Stats - Freo v Dogs

[jnrmac 20,189]

INteresting stats 22m 3rd quarter

Dogs Freo

k 136 130

h 92 82

Disp 233 217

Clearances 28 20

i50 26 36

Cont Poss 92 91

Tackles 44 37

HO 31 to 23

Dogs leading everything except i50 and yet they trail by 38pts. If that demonstrate the reliability of stats nothing will...

[autocol 448]

If you're unable to understand that statistical trends predict probabilities (not certainties), and that a sample size of one is statistically insignificant... then yes, I recommend you stay away from statistics.

[CHAMP 347]

INteresting stats 22m 3rd quarter

Dogs Freo

k 136 130

h 92 82

Disp 233 217

Clearances 28 20

i50 26 36

Cont Poss 92 91

Tackles 44 37

HO 31 to 23

Dogs leading everything except i50 and yet they trail by 38pts. If that demonstrate the reliability of stats nothing will...

Freo cleaner and classier. Dogs getting nothing from either key forward and opportunities too often coming to nothing.

[autocol 448]

That was a rude response. Let me put it another way.

Let's say (I'm making up the numbers for this hypothetical) that in 75% of games where Team A wins the tackle count by 20 or more, they win the game. This stat would be taken from the last 1000 games of footy (a statistically significant sample).

What you could then say, with great confidence, is that if you win the tackle count by 20 or more, you're much more likely to win the game. What it DOESN'T mean is that, if you find a game where Team A wins the tackle count by 25 but loses the game, you can say "statistics are crap". That's completely wrong. All it means is that your sample of one was the more improbable result.

Whilst I don't know the numbers, usually you would think that a team winning clearances by 8 and tackles by 7 would be winning the game maybe 60-70% of the time, but then again a team leading I50's by 10 would usually be winning about 70-80% of the time.

In short... I don't see anything in these stats that suggests that statistical analysis is an improper method of predicting the result of a game.

[daisycutter 29,743]

That was a rude response. Let me put it another way.

Let's say (I'm making up the numbers for this hypothetical) that in 75% of games where Team A wins the tackle count by 20 or more, they win the game. This stat would be taken from the last 1000 games of footy (a statistically significant sample).

What you could then say, with great confidence, is that if you win the tackle count by 20 or more, you're much more likely to win the game. What it DOESN'T mean is that, if you find a game where Team A wins the tackle count by 25 but loses the game, you can say "statistics are crap". That's completely wrong. All it means is that your sample of one was the more improbable result.

Whilst I don't know the numbers, usually you would think that a team winning clearances by 8 and tackles by 7 would be winning the game maybe 60-70% of the time, but then again a team leading I50's by 10 would usually be winning about 70-80% of the time.

In short... I don't see anything in these stats that suggests that statistical analysis is an improper method of predicting the result of a game.

you can't beat the best stat of all - the scoreboard 100% predictability

[Chook 15,022]

I notice that disposal efficiency wasn't one of those stats. From what I've seen today (which admittedly hasn't been much because I've been mowing the lawn) the Dockers' pressure has mucked up a lot of the Bulldogs' kicks and stopped them from scoring.

[Demon Jack 6,930]

Players responsible for umpires being taken off the field on a stretcher

Bulldogs

0

Freo

1

Edited April 6, 2013 by Demon Jack

[P-man 13,366]

Pre-season predictions were that forward potency would again be an issue for the Dogs this year and despite their strong start I still think that will be the case.

With the re-emergence of Cooney, they actually have one of the better midfields going around, hence their dominance in those key stats. It is at the opposite ends of the ground where they will struggle again.

[Chook 15,022]

Players responsible for umpires being taken off the field on a stretcher

Bulldogs

0

Freo

1

I think there are a few fans who believe that should be the sole determinant of who wins the game.

[Muvver Jones 64]

The Dockers under Lyon are one of the most efficient teams in the comp. Supremely organised and adaptable. They make a mockery of stat counting.They will be grand finalists IMO. A far cry from the distracted and effette team 2 years ago who were one of the few sides we used to beat on a regular basis BTW.

Edited April 6, 2013 by Muvver Jones

[Redleg 41,492]

If you're unable to understand that statistical trends predict probabilities (not certainties), and that a sample size of one is statistically insignificant... then yes, I recommend you stay away from statistics.

Statistically speaking.

[jnrmac 20,189]

That was a rude response. Let me put it another way.

Let's say (I'm making up the numbers for this hypothetical) that in 75% of games where Team A wins the tackle count by 20 or more, they win the game. This stat would be taken from the last 1000 games of footy (a statistically significant sample).

What you could then say, with great confidence, is that if you win the tackle count by 20 or more, you're much more likely to win the game. What it DOESN'T mean is that, if you find a game where Team A wins the tackle count by 25 but loses the game, you can say "statistics are crap". That's completely wrong. All it means is that your sample of one was the more improbable result.

Whilst I don't know the numbers, usually you would think that a team winning clearances by 8 and tackles by 7 would be winning the game maybe 60-70% of the time, but then again a team leading I50's by 10 would usually be winning about 70-80% of the time.

In short... I don't see anything in these stats that suggests that statistical analysis is an improper method of predicting the result of a game.

I accept your apology of sorts.

And yes I know that 7 out of 5 people don't know a thing about statistics.