nzfooty wrote:Who said it was fair and square? Ah, you did. Possibly the entire Swedish team went down with food poising on the day. Perhaps the referees decision sent them the wrong way. Perhaps they had key suspensions etc. etc. That's why you look at the total 270 minutes, not just the 90.
That's rather the key point of cup football though isn't it? Barcelona may have done better against Bayer Leverkusen than Chelsea did, but Chelsea won the head-to-head. It doesn't mean that Chelsea are scientifically the better team, or Barcelona don't play the more attacking football, but it does mean that Chelsea deserve to go through , because they won the head-to-head fair and square. We don't decide the European Champions by a marathon league season, it's a cup tournament. The groups are there to make sure everyone who travels gets a few games, but the knockout mentality, best team on the day, is still very much part and parcel of becoming champion. If you're comparing with 36 other matches over an entire year then maybe it is fairer to look at the bigger picture, but we're only comparing with two other matches and the possibility for large unfair fluctuations is still very much there. Anyway, it was 3-0 in the example, imaginary Sweden can have no complaints.
I don't expect people are following the details of team A beating team C hypothetical examples (you're not even following them), but others obviously are reading the thread and others might be interested in the general topic of whether head-to-head is fairer than goal difference, particularly after last night. We're not the only football fans discussing this topic at this very moment.
nzfooty wrote:It also does not offer incentives for attacking football. It leaves teams fearful of losing.
Your glass is half empty. If you beat a team in the first match you're almost guaranteed to finish above them.
nzfooty wrote:I am looking forward to seeing what "3rd system" you come up. Good luck.
A modest proposal: If teams end level on points then for every non head-to-head match they win by one goal deduct one tie-breaker point and for every non head-to-head match they lose by one goal add one tie-breaker point. If that doesn't separate them, then do the same by changing each result by two goals. Keep going until your total goals added or subtracted is equal to the highest score difference.
That would see Spain through in your example if they beat Turkey by two goals in the final match.
That would see Norway through in my example.
That would mean team A hadn't necessarily won the group in your other example. B win 2-0 and A lose and B top the group.
The philosophy behind it is that matches decided by one goal are more "flukey" than matches decided by five goals. It gets at the "marginal goal" concept, what would've happened if. In my example it would send Norway through no matter how many goals Sweden scored against Turkey. It would also send Greece through last night. It would've knocked Italy out in 2004 but without the stitch-up accusations. It's not particularly simple and it's not clear what should be done with draws, but I think it covers all our objections so far. I actually still think I prefer the head-to-head rule though. In a three-way tie with everyone beating everyone, you would still end up going to results against the fourth team. Maybe if you ran it first on the head-to-head matches only, and then ran it on non-head-to-head. Getting complicated...