But, who gets to decide who is the most convincing  

Right....   So what option to we have to decide from?

Fully agree with that. I thought what you were saying before was to keep the bonus points, but then remove them if there is a tie (ie, person who did better overall wins rather than the person who just tied because of the bonus points). I agree with this new suggestion however (maybe it's the same suggestion and I misunderstood previously). Use the final prediction as a tiebreaker rather than points. So if 2 people finish tied, and one got the final correct, he wins. You would just need to define what "correct" means in this case. Is both players and the correct winner enough (surely it's hard enough to get that so there should be some reward for doing it), or do you need the exact amount of sets too?



Pretty much exactly the point I've argued. With the "closer" final thing, there will be plenty of times you can't separate the 2. That is if they had the same final prediction (not an unlikely occurence given almost everyone picks 2 out of the same 4 or 5 players for their final anyway, and especially if Federer, Nadal and Djokovic are in the tournament), or if they picked players who reached the same stage of the tournament but not the final (again, also not unlikely to both pick a semi and quarter final loser for instance). It's worked in this case, but it wouldn't have taken much for it to not work. This will happen more often than not IMO so most of the time we'll be saying it's a tie. If you look at previous week's performance, there will never be a tie.

And by the way, noone argues that looking at the previous weeks result isn't the easiest or that it won't guarantee an answer, its just that its unfair to use last weeks results for this week.

I don't think it's unfair. No more so than the "closer" final thing anyway.
If anything, I think it encourages you to keep going even if you're having a bad week.
You might be near the bottom and out of the points for that week, but it's worth continuing rather than giving up as next week you may end up tied with someone else who is low down with you this week. I just think that if you want a tiebreaker, it's purpose is to break ties. And hence you should use one that guarantees being able to do so rather than one with a relatively high likelihood of still ending with a tie.

Fair point, but I don't really think it's a comparable situation.
If it was a draw at full time you would have extra time and penalties which guarantees a winner. This is the key point. If the previous week is not the way to go then fair enough, you need something else that guarantees a winner. The closer final thing doesn't do this and that's what I've been trying to say.Iit's a not a good tiebreaker for this reason. I've used the previous week as my suggestion because I don't have another idea to suggest. The main point though is that it gurantees a winner is performs better as a tiebreaker than the closer final method. Perhaps someone else has some other fairer idea that gurantees a winner.

Hmmm yeah those should guarantee an answer. I like the idea of the most points in a single day.
I still don't like the least completely correct one though. IMO predicting a winner is easier, and more common than predicting the winner and correct number of sets too. So I think the exact result should be valued more highly given it's harder and less common.

But they still both work well as tiebreakers.

But on the other hand, if you got more exact predictions, it means you got less winners right, so you did worse on an easier sort of prediction! Do you understand what I mean? I haven't worded it very well.
My main reason why i prefer what i said is because i feel getting the amount of sets right is more guesswork then getting the winner right, so i value getting more winners right, rather then getting the sets right...