefully push on to the next level.
    Would it be possible to add wages to transfer spend and perform this analysis? Highly relevant since majority of the money spent at all clubs is on wages. For example, Flamini on a free probably cost us more than big signings for manyother clubs.

  20. Bootoomee

    “It is no joke that the European body of Football statisticians voted Arsene Wenger as the best manager over the last decade.”

    Is that so? Where can I find an article relating to this?

  21. Gord

    Thanks for that. Problem is for about the last 2 Months or so I haven’t been able to access the ‘Mail’ on line. I used to be able to but all of a sudden I couldn’t. Don’t have a clue whats gone wrong, but as I don’t really have reason to visit it often I haven’t bothered to find out why, so unless you can help with that it would be great if you could point me in another direction to see the article, Thanks.

  22. I am seeing the odd other link, but few to reputable news organizations. Some of the links are to blogs, where it appears the intent is to berate this award.

    Perhaps you have a bad cookie for the Mail? You should be able to look at all the cookies stored on your machine, and delete the ones from the Mail. I would imagine that should work to let you view them.

    In other news, Chelsea is successfully parking the bus against Shrewsbury Town in the Capital One Cup (still 0-0).

  23. @ tinalex
    October 28, 2014 at 8:45 pm

    Arsene made his name as a successful coach while he was at Monaco by winning the French league, the French cup and getting to the quarter finals of the Champions League (back then it was called The European Cup). This at a time when Marseille were riding high on corruption. The club won the EC by bribery and corruption.

    Glen Hoddle won his only league champion medal at Monaco playing for Arsene as did Mark Yates.

    From his time at Monaco Glen came to regard Arsene as his mentor.

    You have to feel for Glen and other spuds as Arsene has come so near and yet so far away from spudland. While Arsene stays loyal, time and managers go marching on at their beloved club.

    Think about this it was bad enough losing Sol Campbell to us but Arsene did us a big favour high jacking Glen. If instead of being at Monaco Arsene had been our manager and Glen had crossed Finsbury Park to come to us.What a thought, the mind boggles at the uproar that would of caused.

  24. Amazing! Nearly twice as many points per million pounds as the next highest! Fantastic research!

  25. Can anybody here guess the fans of the premier league club that holds Arsene in the highest (albeit grudging) respect?

    Answer: Tottenham fans!

    I have heard them say on many occasions that they cannot wait for the day Wenger leaves. Quite a number of them are convinced that Arsene has been the difference over the last 2 decades. It is difficult to argue against that point.

    On the hand we have some Arsenal fans crying like babies because Arsenal is winning trophies (WHICH IS ALWAYS THE REASON BEHIND ALL THE MOANING) while Arsenal built and was paying for a stadium. People seldom appreciate what they have till they’ve lost them.

  26. Thanks to Mike and Walter for this astounding bit of information and well written post. Went off to spread the word on another Arsenal blog not sure it will make any difference to those who blame Wenger for all of Arsenal’s perceived problems.

  27. I am not a big fan of numbers unless its 36-24-36 or maybe 40. Anyway jokes aside, its quite an eye opener, the research that’s done. I have quite a few of manu friends who jumped on the Manu bandwagon & I keep telling them that ask fergie or egorinio to take an average midtable team & do half the job what AW has done and I will be their slave. They have no answer, just oh we won X number of cups etc.

  28. This article is Untold at its finest, these numbers proving the club truly is in the presence of genius. Quite incredible..
    The board know it, others in football know it, have a feeling even the media and their pgmol allies know it.

  29. It would be enlightening to see this table with combined net transfer spend and wages. I don’t know this, but my thought is that Arsenal would be around 5th. In which case, Arsene would still be overachieving, but perhaps not by as much as some might believe

  30. A County FA official has been suspended for 4 months for sexist remarks to a refereeing official.

    There is no mention of Mike Riley 😈 in the article, so we don’t know if this John Cummings is related to Mike Riley or not. The story rementions David Elleray (former Select Group referee) putting his foot in his mouth with racist remarks.

    In other news, Chelsea is successfully parked the bus until half time against Shrewsbury Town in the Capital One Cup, and managed to win the game when “Own Goal” scored in the 81st minute.

