The Premier League Meets Away Goal Tiebreakers

By Julian Ryan

When soccer was invented, all goals were created equal, with the team that scores more goals being declared the victor. However, the UEFA rulebook strays a bit from this basic guideline by implementing the well-known “Away Goals” rule, whereby over two legs, if the total number of goals is level, the team with more away goals is victorious. The rationale behind this is to try to avoid extra time if possible and ultimately avoid penalties – which we can all agree is a cruel way to exit a competition.

This occasionally results in particularly bizarre incidents, such as when AC Milan beat Inter Milan on “away goals” despite playing both games at the San Siro. In general, the system can sometimes seem kind of silly.

I wondered what would happen if the Premier League, for no apparent reason, adopted this strange two-legged, away-goals-tiebreaker format.

Rather than using the standard thirty-eight game format with equal weighting on home and away results, I have recalculated the 2012-13 season league table by the following system: Each team’s two games against each other team is considered as one complete fixture with a home and away leg so that each team plays a total of nineteen “games”. A win is awarded for beating the other team over the two legs, including the “away goals” rule as a tiebreaker. (But ties do occur if the score was identical both home and away). Ties were considered a half-win, and then if overall score at the end of the season was the same, head-to-head results were used. This produced the following new league table:


Record Actual “Points”

Actual Finish

 1. Man U 18-1-0 89 1st
 2. Arsenal 15-3-1 78


 3. Chelsea 15-4-0



 4. Man City 14-4-1 73


 5. Tottenham 13-4-2 72 5th
 6. Everton 12-7-0 63


 7. Liverpool 10-7-2 61 7th
 8. West Ham 10-9-0 46 10th
 9. Southampton 8-11-0 41


 10. Aston Villa 8-11-0 41 15th
 11. Stoke 6-9-4 42


 12. Sunderland 8-11-0 39


 13. Norwich 6-11-2 44 11th
 14. Fulham 6-12-1 43


 15. Swansea 6-12-1 46 9th
 16. Newcastle 6-12-1 41


 17. Wigan 6-12-1 36 18th
 18. West Brom 6-13-0 49


 19. QPR 5-14-0 25 20th
 20. Reading 4-15-0 28



The gulf between the top seven and the rest of the league remained, with no changes at the top and another runaway league title in Fergie’s final season. However, despite a very credible 8th finish in reality, under this system the poor Baggies end up getting relegated. Their formula of close home wins and poor away performance saw them come up short. Stoke, traditionally thought of as very strong at home, hold the dubious record of most ties for the season with a very impressive four. Finally, QPR’s strategy of only winning four very carefully chosen actual games but ending up with five victories pulled them above Reading but not out of relegation.

There is still a 0.82 correlation between rankings under the two systems and their outcomes are not particularly dissimilar, though the away goals system does seem to keep the fight to avoid relegation very close, with four teams within half a game of the drop.

In the end, any set of rules for determining order of finish is arbitrary to some degree. When we go back and evaluate the success of teams at the end of each season, we often forget to account for this fact. The format that a league chooses for scoring, seeding, elimination, etc. can randomly elevate or depress the fortunes of any given team. At least that’s what I think the takeaway is from this. It’s either that, or the fact that a kid stuck in Thanksgiving traffic with a spotty WiFi connection will do anything to entertain himself. But mostly the first thing.

About the author


View all posts


  • One factor you have overlooked in your proposed scheme – and it’s an important one – is that strategy and tactics of the 2nd match in the series would be greatly influenced by results of the opening match…much as it is in Champions League play. That’s a major flaw to any system that anoints a victor based upon goal differential (or total goals). In my opinion, all wins should be equally weighted regardless of score. Too many nuances throughout a match to think that score alone is a true reflection of the difference between the teams.

    For example, the winning team from the opening leg would undoubtedly structure their play for a tie and/or a low scoring match, especially if they were the benficiary of a win by two goals or more in the opening leg of play. Doesn’t that take away from the question of which team is better?

    As I see it, the winner of any home and home series should be determined on the field. And the best way to achieve that when a series is split (one win apiece, or two ties), is to play a 3rd match at a neutral site. It may not be practical but it makes alot of sense. Rules that incorporate total goals (or away goals) instead completely miss the point.

  • This was fun to read! But one should keep in mind that in genuinely two-legged matches, teams will chase an equalising goal – or one that turns a loss into a win – if they know they need one. In this analysis some teams would be defending a lead and not realising they’re on the short end of this retrospective analysis.

    More disappointingly – or entertainingly? – for soccer, individual team scoring averages < 3 per games and follows a Poisson distribution, leading to victory at times, for underperforming teams.

    Tangentially related:

    • To be clear: this wasn’t meant to be a better approach to ranking teams. Of course strategy would change if teams knew this was the format. It was just meant to be a fun post. If anything, I would actually be in favour of getting rid of the away goals rule in the champions league

Leave a Reply

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