Teams with same score should be tied! (until the tiebreaker kicks in of course)

Submitted by mjf on 18 April, 2009 - 18:56

Currently if you look at a league, you'll notice that teams with the same score, in fact the same team components, are being scored not at the same rank, but instead rather arbitrarily listed as being higher or lower than others.

It would be nice, if say there were 10 teams tied for first place, they were all listed as 1st, and the next group of teams would be tied for 11th.

It doesn't really make much of a difference other than bragging rights and a "feel good" feeling.

Anyway, that's my suggestion. Enjoying the site thanks for putting it together.

mjf

‹ Points behind the leader Cup Comp to keep the lower teams intrested ›

Comments

I'll look in to it

Submitted by ff1 on 19 April, 2009 - 20:42.

It's not really important to me, but you're not the first to ask, so I'll look in to it for you. I have a feeling it won't be a simple change because there will need to be a different condition after the last race when the tiebreakers are used. It will be a couple of weeks before I can look at it though. Hope that's ok?

Ian :o)

Possible way

Submitted by Ricardo on 1 May, 2009 - 16:33.

You shouldn't need a different condition for the last race.

I've had to do the reverse (create seperaters for tied teams that don't override the main scoring element) for a football prediction league so this might help.

I don't know how it works out rankings now, but if you could do it as something similar to this (and I don't mean it to sound patronising):

(points x 1,000,000) - 1/((time difference in seconds)^2)^1/2

If you then rank us by the result of that it should work.

Until the final race the second part is going to be the same for everybody as our predictions are all blank, so we'll be ranked solely by our points. Anyone with the same points will therefore be the same rank.

When it comes to the final race the squaring and then square rooting ensures that everybody's time difference is positive. By doing it as 1/ that it makes the smallest errors have the most 'points' with one one thousandth of a second giving 1,000 points. Thanks to the multiplier everybody on the same points will be grouped together but then sorted by that time gap.