  31. Untold Arsenal member Mike did a fine job with the table, I just don’t want Mike the table maker to be confused with Mike Riley :twisted:, the referee assigner who is bent. As I mentioned a couple of posts above.

    There are some limiting situations for a league. Here are some.

    If a team was to get a perfect season (38 wins), that is currently 114 points. If some team was to have a perfect season, my wish would be for Arsene Wenger to be the first. If any team was to get a perfect season, there is no reason to expect other teams from having any particular win/loss/draw record. But, theoretically there is a possibility for a perverse, perfect season. That would be the season where the team that wins the league wins every game they play. The team in 2nd place only loses to the team which finished first, and so on down the league. Champion’s League play revolves around 4th, which would be 32 wins and 96 points. The 18th place team would have 4 wins (12 points), and a person could say the relegation line was either 12 points (highest number of points which were relegated) or 17 points (one less than the last team that stayed up).

    This situation results in maximum issuance of points, at 1140 points.

    The lowest number of points possible, results from every game being a tie. In the current scoring scheme, this would result in 760 points being issued. (I’ve mentioned before that I think FIFA should enact a change, such that points are only ever issued if goals are scored. Hence, any 0-0 ties would result in 0 points being issued.) I call this situation, the most boring season.

    I would suggest that the fraction between 760 and 1140, that any particular season falls, is a measure of how exciting the season was (possibly just to the neutrals).

    I am going to followup with other things. In places, I am looking for outliers. For most people, what outliers seem to indicate is data that can be ignored. And this is NOT correct. There are methods to flag data as possibly being an outlier (and usually one can only flag a single datum, and the method cannot be applied iteratively). But what a person needs to do with a data point which MAY be an outlier, is find an explanation for why it can be ignored. Why is that data point sampling a distribution different from what you thought you were sampling. In other submissions (below), I am ignoring outliers just on suspciion, which isn’t correct.

    For example, let’s say we are looking at how many points the lowest place team has at the end of the season. Perhaps a team has exceptionally bad luck with injuries, in particular injuries to whoever is in goal. At the end of the season, this team may have an exceptionally small number of points. By explaining why this particular data point is unusual, it may be possible to ignore it for the purposes of why you are examining the data.

    In much of what I am doing, I am looking at the median, not the average (or the mean, sometimes called the expected value). The median is a more robust indicator of central tendency than the average is.

  32. Exceptional work, Mike! Thanks for the heads-on, Walter. We do need readers like you in this blog. The stats were mind-blowing. Arsene Wenger should be forever remembered as visionary and selfless person for what he did. I’m sure there will be a gold statue of him at the stadium he helped build which I hope will be renamed Arsene Wenger Stadium when he retires. If he retires, of course! Man, can’t imagine Arsenal FC without Arsene Wenger. God bless you!

  33. @Quincy

    I need to leave room and opportunity for people to respond to Walter’s posting of Mike’s valuable work. But I hope that this becomes somewhat useful, and not just a typing exercise.

    I guess you could say I am building infrastructure.

    Most people post on Untold during the UK daytime. But, there are people from elsewhere who post outside of normal daytime UK hours. And I don’t want to swamp them either.

    All clubs (or almost all clubs) in the English Premier League at the beginning of the season expect a reasonable chance at winning the league. Luck (random chance) does play a part in how games end, and at the beginning of the season all teams are equal (at zero points).

    What is expected as a season progresses, is that the “quality” (Wenger’s choice for describing things) of a team will develop, and we see the ranking of teams develop. What isn’t wanted; is that the referees can influence games by allowing conditions such that players can be injured, that referees can make decisions which influence the outcome of the game, or that referees can make decisions which result in players being suspended when they should not have (excess cards).

    A primary expectation of referees, is that they protect all players equally. In physics, a a person might label this as a zeroth order expectation. A consequence of it being zeroth order, is that it isn’t actually part of the laws of the game. Another primary expectation, is that the referee treat all players equally. If player A takes 3 fouls before being carded for persistant infrigement, and player B has been called 4 times for a foul before they are carded for persistent infringement; everything is fine. The number of fouls (3 and 4) are about the same. But, if we see a player being carded for persistent infringement after doing something twice, and another player fouling 6 times before being carded; we should suspect that the referee is not treating the two players equally, as 2 and 6 are (usually) not about the same.

