Why Goal Difference Is Not the Best Tiebreaker for the Premier League Table

Caught your breath yet from Sunday’s Premier League season finale excitement? What a way to end the 2011-12 season. After 38 games, the table doesn’t lie, or so says the time-honored cliché, and Manchester City is deserving champions.

With that said, I’ve been surprised at how little outcry there has been, or discussion at the least, about using goal difference as the first tiebreaker when determining final positions in the league table. I’m not exactly sure why that method has been so accepted for so long and not questioned at all. Could it simply be because that’s how things have always been? If that is the case, that’s ridiculous.

By looking purely at goal difference, we gain, admittedly, a pretty decent picture of how the table is ordered by points. Logic tells us that the best teams win more games, outscoring their opponents and thus accumulating a better goal difference. On the flipside, teams near the bottom will have lost more games and thus accumulated a worse goal difference. Just as a broad, general tool, goal difference can be a useful tool to capture the state of the league.

Broad and general is not good enough, though, when millions of pounds in prize money and potential European qualification places, as well as relegation, is on the line. Using head-to-head results is by far the better method to order teams, one that actually prioritizes what the tied teams have done against each other. If teams are tied, and the definition of “tie” is that they are supposedly equal, surely the best way to determine who is better would be to look at how they fared on the field. Goal difference can be fluky, and significantly altered by just one or two good or bad results with high-scoring final outcomes. But it must be said that whether goal difference or head-to-head results were used Sunday, Manchester City still would have been crowned champions.

Liverpool, however, finished on 52 points, good for 8th-best in the league but exactly equal with Fulham in wins, draws, and losses. Even though Fulham beat Liverpool twice this season, they finished in 9th, behind Kenny Dalglish’s woefully disappointing outfit. Why? Because Liverpool managed a superior goal difference.

West Brom, who had a terrific season under Roy Hodgson, lost to Swansea home and away but finished ahead of the Swans on the basis of one goal scored, the second tiebreaker behind goal difference. Norwich City, who like Swansea had a fantastic first year in the Premier League, tied on points (47) with both Swansea and West Brom, but finished behind BOTH of them despite defeating Swansea twice and splitting victories with West Brom. That dastardly goal difference again, up to its unjust tricks. If head-to-head was used instead, Norwich (with 9 points gained from their 4 games against tied rivals) would finish in 10th, ahead of Swansea in 11th (with 6 points gained) and West Brom in 12th (with 3 points gained).

According to the Liverpool Daily Echo, figures from the 2010-11 season indicate that each higher place is in the standings is worth an extra £800,000 in prize money. Fulham, for me, should be awarded that ahead of Liverpool, and Norwich an extra £1,600,000. On the flip side, of course, Liverpool would miss out on £800,000, and West Brom would lose £1,600,000, a nice bit of cash for a mid-table outfit.

You may think to yourself, “£800,000?? That’s nothing”. In some cases, to some teams, you may be right. However, Sunderland bought the outstanding young left winger James McClean for £350,000 from Derry City last August. Shaun Maloney, who had a terrific game for Wigan against Manchester United in April — getting a goal and helping lift Roberto Martinez’s side out of the relegation zone — cost the Latics only £1 million. Swansea paid £1.6 million for Michel Vorm — one of the best goalkeepers in Britain this year.

Good, inexpensive players can be found through scouting and development. Extra prize money can go into the transfer kitty, pay loan fees, or help pay wages.

It’s not just prize money in the hundreds of thousands of pounds or few million. What if goal difference decided even more valuable positions like the difference between staying in the Premier League or being relegated to the Championship? Or what about the financial and prestige difference between playing in the Champions League and playing in the Europa League, or even playing in the Europa League and playing in no European competition at all?

If you take a look at the final Premier League table this season, simply organized by goal difference rather than points, you’d find an interesting result. The top four and bottom four teams would stay the same, and that’s to be expected.

