Percentage

[1eyedog]

Re: Percentage

That's weird as we have had a win of 60 points and one of 90 odd. The Swans always seem to have a healthy average come season's end (as stated above) they don't kick big scores but they have bugger all kicked against them. I would love to beat them this week.

[Sockeye Salmon]

Re: Percentage

Starting this week.

If the Swans were to beat us this week 90 points to 60, our % will be 129% and theirs will be 128%.

If the Swans were to beat us this week 110 points to 80, our % will be 128% and theirs will be 126%.

But it works the other way, too.

If we beat them by 90 to 60, our % will be 140% and theirs will be 114%.

[ledge]

Re: Percentage

If we beat Sydney and North in the next 2 games percentage means jack, rather we win then percentage is irrelevant.
Looking at ladder now 3 teams are near enough 2 games in front already , we win next 2 and Adelaide might be the only team that worries us rest of the year.

[Dancin' Douggy]

Re: Percentage

Ahem,,,,, There may be more draws this year.
Percentage is always important. One percentage point could be the difference between a home or away final.

[Topdog]

Re: Percentage

If we beat Sydney and North in the next 2 games percentage means jack, rather we win then percentage is irrelevant.
Looking at ladder now 3 teams are near enough 2 games in front already , we win next 2 and Adelaide might be the only team that worries us rest of the year.

Very simplistic way of looking at things IMO. What if we beat Sydney and North but only win 13.5 games for the year. Then percentage MAY come into things.

[Sedat]

Re: Percentage

If we beat Sydney and North in the next 2 games percentage means jack, rather we win then percentage is irrelevant.
Looking at ladder now 3 teams are near enough 2 games in front already , we win next 2 and Adelaide might be the only team that worries us rest of the year.

It's better to put forward a hypothetical to demonstrate how percentage works and how it greatly disadvantages higher scoring teams:

To date in season 2008, Sydney are scoring 87 points and conceding 70 points a game. If they continued at this rate for the rest of the season, they would end up scoring 1914 points and conceding 1540 points, which a differential of +374 points. Their percentage would end up being 124.3.

The Bulldogs are currently averaging a staggering 134 points a game and conceding 96 - we are on track to almost break the 3,000 points for barrier for the season. Say this dropped back to a more realistic 113 points a game (giving us a season total of 2,486 points for), and we continued to concede 96 points a game (2112 points against for the season) - giving us exactly the same differential as the Swans of +374 points - our percentage would be 117.7, which is almost 7 percentage points less than Sydney's. Theoretically, Sydney could actually score 100 less points (1,814 for the season), concede the same points, and their percentage would be 117.8, still higher than ours, despite their differential between points for and against now being only +274 points, 100 less than ours. Very unfair on the higher scorring teams.

So if Sydney can make up the two games they are currently behind the Dogs and continue to win low scoring slugfests, their percentage has a greater capacity to rise at a faster rate that ours, if we continue to play free-wheeling high-scoring football.

[The Coon Dog]

Re: Percentage

^^^^^

Thanks for the analysis Sedat.

One question. I believe they calculate % different in the SANFL. Do you know how they calculate it.

It's a % of the total points scored.

If you won a game & the scores were 122 to 78 the you have scored 61% of the total points scored so your % would be 61, assuming it's round 1.

The following week, you score 94 & the other team scores 101 then your % is 54.68 (as you have scored 216 of the 395 points kicked).

[Sockeye Salmon]

Re: Percentage

^^^^^

Thanks for the analysis Sedat.

One question. I believe they calculate % different in the SANFL. Do you know how they calculate it.

We've been down this road before. They calculate it slightly differently but it doesn't make any difference in the end.

Instead of 'points for/points against x 100' they use 'points for/total points x 100'.

[Sedat]

Re: Percentage

^^^^^

Thanks for the analysis Sedat.

One question. I believe they calculate % different in the SANFL. Do you know how they calculate it.

No probs Aker39.

I believe the SANFL multiply their points for by 50 and the divide that number by points against, hence their mid-point is 50%. We mulitply our points for by 100 and the divide that number by the points against, so our mid-point is 100%. They also give their teams 2 points for a win and one point for a draw. EDIT: I clearly have no idea what I'm talking about!!

I've never understood why we don't just have points difference as the measuring stick when teams are locked on the same number of wins and losses. In soccer, goal difference works simply and easily, and it would be just as simple and effective in AFL - it would also take away the obvious advantage that lower scoring teams currently enjoy with the percentage model.

[Go_Dogs]

Re: Percentage

It's a % of the total points scored.

If you won a game & the scores were 122 to 78 the you have scored 61% of the total points scored so your % would be 61, assuming it's round 1.

The following week, you score 94 & the other team scores 101 then your % is 54.68 (as you have scored 216 of the 395 points kicked).

Better way of doing it imo.

[Sockeye Salmon]

Re: Percentage

Thanks TCD. I couldn't resist, so I've compared the top 8 teams using both calculations.

Geelong 146.17 59.37
Hawthorn 139.42 58.23
Bulldogs 139.10 58.17
Adelaide 115.53 53.60
Sydney 124.35 55.42
North 104.48 51.09
Pies 119.72 54.48
Brisbane 104.89 51.19

What this shows is that it makes no difference either way.

[Sedat]

Re: Percentage

Under a points for model, the ladder would look like this (apologies for formatting):

W L D F A PD
Geelong 6 0 0 706 483 +223
Hawthorn 6 0 0 718 515 +203
Bulldogs 5 0 1 804 578 +226
Adelaide 4 2 0 610 528 +82
Sydney 3 2 1 526 423 +103
North Melb 3 2 1 583 558 +25
Collingwood 3 3 0 674 563 +111
Brisbane 3 3 0 622 593 +29

Interesting that we actually have the best differential between points for and against in the competition but not the best percentage. Both Hawthorn and Geelong have a better percentage than us but we are actually 23 and 20 points respectively ahead of them in the points for and against differential. Similarly Collingwood, who are also a high scoring team this year, are 5% behind Sydney but actually ahead of them in the differential between points for and against.

[ledge]

Re: Percentage

I say go for points differential, it keeps teams going for 4 quarters because they get a better look at how its panning out and how far ahead or away they are with a quick glance.
Also encourages more attacking football, maybe cutting back on this tempo footy everyone seems to have grabbed onto.
The way it is now its better to not score much but win. Totally wierd set up.

[Sedat]

Re: Percentage

Massively important win today to keep 3 steps ahead of Sydney - their percentage has now jumped to 130.0, which is only 1.6% less than ours. Their low scoring routs have dramatically increased their percentage in a very short space of time.

[1eyedog]

Re: Percentage

Massively important win today to keep 3 steps ahead of Sydney - their percentage has now jumped to 130.0, which is only 1.6% less than ours. Their low scoring routs have dramatically increased their percentage in a very short space of time.

Yep we cannot possibly go down the ladder if we keep winning, regardless of our percentage.