  34. Okay, building on the idea that teams desire to win the league, to be first.

    The median winning point level is 87 points, and we may be able to ignore 75 (1996) and 95 (2004) as outliers. The data is quite sharply spiked, and it does tail off towards lower values (it is more likely to have a winning point level of 87-N, than it is to have a winning point level of 87+N). Our upper outlier is +8 on 87, whereas our lower outlier is -12 on 87. The ratio of lower to upper is about 1.5 (ignoring signs). As pointed out above, the maximum possible is 114 (+27) and the minimum is 38 (-49). This ratio is about 1.81, which is about the same as 1.5.

    We see a mean squared deviation of 19 from our median, which suggests a standard deviation of about 4.5. If 95 is 95 percentile, we get a standard deviation of about 4. It seems likely that 75 is about 5 percentile. The EPL is about 20 years old, and we have seen the league winner get as low as 75 or as high as 95 once (each) in those 20 (or so) years. So the two ways of guessing at some unknown distribution seem to be in agreement.

  35. As Brickfields isn’t here inserting jokes, a pause of my own:

    Two hunters are out in the woods when one of them collapses. He doesn’t seem to be breathing and his eyes are glazed. The other guy whips out his phone and calls the emergency services. He gasps, “My friend is dead! What can I do?”. The operator says “Calm down. I can help. First, let’s make sure he’s dead.” There is a silence, then a shot is heard. Back on the phone, the guy says “OK, now what?”

  36. The difference between a perfect season and our 95 percentile marker is 19 points, or about 6 1/3 games. A naive estimate for the number of losses the top team would have, just be chance is square root(38) times an unknown multiplier. Well square root(38) is about 6.16, so it looks like our unknown multiplier may be close to 1. And hence just be chance the top team manages to lose as many games as one would expect from Poisson statistics.

  37. If we look at how many wins the relegated teams get, we should be seeing the flip side. Or, about the flip side. The median number of wins for a relegated team is 8, and their appear to be outliers near 0.

    It would seem that it is slightly more likely for a weak team to win, than it is for a strong team to lose. Is there a real difference between these numbers? One can imagine human reasons for there to be a difference, but if the numbers are statistically the same, we shouldn’t entertain reasons for why they might be different.

  38. On the loss side I talk about the median, and on the winning side I talk about 95 percentile. If I look at the difference between the maximum 114 points and the median (87), the difference is 27 points, which is about 9 games. The winning side loses 9 games, and the losing sides win 8 games, is about the same. It would seem that random chance is equally affecting the best and worst in the EPL.

  39. Another reprive. Looking for jokes about not finishing first or last, this is the best I could do:

    From an anonymous contributor:

    A frog was finishing up its doctoral dissertation and had just stepped out from under its mud bank to get a breath of fresh air when it was surprised by a water snake. “Any last requests before I eat you?” said the snake.

    “Well, would you let me finish my dissertation first?” asked the frog.

    “That’s the silliest thing I’ve ever heard! What’s your dissertation on, anyway?”

    “Well, the title of the dissertation is ‘The Superiority of Frogs to Snakes and Herons.'”

    “I’d like to see that just for the sheer stupidity of the thesis!” said the snake. So he took the frog, went under the mud bank, and never came out again.

    A week or so later, the frog came out from under the bank to get another breath of fresh air, but this time it was snatched by a heron. “Any last requests before I eat you?” said the heron.

    “Well, would you let me finish my dissertation first?” asked the frog.

    “What’s it on?”

    “Well, the title of the dissertation is ‘The Superiority of Frogs to Snakes and Herons.'”

    “That’s the stupidest thing I’ve ever heard! I want to read it first — they say laughter aids digestion.” So the heron took the frog under the bank and never came out again.

    A week or so later, the frog came out from under the bank and yelled, “EUREKA! I’m finished!”