However, Newcastle, who everyone would agree had a great season with Premier League Manager of the Year Alan Pardew guiding a team built on cheap signings, finished 8th instead of 5th. Sunderland’s revitalization under Martin O’Neill helped boost them to the 9th-best goal difference, but they still finished 13th in the league. Aston Villa, my favorite club, was simply awful to watch this season yet still managed the 14th-best goal difference, two places ahead of their actual final position. Stunningly, they were only two goals in difference worse off than Norwich, who had a far more impressive season and should have finished 10th, as I mentioned earlier.

The point is, goal difference simply just doesn’t matter that much, at least not enough to be the first tiebreaker when there are, potentially, millions of pounds on the line. Head-to-head results are a better option, one that some other leagues in Europe have been smart enough to implement, including Serie A, La Liga. If it’s good enough for two of the best leagues in the world, I don’t see why it shouldn’t be good enough for the Premier League as well.

38 thoughts on “Why Goal Difference Is Not the Best Tiebreaker for the Premier League Table”

  1. Head-to-head only takes into account 2 games out of a 38 game season. Goal difference takes every game into account. Every team played the other teams home and away so there’s no unjust reason to use goal difference as a tie-breaker if each team has earned the same amount over the full 38 games.

    1. Exactly right. Why reward Fulham for 2 good games when Liverpool must have outperformed Fulham for the other 36? I am in no way a Liverpool fan, just using the example provided.

  2. One point in goal differential’s favor is that it keeps teams pushing for a full 90 min. Every minute and every goal counts and to the neutral this can heighten entertainment

  3. Goal difference is superior to ties, because it’s a better measure of the team performance during the course of the season. Any match can be lost on a fluke and should, in my opinion not be used in a league as a measure of performance. Cup’s on the other hand is another thing, there you have to be at your best at a specific time, and thus, is not played with a league table.

  4. I would like the first tie-breaker to be head-to-head, but I think that it won’t change because those in power seem to think that using goal difference is supportive of teams like manchester united who, because of their relentless style of play, usually get have an extra bonus if they are tied with other teams.

    This is my speculation and it doesn’t necessarily hold up this year (as man u lost title on g/d), but the point is that making the first tie-breaker goal difference is a statement and vote of more attacking style in the minds of decision-makers. They may be right, but I think they won’t change because to change to head-to-head would be tantamount to endorsing more negative playing styles (which then be perceived as a change that betrays the traditional English values in the game).

  5. While the argument is well articulated I disagree and hope it never happens.
    Here are a few thoughts; Liverpool above Fulham and West Brom above Swansea seems wrong but what GD does is show that Liverpool’s overall performance was more emphatic, that the Liverpool defensive performances have been consistently better, maybe Liverpool defend from the front or the formation they play or tactics used are superior? I am a big fan of discernment but this method is more snapshot and only looks at 2 games for 2 specific teams not the body of work for the season.

    eg…I know there are a lot of people that believe Chelski are a great team (IMO delusional looney tunes for the most part) and using the method you propose the 2 leg fixture in the UCL would suggest that. However, I think that if people are honest they would NOT say that Chelski are a better team than Barca, they just match up well.

    Your idea works for cup competitions but not for a league. The special one himself said before the Bayern tie that the UCL doesn’t truly say who the best team in europe is.

    That’s why there are cup competitions because that way a team can have their day and have a run at the cup but the best team wins the league, and great teams win the double and just off the wall looney tune teams win the big three treble (domestic league, domestic cup and the UCL).

    All american made sports use the game tie breaker (as you are suggesting), and are setup like the UCL, a period of league play that qualifies you for a play off, with a little variance depending on the sport (NFL single knock out games, with baseball, basketball and hockey using a “best of x series”). Which isn’t really the best way to determine the “Best” team over a 9 month season.

    There is a certain purity to seeing who can perform consistently in the league with all the different cup competitions and internationals causing distractions.

  6. “Exactly right. Why reward Fulham for 2 good games when Liverpool must have outperformed Fulham for the other 36? I am in no way a Liverpool fan, just using the example provided.”

