Jump to content

The Road to Qatar.


Binky

Recommended Posts

11 minutes ago, Stryker911 said:

I was curious what people's thoughts were on who do you cheer for in the other games? Do you want US and Mexico to win all their games? Do you cheer against Honduras regardless of opponent?

Won’t start worrying about other games until the new year.  But on your last question, yes!

Link to comment
Share on other sites

57 minutes ago, Stryker911 said:

I was curious what people's thoughts were on who do you cheer for in the other games? Do you want US and Mexico to win all their games? Do you cheer against Honduras regardless of opponent?

Yes, I do think this would be beneficial.  Hopefully we can pick up points against both at home. I doubt the US will have many screw ups, even away, this campaign… I pick them to win the group… and I’d say “comfortably” if Berhalter wasn’t the coach.

Link to comment
Share on other sites

38 minutes ago, youllneverwalkalone said:

Hmm, every fibre of my being says "Yes, I do cheer against Honduras no matter what." But then, I really think that I want them to do well enough for Canada to be the last thing they see before missing the World Cup.

I just love revenge. Carolina Hurricanes are my new favourite NHL team, strictly on that basis.

I get the poetry of that scenario, but I don’t want to have to go down to Honduras with both teams needing a result.  New team or not, that is a demon I don’t need to revisit.    

Link to comment
Share on other sites

16 minutes ago, dyslexic nam said:

I get the poetry of that scenario, but I don’t want to have to go down to Honduras with both teams needing a result.  New team or not, that is a demon I don’t need to revisit.    

"Revenge is a dish best served cold"

Link to comment
Share on other sites

My default will be to hope USA and Mexico win all their games not against us. But I do like to see USA lose so I won't be too upset if they do lose a game, at least this early on when we don't have much of the picture established yet. The real scoreboard watching won't start until at least November I would think.

Link to comment
Share on other sites

Alright, so I couldn’t help it but try out this simulation concept. Warning: if you hate analytics and simulations, please skip.

I developed a MATLAB script that would simulate the Ocho 1,000,000 times, then analyzed how many points it would take to get 3rd or 4th. There are a few inputs that affect the simulation: 1) strength of home field advantage, 2) how often a draw takes place between two teams, and 3) the relative strengths of each of the teams. If, for example, you made all 8 teams equal in strength then draws would occur quite often, and teams would combine for 2 pts (1+1) as opposed to 3 pts (3+0) for any given game. And so the spread from top to bottom would be much more clustered together. Anyway, I tweaked the 3 input parameters to best match KJ’s historical analysis. This means that the simulation is not going to tell you anything different from the ‘average’ pts needed, but rather will give you an indication of the variance.

Results:

3rd place - 22.6 +/- 3.8 pts (2 sigma / 95% of simulations)

4th place - 20.3 +/- 3.6 pts (2 sigma / 95% of simulations)

What this means is that if we get 26.4 pts, we will get at least 3rd 95% of the time. 23.9 pts gets you at least 4th 95% of the time. On the low end, 16.7 pts is the bare minimum to get 4th place, and 18.8 pts the bare min for 3rd (again, 95% of the time).

Bar plots with 2-sigma confidence intervals:

image.png.0e5fe0f511732df4d2a6803a721646a1.png

Histogram for 3rd and 4th

image.png.99141f40b4d3e2b97ec38f82c2dd8754.png

Takeaways: Win 5 non-MEX/USA home games, and 3 away games and we’re at 24 pts and basically guaranteed we’re still in it. And quite likely direct qualification. If we play for draws, especially at home, we’re playing with fire.

 

If anyone wants to see the source code to play around with it yourself, PM me. But beware, I’m a hardware engineer by training and my code is not well written/efficient. My software engineer wife was shaking her head at my code the whole time.

Link to comment
Share on other sites

3 hours ago, maplebanana said:

Alright, so I couldn’t help it but try out this simulation concept. Warning: if you hate analytics and simulations, please skip.

I developed a MATLAB script that would simulate the Ocho 1,000,000 times, then analyzed how many points it would take to get 3rd or 4th. There are a few inputs that affect the simulation: 1) strength of home field advantage, 2) how often a draw takes place between two teams, and 3) the relative strengths of each of the teams. If, for example, you made all 8 teams equal in strength then draws would occur quite often, and teams would combine for 2 pts (1+1) as opposed to 3 pts (3+0) for any given game. And so the spread from top to bottom would be much more clustered together. Anyway, I tweaked the 3 input parameters to best match KJ’s historical analysis. This means that the simulation is not going to tell you anything different from the ‘average’ pts needed, but rather will give you an indication of the variance.