    So one of his fellow frogs came over and asked, “So, what’s the dissertation on?”

    “Well, the title of the dissertation is ‘The Superiority of Frogs to Snakes and Herons.'”

    “That sounds pretty tough to prove! Could I look at it?”

    When they went into the frog’s bank, they saw a word processor, a number of empty pop cans and food wrappers strewn about the room, a scattering of books lying here and there, a pile of heron bones, a pile of snake bones, a finished dissertation, and an alligator.

    The moral of the story: It’s not the title of your dissertation, but who your director is.

  40. A fantastic article and a great body of work . So simple yet so brilliant . Great job , guys , well done !

  41. ” And most painfully he had to stand up against a part of the club fans who don’t look further than their own little noses.”
    Walter , quoting your statement in the main article , those AAAA’s noses are no longer little .
    Since they are a deluded lot , they lie through their teeth and distort the truth ; are heartless unbelievers ; never hear nor heed the logic and sense ; close their eyes to the facts and figures ; but always and foolishly chose to mouth off ( when they really should be shutting the fuck up!),
    ergo , I now have come to the conclusion that they have all collectively developed the dreaded Pinoccio Syndrome.
    The good news ( For us AKBs at least !) is that they are no longer able to thumb their noses at us !
    Unless of course, if their upper limbs have similarly ‘over developed’ with all that wanking that went before !
    The Medical Division of UA will be presenting the official paper if and when the researcher gets his hands on some exquisite fine wine !

  42. Arsenal became a big club in the thirties and have remained so. We’ve never been so consistent over a long period definitely but let’s not pretend we suddenly became a big club in the late 90’s.

    Considering some of the rather ordinary players we bought after 2005 it’s certainly a great achievement that we maintained a high profile in the league. It does beg the question though that if we’d been a little more adventurous with our money we might have won the odd league title and probably not suffered financially.

    The Wenger factor is the reason why we stayed up there, after all he managed to get the team over the fourth finishing line despite having the likes of Squillaci, Senderos, Eboue, Chamakh and Almunia in the line up, to name a few. Not really awful players, just not the standard of players in the teams prior to the stadium build. I doubt if those players would have done much at all managed by anyone else.

  43. @ Chapmans Ghost October 29, 2014 at 9:39 am

    More like:

    It does beg the question though that if the referees were a little more fair with our players we might have won the odd league title and probably not suffered broken limbs.

  44. @ Mandy Dodd October 29, 2014 at 6:48 am

    Not only do I doubt Riley would want to meet Wenger or anyone from Arsenal, but I am of the opinion that we should not kiss ass just for the referees to do their job. Also, Wenger himself is still hurting from the Invincible match 50; he throws in a jibe about Riley every time he is asked about the Invincibles.

    The referees must just do their job. We don’t want any favours.

  45. Great article, unbelievable stat table. A big thanks to Walter and Mike (Not Riley of course) for a job well done.

    Imho, we should be able to ‘force’ figures like these down the throat of the mainstream media. How exactly, I don’t know. But there should be a way. Articles like this should feature prominently on and stat should be shared on all Arsenal official social media platform. We’ve got to start blowing own trumpets.

    And for AW, I share Bootoomee’s thoughts 200%…”As my hero and role model, it is impossible for me to respect or love Arsene more than I already do but on occasions like this, I wish I could.”

    Kudos Untold Arsenal for being here!

  46. Fantastic job and beautifully expressed. Well done Mike & Walter.

    There are several areas that can be included including: earnings per point where clubs like Man United should soar because of popularity; Value of clubs infrastructure; value of assets (player value);

    This table is wonderful as it is because its focus is ‘return on capital in points’.

    @Gord your comments need to be collated into an article as they cover a very interesting view of ‘crime & punishment’ as meted out by PGMOL.

  47. As all are aware, the top 4 places in the EPL determine which teams get a chance a Champion’s League in the next season.