    Jack, they finished with the EXACT same record (14-10-14). How exactly did Liverpool outperform Fulham by finishing with the same record, the same points, and losing twice to them? If two teams are tied that closely, surely looking at how they did against each other would be helpful?

    1. I think you answered your own question. Because they have the EXACT same record, the fact that Liverpool had a much better GD would seem to be the ONLY way to judge which team was truly better on the season. If we have to differentiate between two records that are exactly the same, we have to have some way to quantify the quality of the wins and loses. What better way than an aggregate score for the whole season?

      1. Between Fulham and Liverpool, Fulham was the better team, because they beat Liverpool twice. It’s pretty obvious.

  7. Micheal – fair points but as thew other posters have said: Goal difference in my humble opinion, is a far better initial tie-breaker. If I’m not mistaken – it would have gone to head to head if the goal difference were the same for both teams. It’s different in every league, I think in Serie A – it goes to head to head first.

    The whole point of football is to score goals and secondly, to avoid conceding them. A team that does the better job of doing so over the course of 36 matches, deserves higher praise.

    1. I agree that goal difference is best, but head to head is not the “next” decider. I believe it is goals scored. Look at the final table and the positions of WBA and Swansea. Swansea beat them twice–3-0 and 2-1. They both have -7 goal differences. However, WBA scored 45 goals and Swansea scored 44. Thus, WBA end up 10th and Swansea 11th. That’s the way it goes.

      No system is perfect, but I think the Prem has it right.

  8. EPL, Bundesliga and FIFA World Cup use goal difference

    La Liga, Serie A, Champions League & Europa League group phase and European Championships use head-to-head

  9. So, if two teams are tied with the identical record…but one team has a higher gd by 2 or 3, they get the spot because they performed ‘better’ during the whole season? Perhaps they played a match against a team that had 2 guys get red cards, which allowed them to score 6 or 7 times. By your theory that one match means they played a better ‘season’ than the other team because of the gd. I disagree.

    gd is too variable. A better 1st tiebreaker would be wins. Two teams can tie on points, but let the team with the most wins take the spot. If those are equal, then fewest losses. If still equal, then head to head (since you are comparing teams that had equal seasons to this point). if equal after that, THEN move on to gd.

    None of it would’ve mattered this year, which is fine, but I’ve always been against gd as the 1st tiebreaker.

    1. its Goal Difference not Goals scored, the “win” column makes no sense, a team dominates for 89 minutes and a fluke goal or own goal or garbage penalty can cause a win. Goal difference looks at attack and defense.

    2. If teams are tied on points and wins, wouldn’t they have to have the same number of losses as well?

  10. This whole article makes very little sense. As articulated by the other comments, goal difference is the only realistic option, hence why it has been used for many many years. Surely if what you are suggesting were better it would have been put into place already? This discussion has been had numerous times and the outcome is always the same.
    Another person who doesn’t really know what they’re talking about.

  11. C Hamilton,

    If it’s (goal difference) the ONLY realistic option, why are other major leagues/competitions not using it anymore, and using head-to-head results, as I suggested?

    1. what other major leagues and tournaments? the american ones that are not set up like english football. The most popular sports in the worlds 3 marque competitions the world cup, the EPL and the Champions league all use GD. as I stated earlier, us major league sorts don’t have a real league system there all knock out competitions.

  12. Here’s a shocking suggestion: how’s about Man U. and Man City actually play each other to decide the title?

      1. This would speak against the Article as well. The EPL should look at the tiebreakers via head to head… Man City won 2, Man United lost 2. Hence little controversy.

      2. So if City is so much better, how did they end up tied in points with United? The fact that they ended up tied means that season should extend for one extra game between the teams who were tied. Tiebreakers to decide championships. Welcome to soccer. Or I suppose you could decide the entire season based on penalty kicks if that would make you feel any better.

  13. Goal differential makes every minute of the match relevant. Or at least more relevant than it might otherwise be if there was no such tiebreaker and one side was up 3-0 or some such.

