Rugby World Cup 2019 predictions

[Groundhog]

View attachment 7284
Not OC but had to post on this thread

I think looking at the world rankings we should run a system whereby the team that finishes top of the Rugby Championship gets the chance to pay Italy and get promoted to the 6 nations. Yeah it might dilute the quality of the 6 nations for a few years, but it would do a lot to spread rugby to minnow nations like New Zealand who were thoroughly beaten by Australia, a country where rugby union is about as popular as yelling bomb at an airport.

 

[Brigantine]

I think looking at the world rankings we should run a system whereby the team that finishes top of the Rugby Championship gets the chance to pay Italy and get promoted to the 6 nations. Yeah it might dilute the quality of the 6 nations for a few years, but it would do a lot to spread rugby to minnow nations like New Zealand who were thoroughly beaten by Australia, a country where rugby union is about as popular as yelling bomb at an airport.

If New Zealand vs Australia is a repeat of Australia vs New Zealand last week, Wales vs England is also a repeat of England vs Wales, and South Africa beat Argentina again, the World Rugby Rankings will be:

1 England
2 Ireland
3 Wales
4 Australia
5 South Africa
6 New Zealand

I think we can just draw a line under that, ring fence it, and call it Six Nations. No need to play a relegation match!

 

[Leinster Fan]

If New Zealand vs Australia is a repeat of Australia vs New Zealand last week, Wales vs England is also a repeat of England vs Wales, and South Africa beat Argentina again, the World Rugby Rankings will be:

1 England
2 Ireland
3 Wales
4 Australia
5 South Africa
6 New Zealand

I think we can just draw a line under that, ring fence it, and call it Six Nations. No need to play a relegation match!

Honestly would prefer a six-way round robin between those teams to a world cup.

 

[Kiwi boss]

I think looking at the world rankings we should run a system whereby the team that finishes top of the Rugby Championship gets the chance to pay Italy and get promoted to the 6 nations. Yeah it might dilute the quality of the 6 nations for a few years, but it would do a lot to spread rugby to minnow nations like New Zealand who were thoroughly beaten by Australia, a country where rugby union is about as popular as yelling bomb at an airport.

This was funny considering you support Ireland and your last semi final was 28 years ago

Lets look at the last WC's -

SH ***les 7 - NH -1

Now lets go to semi finals of NH teams

2015 -0
2011 -2
2007 -1
2003 -2
1999 -1
1995 -2
1991 -2
1987 -2

World Rankings mean nothing

South Africa won the Champions Trophy and are in sixth

 

[Leonormous Boozer]

Is there a force field at the equator that blocks sarcasm or something?

 

[Groundhog]

This was funny considering you support Ireland and your last semi final was 28 years ago

Lets look at the last WC's -

SH ***les 7 - NH -1

Now lets go to semi finals of NH teams

2015 -0
2011 -2
2007 -1
2003 -2
1999 -1
1995 -2
1991 -2
1987 -2

World Rankings mean nothing

South Africa won the Champions Trophy and are in sixth

Have we a corner we can put this fella in to think about his mistakes.

 

[unrated]

Is there a force field at the equator that blocks sarcasm or something?

Blocks what?

 

[Kiwi boss]

Have we a corner we can put this fella in to think about his mistakes.

no mistakes

 

[Brigantine]

The possibility of a Japan - New Zealand QF got me thinking... Here's New Zealand's record in RWC games against host countries:

W 1991 England 12 - 18 New Zealand at Twickenham (Pool game)
W 1991 Scotland 6 - 13 New Zealand in Cardiff (Bronze Final, venue not in Scotland)
L 1995 South Africa 15 - 12 New Zealand (Final, a.e.t.)
L 2003 Australia 22 - 10 New Zealand (Semifinal)
L 2007 France 20 - 18 New Zealand in Cardiff (Quarterfinal, venue not in France)

We always seem to choke when we have to knock out a host country.

Outside of that we have only lost 3 other world cup games - once each to Australia (1991 SF), France (1999 SF) and South Africa (1999 BF). I'm getting deja vu...

 

[Umaga's Witness]

Honestly would prefer a six-way round robin between those teams to a world cup.

Just have three years to work on your world rankings, with a more even, randomised, distribution of who plays who during that time, and then the top 6 ranked teams can compete in the fourth year. With a final at the end.

 

[TheOvalBall]

I'd love it if Ireland won it. I'm 1/8 Irish.
And yes, I still love France.

 

[TheOvalBall]

Is it possible to have a poll ?

 

[bushytop]

I'd love it if Ireland won it. I'm 1/8 Irish.
And yes, I still love France.

You seem to have a few bases covered what with your flag too!

 

[TheOvalBall]

I cannot see South Africa winning it. The squad for the World Cup has not been announced yet, and the government requires them to have a lot of non-white players in the squad, which will leave out a lot of good players.

 

[Umaga's Witness]

Well the all blacks have been honing a dual playmaker game plan this year and last; they lost Damian McKenzie but decided to stick with the plan, shifting barret to fullback and starting mounga at ten. Now maybe mounga is out, meaning they will have to change their game plan because there are simply no playmakers left of international standard.

My biggest worry for the wc is that the team aren't smart enough to adapt on their own, that they need their coaches to tell them what to do. They lost last week and then their coaches told them what to change and then they won. You don't get a second chance in a World Cup

I cannot see South Africa winning it. The squad for the World Cup has not been announced yet, and the government requires them to have a lot of non-white players in the squad, which will leave out a lot of good players.

How many? Is this a new thing? Because they have had targets for a while and rassie hasn't done too bad a job of sticking to them. Unless there are new requirements from the government their squad will be similar to what has been playing this and last year. And they have been quite good since rassie started.