    The median difference between 1st place and 4th place, is 16 points. The year that seen the highest number of points to the league winner, is the same year we observe the largest difference between 1st and 4th. Coincidence? I am going to suggest that all the money flowing into the league, is going to drive the difference from 1st and 4th towards 0, which means not only is 34 an outlier, but the second largest difference (30) is also an outlier, and is from the year before. One extreme value in the first year of Abramovich, and 2 in the second year? One wonders what other statistical curiosities exist in 2002-2004 (two seasons).

    But, in any event it appears that an average season should see the winner get 87 points, and the 4th place team should be at 71 points (almost 24 wins).

    Is UEFA league considered desirable by EPL teams? The median finish for teams in 3rd place is 11.5 points behind leaders (4th was 16 points back), a difference of 4.5. The median finish for 5th is 21 points behind the league leaders of about 5 points, which is about the same as the 3rd-4th difference. Four times a team playing in the UEFA league has been relegated, and an additional 2 times a UEFA team has been close to being relegated. The median finish of 6th place, is 24 points behind the leader. It would seem unlikely that teams are trying to avoid 5th place, but there has been a tendency for the team(s) going into UEFA league to have smaller rosters. It will be interesting to see if some of the newer clubs supposedly throwing money at the problem decide to make efforts at UEFA, as there are now supposedly more rich teams in the EPL than there are Champion’s League places.

    Not every country gets 4 places for Champion’s League. I can imagine that one of the barometers for fiddling with allocations by UEFA, is the ongoing presence of Arsenal in Champion’s League.

  48. Interesting article. I wonder what the table would look like including wages, and what the last 5 years would look like.
    I agree that the the extra points needed to win the league would have been more expensive and brought our ratio down, as the price of the best players rises out of proportion, however we haven’t been in a position to go that extra mile in the oil dominated market.

  49. Ignoring the first 3 seasons (22 teams), the median finish for last place was 28 points. And perhaps one should look at finished below about 19 points for unusual reasons. Years to look at would be 2002, 2005, 2007 and 2009.

    The median difference between 18th and 20th, is 8 points. There seem to be unusual circumstances in 2005(19), 2002(23) and 2007(25) generating excessively large differences between last place and 18th.

    How does one specify the relegation line? Is the the level of the 18th place team, or it related to the finish of the 17th place team? The level of the 18th place finish has a median of 36 points, and a MAD (Median Absolute Deviation, like a standard deviation) of 2. If we say the relegation line is 1 point below the 17th place finish, except when 17=18, we see a relegation line of 38 points. The MAD is the same (2 points).

  50. This is a very interesting post.

    However, I wonder if there could be more clarity on where these numbers come from?

    (I like the fact it’s overall numbers, which don’t rely on undisclosed transfer fees etc – I assume these are all available from the various clubs’ public accounts – but of course you should never assume!)

    Other aspects: this only covers transfer spending. As we know Wenger has been able to successfully ‘game’ the transfer market through his knowledge of other leagues and our scouting network. But a common criticism these days is that this advantage is now gone as other clubs have caught up in this regard. How would a table look that only covered the last,say 5, or 3 years?

    We also know that Wenger has been forced to sell players in order to keep up payments on the stadium for several years (£15m required every season prior to 2014 renegotiations of commercial income) – I refer to Swiss Ramble’s articles over the years for this.

    So naturally his net spend will be lower, which will increase his final points-vs-spending ‘score’. Is this a sign that he’s a fantastic developmental manager, able to ‘polish gems’, ‘create stars’, etc? I think it is – but is it possibly also something that skews the methodology?

    As another poster suggested: factor in wage spending. Now how does the table look? We don’t know individual wages but we do know the overall figures for entire clubs I believe, again via the public accounts. Make a table for just wages (the ‘Soccernomics’ concept that wages,not transfer spending, correspond to table position). Then make a table combining both.

    Factor in ‘football inflation’, as described in the Pay As You Play book. How does that change the table? What about exchange rates, when signing players from abroad?

    I’m not trying to discount this chart at all, I find the topic fascinating – however it seems that it’s a deep topic that merits far more scrutiny.

  51. Looking back at the observations that the winning team seems to lose 9 games a season and that the losing teams seem to win 9 games a season, combined with the fact that 9 games is about a quarter of a season (where we are now); a quarter of a season does not determine a season.