  14. Dust,

    The Champions League uses, as its FIRST tiebreaker, “a) higher number of points obtained in the group matches played among the teams in question”, followed by goal difference. The same is true for the World Cup, so check your facts. La Liga and Serie A use head-to-head results as their first tiebreaker as well.

    Fair enough if you say that goal difference is a better tiebreaker, we can agree to disagree on that. Don’t say that “The most popular sports in the worlds 3 marque competitions the world cup, the EPL and the Champions league all use GD” because that’s patently untrue.

  15. It’s a hard call to make. I don’t love goal differential, head-to-head seems more appropriate, but the counter is that one or two results (which could be flukey) over the course of a 38 game season don’t really tell you anything.

    2011-12 Manchester City isn’t a great case for this discussion, they dominated, most goals scored, fewest goals allowed, two wins against second place United. No disputing their title. I just don’t understand why it’s goal differential (net difference) that’s the obvious first tiebreaker.

    Say some team was just crazy efficient, dominant, didn’t score a ton of goals but rarely ever conceded. Let’s use a ridiculous example – say some other team was in the league this season, 1 goal allowed all year. Team X, brick wall of a defense. Beat City 1-0 once, 0-0 draw the other game. Won 26 games 1-0, lost one game 1-0, eleven 0-0 draws. 26 Goals scored, 1 goal allowed, +25 differential, basically impossible to score on, and they beat City once. But they finished with 89 points and City was +64, so City wins. Why? Is having a +64 goal differential more impressive than +25 while only allowing one goal all season, and going 1-1-0 against City?

    Why not measure the number of goals allowed as a percentage of goals scored rather than net difference? In the example above, Team X allowed 1 goal, or 4% of their season output of 25 goals. City allowed 29 goals against 93 scored, or about 30% of their season output. Should that matter? I think that as an indication of quality, it’s at least as telling as net difference. The problem is it would encourage a boring defensive style of play while punishing more wide-open, high scoring teams. In that sense it’s not better, it would still be style-points, just favoring a defensive style over an attacking style. I wouldn’t advocate it because it’d encourage conservative, boring play – but I don’t see why it’s actually any worse than simple goal differential.

    There isn’t a perfect system. Again, I’m not a huge fan of simple goal differential. I would prefer head-to-head while conceding that it’s not actually a better system, just more entertaining. Those Top-Four games would carry extra weight.

    1. Many leagues, including England, used Goal Average for years. It stopped because it generally rewards defensive play. 1-0 is an infinitely better score than 4-3 in goal average, goals dropped across the board, so they went back to goal difference.

      I’m torn on this. I like the idea of head to head but it gets tricky with 3 or more teams and I can see the whole season v 2 games logic, I like GD and I have a small inkling for simply making it goals scored just to encourage attacking play

  16. Michael,

    So I checked my facts, and admit I have to amend my statement…”the 2 most prestigious competitions for the worlds most popular game, The World Cup and the EPL use GD as the first decider”.

    What! I hear you scream, how can that be?? the idiot has missed the of another one!, well calm down kermit, don’t give yourself a hernia, let me explain.

    I can’t find a doc on the FIFA site like I misread on UEFA’s. (not that surprising considering the shady nature of Blatter and co.) However, I did find this

    http://en.wikipedia.org/wiki/2010_FIFA_World_Cup (scroll down to group stages)

    Teams were ranked on the following criteria:
    1. Greater number of points in all group matches (This just means point totals, as there are not home and away legs, you play each team once in the world cup)
    2. Goal difference in all group matches
    3. Greater number of goals scored in all group matches
    4. Greatest number of points in matches between tied teams (this is what UEFA CL use as the 1st tie breaker)
    5. Goal difference in matches between tied teams
    6. Greatest number of goals scored in matches between tied teams
    7. Drawing of lots by the FIFA Organising Committee

    And to your point here, is an example of the UEFA CL from this season, with Apoel ahead of Zenit in-spite of them having the smaller GD . So, Zenit scored more and conceded less but because of the stupid head to head Tie breaker, Apoel had top spot and the easier rounds ahead of them–You see, proof! it is stupid and doesn’t make sense, and looks strange. don’t forget the whole point of football is to score more than the other team, so on that fundamental principal alone the head to head is stupid.

