Cats and the Bye - Some Numbers

[CatToTheFuture]

Look at Port's record after the bye. And people are worried they'll smack us lol

Port isn't coming of the bye, they're coming of a loss away and bring back 3 good players

 

[winty]

Port isn't coming of the bye, they're coming of a loss away and bring back 3 good players

Well, two good players and Dixon.

 

[Landgraft]

Correct me if I'm wrong here but Chris Scott has coached 202 games at a 70% win rate. If we had only ever seen the 11 post bye games and we estimated his performance as a coach based on his 18% win rate there then a statistician would be 90% confident that his win loss for the full 202 games would fall within 1% and 43% which doesn't match reality.

The kicker here is that we've selected non-randomly. We've introduced selection bias. If we were going to be 90% certain that the results of this selection were random chance then we'd expect a minimum 5 wins from an 11 game sample. A statistician would I think be 90% certain we're selecting from a different subset when looking at post bye games.

Fell free to check my maths, it's been a long time since I studied confidence intervals.

Edit: would like to add that I'm confident we win this match. As much as the post bye hoodoo is a different subset of Scott's coaching career, so is the 2019 cats side. Rules are out the window this year I think

You would be very poorly placed to put any confidence in extrapolating from a very small and cherry picked data set like that.

See also: finals w-l, except that has even more flaws all of its own

 

[CatToTheFuture]

You would be very poorly placed to put any confidence in extrapolating from a very small and cherry picked data set like that.

See also: finals w-l, except that has even more flaws all of its own

That's why the error margin is 25% which adjusts for the 90% confidence interval. You couldn't do an estimation with a 5% error margin but if you're willing to set the brackets wide enough it technically fits the maths.

 

[Landgraft]

That's why the error margin is 25% which adjusts for the 90% confidence interval. You couldn't do an estimation with a 5% error margin but if you're willing to set the brackets wide enough it technically fits the maths.

Only for a random sampling, not for a predetermined one.

 

[CatToTheFuture]

Only for a random sampling, not for a predetermined one.

Definitely. The average not falling within the error bounds is what indicates that the trend is non-random if I'm interpreting that right. The implication of the comment itself is that our performance after the bye only has a 10% chance of being a fluke right?

 

[King Cold]

Good post we had an average side in them years with crap bottom 6 players

Jordan Murdoch played 100 games over that time period, enough said.

 

[BotsMaster]

Jordan Murdoch enough said.

EFA

 

[catstorm]

The club is clearly clueless why this is happening because despite it going on for years absolutely nothing changes post bye.

 

[foxdog50]

 

[UnitedWarrior93]

7 years on, a whole list turnover and still we are no closer to getting that post bye win and are as clueless as we’ve ever been.

I really have no answers, club obviously the same, it’s just incredible.

 

[Herne Hill Hammer]

That's 2 / 11 now .

 

[fpm84]

I’m undecided, I don’t think ‘the bye’ has been a factor in most of these losses.

I can only think of a couple, most notably the 2016 prelim, where we’ve come out flat as a tack and the game has been over by quarter time.

‘The bye’ didn’t make Taylor miss after the siren v Bulldogs last year after a close match that could have gone either way.

I don’t think we were flat tonight FWIW, Port brought four quarter intensity and it was a scrap for the most part.

 

[lionbear]

Look at our draw off the bye since 2016.

It’s a myth getting fueled by the AFL fixture department.

 

[Stocks_go_Up]

Chris Scott will brush it off like all the other times.

 

[Herne Hill Hammer]

Since the SA sides started playing at the AO, Geelong before tonight were leading Port in the head to head there 3 /1, the only loss coming in the two clubs' very first meeting there back in 2014. In that time, the closest that Port had got to Geelong was 2 pts in their only game in that 5 period that was played at Kardinia.

 

[fpm84]

Look at our draw off the bye since 2016.

It’s a myth getting fueled by the AFL fixture department.

Tend to agree.

Give us a bottom 8 side at GHMBA after the bye and then we’ll see if it’s really an issue.

 

[Micksy]

 

[Biggy_Boy]

Tend to agree.

Give us a bottom 8 side at GHMBA after the bye and then we’ll see if it’s really an issue.

And give Scott more bottom 8 sides at GHMBA during finals, and I'm sure he'd rectify his finals record.

 

[Mad_Hatter]

Clearly it's a massive issue, but you can look at WHO and WHERE we have played as a genuine reason as to why we lose so many of them.
However, come finals you play good teams every game and if we are to win the flag then we must navigate effectively 2 byes.
I have no confidence the club will be able to do that.

 

[Herne Hill Hammer]

It's all well and good for people to say that we've played good sides when coming off a bye but we have been one of the best h&a sides during that time.

At the end of the H&A :

2012 - 6th, 2 wins from top spot.
2013 - 2nd, 1 win from top spot.
2014 - 3rd, level on points with Sydney and Hawthorn.
2015 - 10th, a game and a half from 8th and 5 and a half from top.
2016 - 2nd, level on points with Sydney on top.
2017 - 2nd, level on points with Adelaide.
2018 - 8th, 5 games off top spot and a game clear of North in 9th. Had the 4th best attack and the best defence and 2nd best percentage.
2019 - 1st, 11-1, two games and almost 20% clear on top of the ladder.

Winning h&a percentage from 2012 up until this week is 71.25%. Take out those 8 byes and the losses and the winning percentage sits at 80%.

 

[xtatik222k]

I have a bit of a theory about our lacklustre performances in the last decade. I don't have much knowledge about the inner workings of the club, but this might be food for thought.
I've heard a lot of people in the media over the years talk about the management of internal process at the club and people say Geelong is one of the most tightly-run and well-oiled clubs in the competition. There's strict process and adherence to procedure that many people have made special mention of.
I wonder whether this procedural methodology becomes something the players rely on, such that breaking that pattern has a negative impact on the play of the team as a whole. I wonder whether the processes are so stringent and micro-managed that maybe straying from that path mid-season causes things to break down a little and causes the team to lose form somewhat?
Essentially, the routine is a defining factor at the club, moreso than others, which negatively impacts our performances after that routine has been disrupted?
Other than that, I can't figure out why it happens to us more than any other club, and it seems to be completely unrelated to how we're going generally for the rest of the season...

 

[catempire]

 

[Vdubs]

Well, two good players and Dixon.

Dixon is possibly the MVP of those 3

 

[Vdubs]

7 years on, a whole list turnover and still we are no closer to getting that post bye win and are as clueless as we’ve ever been.

I really have no answers, club obviously the same, it’s just incredible.

But nobody has an answer. It doesn't actually matter at all. We are going to lose a game. I would have selected Port even if no bye. (I'm not selecting Adelaide this week.) We were due a loss, and we faced a team that was better than us all over the ground. No excuses, just better. Port were ready for a win.
Does it mean that because we smashed WCE many weeks ago, they are no threat?
Does it mean that because GWS beat us at home, we have no hope against them in a final?
The prefinals bye is for every team, and I think it's irrelevant. We actually beat Hawthorn