    Looking at the current standings, there is no reason to believe that any of Chelsea, Southampton, ManCity and West Ham are guaranteed to be in the top-4 at the end of the season, or that Sunderland, QPR and Burnley are doomed to relegation. Continued effort and success in one circumstance, lack of effort and failure in the other; will (probably) see the current situation continue.

    I would think about 14 games is where a decision needs to be made: if a team has aspirations of winning the league but at 14 games has (for example) 6 losses it would seem that it would almost require a bit of luck as well as a lot of effort to carry through on those aspirations. On the flip side, if a team gets to 14 games and finds they have (for example) 6 wins, it seems likely that they are not going to have to try too hard to avoid relegation.

    Except that shortly after 14 games, the Christmas rush comes along, and then injuries can take over.

    Metro is pretending to think, and had this:

    The article presents the idea that there is only a single factor involved in penalties. It is more likely there is a cluster of things involved. Metro seems to think club size is the determining factor (and lumps Arsenal in there). There is one truth behind penalties; if you don’t get into the box with the ball, you will not get a penalty. Park the bus is a poor strategy if you desire penalties. The only thing park the bus has to offer, is the counterattack.

    I am not sure why Metro works this around Crystal Palace, as I recently seen an article about Crystal Palace being in the process of getting a suagr daddy who would buy them a Champion’s League birth. We must have about 7 teams in the EPL all bound and determined to buy Champion’s League places, and there are only 4 places available. Add to that a team consistently challenging for one of those places on merit, and it seems likely that every year will produce 4 or 5 miserable billionaires.

  52. Keeping competitive in the top four when the oil money came in = miracle
    Keeping competitive in the top four when building a stadium = miracle

    Doing both of these at the same time is an achievement of biblical proportions akin to the ascension of the messiah, the parting of the seas and the building of the arc (like the reference 😉 ) all rolled into one.

    Apologies to anyone of faith that may take offence to these analogies

  53. In the refereeing news today, is mention that Clattenberg is being dropped (for a while, for a long time?). Somewhat to do with him driving himself so that he could attend a concert after the game, but possibly other issues. There was a different set of notes which mentioned that Stuart Attwell may be re-instated tot he Select Group of Mike Riley 😈 officials.

  54. @Gord
    Very interesting stuff, you should start your own blog.
    Regarding your 3:13pm post, can’t you look at past seasons and see after n games how the top four and bottom three correspond to the respective teams at the end of the season? I think 14 games is a little early, perhaps around 20 games.

    @Mare St Gooner, 2:53pm
    Of course you are correct (which others have also pointed out), wages should be included, and this data is publicly available, for example:

    As for ‘gaming’–a highly inappropriate word–the market, Wengers advantage due to his superior knowledge of players across Europe and scouting was prbably closed greatly by the time the oil money arrived, so was probably not relevant at the time.

    As for your fifth and sixth paragraphs, that’s the entire point of this article, that despite not being able to spend much, Wenger was still able to gain a relatively huge number of points. How is not being able to spend any sort of advantage? Your point is very confused.

    As for your second last paragraph, I’m not sure how any of those things would favour Arsenal specifically, as they would affect all clubs equally, so they are not relevant.

  55. A serious flaw in the Metro article linked above is that it gives the total over the history of the Premier League, during which time some clubs, like us, have never been relegated, while others might have been playing in the Championship for half of those seasons. We need to be looking at the average number of penalties awarded per season. A rather poor attempt at an intelligent football article.

    The Metro also has some guff about who will replace Wenger *next season*. Deluded bastards.

  56. Quincy. Ignore the Metro. Who can forget this classic by the dissenter Peter Wood entitled “Tottenham Give Arsenal a Lesson in Transfers”

    This is from the same guy who claimed yesterday on his anti-Wenger blog that Giggs and Scholes played for Tottenham. You have to laugh thst people like this are writing football articles about our club and yet have not git a clue what they are talking about.

  57. Dear Walter and Tony, how could you?

    The table has been based on Transfermarkt?

    The table has an obvious flaw, The Arsenal’s seasons are 11, and not 10 seasons. 17/11 = 1.55.