    (i hope the formatting of the table is ok)

    Group G

    Team Pld W D L GF GA GD Pts
    APOEL 6 2 3 1 6 6 0 9
    Zenit St. Petersburg 6 2 3 1 7 5 +2 9
    Porto 6 2 2 2 7 7 0 8
    Shakhtar Donetsk 6 1 2 3 6 8 −2 5


    This may look normal if you follow the Ponzi or Madoff school of accounting, but its not right given the fundamentals of the game. So, not quite “patently false” but your passion is adorable.

    Please don’t hesitate to let me know if these writings of mine are also patently false so I can amend my previous statement to read “The home of football, the original and most incredible, supercalifragilisticexpialidocious league of the worlds most popular sport uses GD”

    1. Yes if you are EQUAL IN POINTS it is only natural to look at common games.

      “We are equal in points and I beat you twice!! But wait you know that team that we both beat badly four times! I beat them just a little more badly haha so I am better than you”


  17. This article misses the point that the game is supposed to provide entertainment for the fans. How we do we do that? Simple – encourage and reward the scoring of goals. Usually goal difference is the forgotten decider lurking somewhere in the background until this season. There have been all types of ideas put forward to encourage goal scoring from extra points for more than three goals etc etc but this year has shown none are needed. Man City scored a boatload of goals and their defence prevented a boatload going in. Therefore, it is as good a system as we’re going to get (and understand) without employing Harvard geniuses to come up with some sort of system relying on statistical hell (typical of what plagues American sports today). The season is long enough for goal difference to be a sensible decider and I don’t hear Man U crying “foul” about the system.

    We don’t need any more half baked schemes thought to “improve” the game. If I remember correctly it was the US that floated the idea of the 94 World Cup games being played in quarters for advertising purposes, and more recently the frenzy about youngsters under the age of eleven or something should be required to wear helmets in case they head the ball and damage their brains – come on!

  18. H2H seems intuitively fairer but as many have pointed out, taking 38 games into consideration is going to give you a better statistical assessment a team. Goal difference is not perfect but its probably a better indicator of what a team is likely to do. Statheads would tell you…. better than points. If little changes in the Northeast put a few quid on on Sunderland (who were unlucky based on GD) to pass Newcastle (who were pretty lucky based on GD).

  19. Goals scored seems a better tie breaker to me if you want an exciting league. Goal difference just tells you how bad your defence is with respect to your offence.

    1. Not true!!! An opportunistic team that plays defense and protects a lead of 1-0 can be a great football team. Yet they are punished by the system. 38 games the first tie breaker should be head to head then total wins

  20. I am assuming you’re American due to your deluded logic and incorrect grasp of the British leagues. It all makes sense now. Stick to your own sports kiddo.

  21. I am a true fan of all sports world wide. I have never agreed with the tie breaking system in football. It is seriously flawed!! Those of you that think Goal difference is better support this very possible scenario for me.

    Man United 89 pts and Man City 89 pts.

    Man City has one more in goal difference. However, Man United has two more wins and has beaten Man City twice. How on earth can you tell me Man City deserves to win the league.

    It is even worse that they follow this format in the Euro 2012 round robin. Four teams and three games. It is common that teams can be equal for second and third place. You are only playing three games so head to head must be the tie breaker!!! But it is not. Who cares who scored more goals against Ireland!!!

  22. Goals buddy. 1. If you can’t score you can’t win — that’s simple in nature. 2. Why do there is a away-goal rule ?because to prevent teams from defending too much……. same logic to GD rule

  23. @Nick
    > … so head to head must be the tie breaker!!! But it is not.
    Oh yes it is — that’s why Russia are going home with 4 points and a much better goal difference than Greece, but Greece beat them on the day. And that’s OK with me.

    It does seem a bit crazy to think, though, that if Germany had lost their last group game, having won their first two, that they still could have ended up going home. It does make things kind of exciting….


Leave a Reply

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