Results:

3rd place - 22.6 +/- 3.8 pts (2 sigma / 95% of simulations)

4th place - 20.3 +/- 3.6 pts (2 sigma / 95% of simulations)

What this means is that if we get 26.4 pts, we will get at least 3rd 95% of the time. 23.9 pts gets you at least 4th 95% of the time. On the low end, 16.7 pts is the bare minimum to get 4th place, and 18.8 pts the bare min for 3rd (again, 95% of the time).

Bar plots with 2-sigma confidence intervals:

image.png.0e5fe0f511732df4d2a6803a721646a1.png

Histogram for 3rd and 4th

image.png.99141f40b4d3e2b97ec38f82c2dd8754.png

Takeaways: Win 5 non-MEX/USA home games, and 3 away games and we’re at 24 pts and basically guaranteed we’re still in it. And quite likely direct qualification. If we play for draws, especially at home, we’re playing with fire.

 

If anyone wants to see the source code to play around with it yourself, PM me. But beware, I’m a hardware engineer by training and my code is not well written/efficient. My software engineer wife was shaking her head at my code the whole time.

Dude! Awesome analysis! Thanks for posting!

Link to comment
Share on other sites

4 hours ago, maplebanana said:

Alright, so I couldn’t help it but try out this simulation concept. Warning: if you hate analytics and simulations, please skip.

I developed a MATLAB script that would simulate the Ocho 1,000,000 times, then analyzed how many points it would take to get 3rd or 4th. There are a few inputs that affect the simulation: 1) strength of home field advantage, 2) how often a draw takes place between two teams, and 3) the relative strengths of each of the teams. If, for example, you made all 8 teams equal in strength then draws would occur quite often, and teams would combine for 2 pts (1+1) as opposed to 3 pts (3+0) for any given game. And so the spread from top to bottom would be much more clustered together. Anyway, I tweaked the 3 input parameters to best match KJ’s historical analysis. This means that the simulation is not going to tell you anything different from the ‘average’ pts needed, but rather will give you an indication of the variance.

Results:

3rd place - 22.6 +/- 3.8 pts (2 sigma / 95% of simulations)

4th place - 20.3 +/- 3.6 pts (2 sigma / 95% of simulations)

What this means is that if we get 26.4 pts, we will get at least 3rd 95% of the time. 23.9 pts gets you at least 4th 95% of the time. On the low end, 16.7 pts is the bare minimum to get 4th place, and 18.8 pts the bare min for 3rd (again, 95% of the time).

Bar plots with 2-sigma confidence intervals:

image.png.0e5fe0f511732df4d2a6803a721646a1.png

Histogram for 3rd and 4th

image.png.99141f40b4d3e2b97ec38f82c2dd8754.png

Takeaways: Win 5 non-MEX/USA home games, and 3 away games and we’re at 24 pts and basically guaranteed we’re still in it. And quite likely direct qualification. If we play for draws, especially at home, we’re playing with fire.

 

If anyone wants to see the source code to play around with it yourself, PM me. But beware, I’m a hardware engineer by training and my code is not well written/efficient. My software engineer wife was shaking her head at my code the whole time.

So basically you are saying we need to start with two wins at home to Honduras and El Salvador, otherwise we need to beat one of Mexico or US at home or take a lot of points on the road which is unlikely

Link to comment
Share on other sites

Simulated the ocho in FM '21 (didn't control anyone just let it run as usual).

Window 1

  • CAN 2-0 HON (Larin, Davies)
  • USA 2-1 CAN (Eustaquio)
  • CAN 1-0 ESV (Buchanan)

Window 2

  • MEX 3-0 CAN
  • JAM 1-0 CAN
  • CAN 2-1 PAN (Arfield, Buchanan)

Window 3

  • CAN 1-0 CRC (David)
  • CAN 4-1 MEX (David x3, Larin)

Window 4

  • HON 1-1 CAN (David)
  • CAN 1-1 USA (Cavallini)
  • ESV 0-1 CAN (David)

Window 5

  • CRC 0-2 CAN (Cavallini x2)
  • CAN 2-0 JAM (Davies, Buchanan)
  • PAN 3-3 CAN (David x2, Osorio)

Total: 8-3-3.  Needless to say if we go undefeated in Central America we will be in very good shape.  The interesting part was that FM absolutely loves Jamaica with Antonio/Pinnock et al to the point where it actually took until the second last matchday to qualify.