  58. Erm,
    (1) 2003/4
    (2) 2004/5
    (3) 2005/6
    (4) 2006/7
    (5) 2007/8
    (6) 2008/9
    (7) 2009/10
    (8) 2010/11
    (9) 2011/12
    (10) 2012/13

  59. Brilliant work.

    Arsenal FC are fortunate to have such a remarkable manager in charge.

    Kudos must also go out to Newcastle United. For all the rubbish spoken about Ashley and Pardew it would appear they also know exactly what they are doing.

  60. Quincy –

    The relative change in football inflation is important, because not all clubs have spent similar amounts in each year. We’re examining a 10-year period – if year-on-year the ‘actual’ cost changes, this will change the effect of the net spend in that 10 year period, thus affecting the final total. This may actually make Wenger’s achievements even greater, I don’t know.

    If you read some of the Pay-As-You-Play articles you’ll see that in certain years (eg just prior to new broadcast deals) prices go up dramatically, to the point where a player’s actual purchase price in the year previous can go down rather than up. I’m not explaining this very well but the authors do a good job of it, you can read some articles here:

    Of course their model is not perfect but it seems more accurate to me than simply adding un-adjusted figures.

    I greatly appreciate the writing done by the authors of this site – however, it feels wrong to blindly accept figures that are neither sourced (apologies if I missed the source, couldn’t see it), nor properly put in context. To put this table in context would make a much stronger case.

    As for the point on wages – yes it’s been mentioned but it can’t just be brushed aside. The fact is we’ve had a wage bill hovering around 4th or 5th and have (according to the Soccernomics theory) predictably been hovering (or overperforming by a place or two) in 3rd or 4th. This doesn’t tally with the frankly massive points haul that the table suggests Wenger has been capable of achieving.

    Surely the truth lies somewhere between the two – and that’s only looking at transfer spending and wages. As we know football isn’t just about money, although sadly that is a large part of it! Wenger keeps his squads small and has techniques for promoting team harmony. The money flowing into the game has damaged this, leading to the premature breakup of his teams. How do we factor these sorts of effects into this table? Does it mean his achievement is even more extreme, or does it mean that he has capitalised on the buying sprees of clubs such as City?

    I would like nothing more than to see this table properly ‘proven’ to the extent that it’s possible. However in this post I don’t feel it’s presented as anything other than a good, but unproven, hypothesis, which warrants quite a few more articles.

    Gord’s posts are interesting but a bit over my head I’m afraid, so to be frank I don’t understand if he’s expanded on the data in the table.

  61. @MareStGooner

    For most people, what I did just provides interesting background. To use any of it with Mike’s work is probably at best non-trivial. It was more meant in the spirit of useful data and methods for all Untold work. Particularly the use of medians instead of averages, and MAD instead of standard deviation.

  62. Thanks for the reply Gord,

    I take it you are a statistician or mathematician of some sort? I’ve re-read your posts with interest but not full comprehension – ah well, someday perhaps. Do you have a blog yourself?

    It doesn’t factor in wages but I found the following quite interesting:

    I disagree with the author’s notes (for me the Invincibles will always be the best, but then of course I’m biased), but it’s quite interesting sorting the table by different parameters. Shame they can’t be ‘cascaded’ (if that’s the correct term – applied in sequence).

  63. Great work and as others said before – an eye opener.
    Two reservations though:

    1. For meaningful analysis the value of the teams at the starting point (that is ten years ago) should be taken into account. In the next ten years Chelsky’s or Sheikhs’ fan could make a similar analysis and these clubs could by then turn out to be perfectly balanced from financial point of view, each point could cost zero pounds! I don’t know whether the future managers of these clubs should be given credit for it.

    2. I have a feeling that if we calculated how much each trophy cost the story would be very different.

  64. Mare St Gooner

    You can always go into more detail. For example, would you use the official inflation, or real inflation? The two often do not coincide. In the case of Arsenal it wouldn’t make much difference, firstly because the nett spending over the 10 years is low, and secondly because the bulk of that would have been near the end of that period. It would probably affect a club like Chelsea more than us; but you are correct, factoring in inflation would give a more accurate analysis. If those figures are taken from Transfermarkt, I believe they are already adjusted for inflation.