 

[BakkiesBokke]

https://www.sarugbymag.co.za/67018-2/

I really don't think transformation targets will affect the squad selection. Rassie has done well there too, although he is also fortunate that the form front row consists of three black players, which I didn't think would be the case a year ago.

 

[Brigantine]

Trying out a new format... Box & Whisker plots.
Is it readable like this?




Here's an alternative, interpolated (smoother) version. Does it make sense like that?



And now another version tinkering with the interpolation. It fixes some misleadingly long whiskers e.g. showing Italy as almost able to reach 2nd in pool.

 

[Umaga's Witness]

Trying out a new format... Box & Whisker plots.
Is it readable like this?

View attachment 7295


Here's an alternative, interpolated (smoother) version. Does it make sense like that?

View attachment 7296

And now another version tinkering with the interpolation. It fixes some misleadingly long whiskers e.g. showing Italy as almost able to reach 2nd in pool.

View attachment 7297

Very cool. Readable and makes sense, France for instance has a tight distribution around their mean but very long tails, which aligns with what you'd expect from them in that they're middle of the road but can beat anyone on their day or lose to anyone when it's really not their day.

I have no idea which version is better though because I'd simply want the one that was closest to the truth, and I don't understand the methodological differences underlying the different versions. I mean I get that the first one just uses midpoints of the places, Except where the separate components fall in the same placing, eg whisker and edge of box. And I think the other two are more intuitive to the eye, showing how close to the adjacent placing they are. But what is the difference between these two?

 

[Brigantine]

It isn't smart enough to factor in the France thing, it just shows France that way basically because the teams in their group are either way above France or way below France (usual caveats about volatility in RP, incl for Argentina atm) - like Italy's situation, but not so extreme. Also it's just a fluke that there is *not quite* a 25% chance of France beating England, Wales or Australia, and *not quite* a 25% chance of finishing below Argentina. (as you can see better in the other versions).

As for the difference between the 2nd and 3rd versions... In small font, because I'm talking about minor details in the fine print

First of all note that the end of the whiskers show where there is at least a 2.5% chance (2.5% above and below, so 95% in between. A common choice for confidence intervals / error bars)
I'll use Italy as an example how the 2nd version works... they have a 0.17% chance of being at least 2nd in pool - way below 2.5%, so basically the same as no chance. Then they have another 95% or so chance of being exactly 3rd in pool.
So where does 2.5% fit in between them? well in total within the "3rd in Pool" area, there's 2.33% above 2.5%, and about 93% below the whisker tip at 2.5%. 93% is way bigger than 2.33%, and in the interpolation, the "3rd in Pool" area on the graph is split in that ratio, so the whisker goes to only a tiny distance away from the border with "2nd in Pool".

I didn't like this, because having the whisker extend nearly into the "2nd in Pool" area seems like it suggests there's a nearly 2.5% chance of making 2nd in pool - which there isn't.

The 3rd version weights the 2.5% 9x more than the next 22.5%, so e.g. New Zealand or Wales have a 0% chance of doing better than Champions, and a ~25% chance of being Champions. This basically means the tip of the whisker is decided equally by the 0% as by the 25%. To reach close to the top or bottom of an area, there needs to actually be a large fraction of a 2.5% chance of a result in the next area. E.g. Canada has a 2.46% chance of making 3rd in pool, so they still get to have a whisker almost reaching the line.

Not so mathematically elegant, and maybe a bit contrived, but what you think you're seeing when you look at it is more like what I'm actually trying to show. (Or at least that's what *I* see when *I* look at it. Do you see it the same?)

 

[Umaga's Witness]

It isn't smart enough to factor in the France thing, it just shows France that way basically because the teams in their group are either way above France or way below France (usual caveats about volatility in RP, incl for Argentina atm) - like Italy's situation, but not so extreme. Also it's just a fluke that there is *not quite* a 25% chance of France beating England, Wales or Australia, and *not quite* a 25% chance of finishing below Argentina. (as you can see better in the other versions).

As for the difference between the 2nd and 3rd versions... In small font, because I'm talking about minor details in the fine print

First of all note that the end of the whiskers show where there is at least a 2.5% chance (2.5% above and below, so 95% in between. A common choice for confidence intervals / error bars)
I'll use Italy as an example how the 2nd version works... they have a 0.17% chance of being at least 2nd in pool - way below 2.5%, so basically the same as no chance. Then they have another 95% or so chance of being exactly 3rd in pool.
So where does 2.5% fit in between them? well in total within the "3rd in Pool" area, there's 2.33% above 2.5%, and about 93% below the whisker tip at 2.5%. 93% is way bigger than 2.33%, and in the interpolation, the "3rd in Pool" area on the graph is split in that ratio, so the whisker goes to only a tiny distance away from the border with "2nd in Pool".

I didn't like this, because having the whisker extend nearly into the "2nd in Pool" area seems like it suggests there's a nearly 2.5% chance of making 2nd in pool - which there isn't.

The 3rd version weights the 2.5% 9x more than the next 22.5%, so e.g. New Zealand or Wales have a 0% chance of doing better than Champions, and a ~25% chance of being Champions. This basically means the tip of the whisker is decided equally by the 0% as by the 25%. To reach close to the top or bottom of an area, there needs to actually be a large fraction of a 2.5% chance of a result in the next area. E.g. Canada has a 2.46% chance of making 3rd in pool, so they still get to have a whisker almost reaching the line.

Not so mathematically elegant, and maybe a bit contrived, but what you think you're seeing when you look at it is more like what I'm actually trying to show. (Or at least that's what *I* see when *I* look at it. Do you see it the same?)

Yeah, I think that's what I see, the third is best then. If you were ok with not having areas, just having a point on a spectrum as your placing, starting with last in pool, ending with champions, then you could get something even more intuitive I think. Maybe? It might not look as good though.