0?ui=2&ik=593450151b&attid=0.1&permmsgid

Edited by theaub
Link to comment
Share on other sites

4 minutes ago, theaub said:

Simulated the ocho in FM '21 (didn't control anyone just let it run as usual).

Window 1

  • CAN 2-0 HON (Larin, Davies)
  • USA 2-1 CAN (Eustaquio)
  • CAN 1-0 ESV (Buchanan)

Window 2

  • MEX 3-0 CAN
  • JAM 1-0 CAN
  • CAN 2-1 PAN (Arfield, Buchanan)

Window 3

  • CAN 1-0 CRC (David)
  • CAN 4-1 MEX (David x3, Larin)

Window 4

  • HON 1-1 CAN (David)
  • CAN 1-1 USA (Cavallini)
  • ESV 0-1 CAN (David)

Window 5

  • CRC 0-2 CAN (Cavallini x2)
  • CAN 2-0 JAM (Davies, Buchanan)
  • PAN 3-3 CAN (David x2, Osorio)

Total: 8-3-3.  Needless to say if we go undefeated in Central America we will be in very good shape.  The interesting part was that FM absolutely loves Jamaica with Antonio/Pinnock et al to the point where it actually took until the second last matchday to qualify.

0?ui=2&ik=593450151b&attid=0.1&permmsgid

Love that a FM simulation has us trashing Mexico at home. Of course anyone who has played FM knows that in 4 years Mexico will have about 10 world class regens. 

Link to comment
Share on other sites

2 hours ago, Jedi Ram said:

Here is link based on ELO and the future probablity of qualifying ... it gives up 52% chance of qualifying directly.

http://www.football-rankings.info/2021/08/concacaf-qualification-world-cup-2022.html

Thanks for that link.

So a 71.5% chance of a top 4 finish. Seems about right.

Link to comment
Share on other sites

COVID in Jamaica is really getting out of control and the vaccination numbers are very low. The UK Foreign Office is now advising against travel to Jamaica and it seems very possible that it will be "red listed" in two weeks.

We play at Jamaica in the next window - I wonder if Jamaica will try to schedule this at a neutral location (USA?) or would rather play us at home with a severely depleted squad. Anyways, we'll see in a few weeks but this will be an interesting development to watch. 

Link to comment
Share on other sites

13 minutes ago, Floortom said:

We play at Jamaica in the next window - I wonder if Jamaica will try to schedule this at a neutral location (USA?)

It's not that absurd given that they will play the USA in Austin right before our match against them.

In terms of costs and logistics it might be easier for them to take a coach bus for 3 hours and host us in Houston before flying to Honduras.

Link to comment
Share on other sites

39 minutes ago, Floortom said:

COVID in Jamaica is really getting out of control and the vaccination numbers are very low. The UK Foreign Office is now advising against travel to Jamaica and it seems very possible that it will be "red listed" in two weeks.

We play at Jamaica in the next window - I wonder if Jamaica will try to schedule this at a neutral location (USA?) or would rather play us at home with a severely depleted squad. Anyways, we'll see in a few weeks but this will be an interesting development to watch. 

We can always offer them one of our venues in Canada.😏

Link to comment
Share on other sites

29 minutes ago, Olympique_de_Marseille said:

It's not that absurd given that they will play the USA in Austin right before our match against them.

In terms of costs and logistics it might be easier for them to take a coach bus for 3 hours and host us in Houston before flying to Honduras.

Best bet is at DRV Pink Stadium in Fort Lauderdale or at RedBull arena in NJ if they want some kind of home support attendance.

Link to comment
Share on other sites

1 hour ago, BigMo said:

This was great. Impressed by the Jamaican supporters knowledge of our NT. He had very high praise for us.

He's pretty knowledgeable overall, but yes we seem to have earned his respect after the Suriname game. He did a lot of analysis in the lead up to it, particularly on the Surinamese side. Because he’s based in Netherlands/Belgium, he had the connections and the incentive to provide coverage on them, and became quite acquainted with us by extension.

Link to comment
Share on other sites

  • Admin locked this topic
Guest
This topic is now closed to further replies.

×
×
  • Create New...