    I’m not familiar with this Pay as You Play thing, it seems to be a model of Tomkins Times, meaning of course that there are other admissable models, and that this is not necessarily the final model.

    And I don’t think I brushed aside the issue of wages, in fact I agreed with you wages should be included.

    And I don’t really agree with the old cliche that Wenger likes a small squad. Compared to most teams in Europe, our squad has never really been small, only in comparison to clubs like Chelsea, Man City, Man U, etc., and that requires no explanation.

    Of course there are many models, some done by economists or statisticians, but this article is just to give a simple picture of what was going on with our transfer spending in relation to our success on the pitch.

  65. I don’t think Gord’s posts are specifically about the stats themselves. The differences between mean and median, standard deviation and mean absolute deviation are quite well know, it’s more the application to football that I found interesting.

    As for Mare’s October 30, 6:31 pm post, that table doesn’t take into consideration any sort of spending, merely results. And I can’t help but laugh at a website about football calling itself objective, even if it is a stats blog.

  66. Right, we can always go into more detail. And of all sites, this one should. (Apologies if this is becoming a circular debate, this will be my last post on the topic.)

    Take for example the yearly ‘price of football’ articles, which have in the past stated the most expensive tickets without regarding the number of games included (Arsenal), or stated the cheapest tickets, without mentioning that these tickets are virtually impossible to buy and sold in very small numbers (City, Barcelona, Bayern Munich etc).

    For me the way the table in this topic is presented is a bit reminiscent of that, I would like to accept it but I find myself resistant. A site such as this one, which has done excellent investigative work, shouldn’t fall into the same traps that it often decries!

    A similar analogy might be the wonderful work of the historians Andrews and Kelly. If the transfermarkt figures are equivalent to, I dunno, the Bernard Joy book ‘Forward Arsenal’, these historians instead pursued the newspaper clippings of the time rather than a secondary source, thus shining light on topics that had been portrayed in a one-dimensional manner. (Apologies, that’s a weird analogy. Ah well, I’m leaving it, in a stream-of-consciousness sort of way.)

    I think you might find the Pay As You Play model interesting – and the further work done by other authors on effective squad costs, and multiples of the median required to win the league.

    They calculate so-called ‘football inflation’ (which obviously greatly outstrips normal inflation) by using the yearly average cost of a transfer, and its relative increase or decrease. I’m sure holes can be picked in this methodology, and I’ve probably greatly over-simplified it – however, it still throws up many interesting ideas and concepts, much as the table in this post does.

    Anyway, the result is that spending in the earlier part of the decade will be mostly adjusted up. I am guessing that this type of ‘correction’ to the raw figures will reduce the points/spend total Wenger has racked up, and might make it tally closer to the wage = league position type of ratio. If the table spanned back another couple years, to 2001-02, we’d find ourselves with the most expensive squad in the league if I’m not mistaken.

    A side note: I’ve often thought the period when City were buying our players was a very strange one. On the one hand they were weakening us, perhaps a clearly targetted methodology (target the team most likely to tumble from the top four, and buy their players, strengthening yourself as you weaken them). At the same time, their cash injections kept us afloat. This is what I am trying to get at by suggesting that having a good net spend is not necessarily a good thing, even if performances were above expectation!

    Apologies again for such a rambly post. I’m just thinking aloud I suppose. I’ve been greatly saddened by two things: the failure of ‘project youth’, and the loss of our plural ownership. Perhaps it’s made me a bit cynical.

    Ah well, good night to all, and enjoy the game wherever you may be watching it. (But do look into those Pay As You Play links! Even if you disagree they are well worth the read.)

  67. Hi Walter (and everyone else who’s said nice things about my work with numbers!). Sadly I’ve been out of circulation for a few days so missed this when it came out. Nonetheless, I have to say that gathering statistics is merely a task that anyone can do; it’s the boys at untold that bring it all to life. Keep up the good work guys and maybe I’ll send you some more numbers one day 🙂

Leave a Reply

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