CONCACAF Nations League 2024-25 - Group Stage & ¼finals - General Thread

[Olympique_de_Marseille]

14 minutes ago, Neil R. said:

If we end up overtaking the US for 2nd in the CONCACAF rankings after the October update

Someone like @Kent or @maplebanana would need to do the calculations.

But I don't think it's possible.

https://www.concacaf.com/rankings/men-s-national-team/

Right now that is:

Mexico 1882

USA 1805

Canada 1769

Panama 1756

So with the win against the USA and draw against Mexico, and supposing a win against Panama.

And with the USA losing to us + their draw against New Zealand, and supposing they lose to Panama and to Mexico.

Would the swing be enough?

 

 

[Kent]

3 hours ago, Olympique_de_Marseille said:

Someone like @Kent or @maplebanana would need to do the calculations.

But I don't think it's possible.

https://www.concacaf.com/rankings/men-s-national-team/

Right now that is:

Mexico 1882

USA 1805

Canada 1769

Panama 1756

So with the win against the USA and draw against Mexico, and supposing a win against Panama.

And with the USA losing to us + their draw against New Zealand, and supposing they lose to Panama and to Mexico.

Would the swing be enough?

 

 

I don't think any of us have figured out the CONCACAF ranking formula completely (I'm trying to remember if maplebanana did come up with a table to approximate it, or if that was related to FIFA rankings), but @shorty had a good break down that sounded very plausible to me, I guess in a different thread (maybe the FIFA ranking thread). With his estimations/assumptions/logic, it had us passing USA at this point already. The most recent rankings had El Salvador and Guatemala gaining/losing 22 points based on one friendly between the two teams. So that was a 44 point swing in just one friendly. So it is very possible that we could have made up the 36 point gap already.

[Obinna]

1 hour ago, Kent said:

I don't think any of us have figured out the CONCACAF ranking formula completely (I'm trying to remember if maplebanana did come up with a table to approximate it, or if that was related to FIFA rankings), but @shorty had a good break down that sounded very plausible to me, I guess in a different thread (maybe the FIFA ranking thread). With his estimations/assumptions/logic, it had us passing USA at this point already. The most recent rankings had El Salvador and Guatemala gaining/losing 22 points based on one friendly between the two teams. So that was a 44 point swing in just one friendly. So it is very possible that we could have made up the 36 point gap already.

Here is his post from the other thread you are referring to:

Formula for Concacaf is Pts = Importance * (Result - Expected Result). Result is either 1 for a win, 0 for a draw, or -1 for a loss). Friendlies are worth an “importance” of 17.5

So, even without knowing the expected result specifically, we know our win is probably worth a swing of at least 35 pts — we gain at least 17.5 and they lose at least 17.5, assuming the expected result was that the Americans would win or at best a draw.  
Then even without considering the Mexico draw, the Americans probably drop even more points for drawing NZ at home when they were probably “expected” to win. 

[Unnamed Trialist]

5 hours ago, Obinna said:

Here is his post from the other thread you are referring to:

Formula for Concacaf is Pts = Importance * (Result - Expected Result). Result is either 1 for a win, 0 for a draw, or -1 for a loss). Friendlies are worth an “importance” of 17.5

So, even without knowing the expected result specifically, we know our win is probably worth a swing of at least 35 pts — we gain at least 17.5 and they lose at least 17.5, assuming the expected result was that the Americans would win or at best a draw.  
Then even without considering the Mexico draw, the Americans probably drop even more points for drawing NZ at home when they were probably “expected” to win. 

If it is that "simple", our friendly win essentially ties us with them, or puts us a point behind them, and as you say, our Mexico draw vs their draw would put us ahead of them. 

What is this important for, the Nations League draw?

[Obinna]

1 hour ago, Unnamed Trialist said:

If it is that "simple", our friendly win essentially ties us with them, or puts us a point behind them, and as you say, our Mexico draw vs their draw would put us ahead of them. 

What is this important for, the Nations League draw?

That is correct. We should be ahead of them now.

The importance is avoiding Costa Rica or Jamaica in the quarter final, assuming both win their groups. Seeds 3 and 4 play the group winners, while seeds 1 and 2 play the runner ups, and there seems to be a pretty big drop off after Jamaica and Costa Rica. Honduras is probably the most dangerous candidate for a runner up in group B, but they aren't what they were previously, while maybe Guatemala or Suriname end up as runner up behind Costa Rica, neither of which should concern us.

Assuming we are 2nd seed now, the QF match ups would be as follows if the current group standings hold:

Mexico (1) vs Nicaragua (group B runner up)

Canada (2) vs Guatemala (group A runner up)

USA (3) vs Jamaica (group B winner)

Panama (4) vs Costa Rica (group A winner)

*Honduras are currently 3rd in group B

This should add context to why I was going on about Greenwood earlier. Jamaica are already a team you want to avoid but you add him into the mix and things get scarier. We don't match up well with Jamaica and never have, as they are the one team in this region that can nullify our athleticism, plus they have some real match winners. As we saw in the last edition, they knocked us out and then pushed the USA to the brink in the SF and that was with them missing Bailey (left the team), Demari Gray (suspended), Antonio (injured), and some others. I think Bobby DeCordova Reid was their biggest player remaining and many called it a B team, yet they took the USA to extra time.

There is a real chance Jamaica could take out us or the USA out in the QF, so we really need to lock down that 2nd seed if we can. We are in a good spot right now, but we need the USA to drop points to Mexico or Panama next window. And if course we must beat Panama.

[maplebanana]

On 9/13/2024 at 9:17 PM, Obinna said:

Here is his post from the other thread you are referring to:

Formula for Concacaf is Pts = Importance * (Result - Expected Result). Result is either 1 for a win, 0 for a draw, or -1 for a loss). Friendlies are worth an “importance” of 17.5

So, even without knowing the expected result specifically, we know our win is probably worth a swing of at least 35 pts — we gain at least 17.5 and they lose at least 17.5, assuming the expected result was that the Americans would win or at best a draw.  
Then even without considering the Mexico draw, the Americans probably drop even more points for drawing NZ at home when they were probably “expected” to win. 

Sorry folks, I saw the math bat symbol while I was on vacation last week but just getting back into it. The post I made a while back on the concacaf index is here. 

The 'expected result' is not binary, and is actually quite difficult to get to the max values. The equation I landed on was: 

P(win) = (10^((B-(A+C))/D)+1)^-1

Where:
- A is rating of team 1
- B is rating of team 2
- C is home field advantage, which I estimated as C=175 based on internal consistency of several friendlies
- D is a denominator that scales the teams in the calculation, which I estimated as D=750

So I would be surprised if there are these huge swings amongst relatively close opponents in friendlies. But I haven't re-checked my model since the last time I tried to calculate it. The +22 / -22 of ESV and GUA doesn't jive with my model, so I'll be the first to say ... I dunno. 🤷‍♂️ Maybe the index is calculated every few months and there were more results?

[Kent]

2 hours ago, maplebanana said:

Sorry folks, I saw the math bat symbol while I was on vacation last week but just getting back into it. The post I made a while back on the concacaf index is here. 

The 'expected result' is not binary, and is actually quite difficult to get to the max values. The equation I landed on was: 

P(win) = (10^((B-(A+C))/D)+1)^-1

Where:
- A is rating of team 1
- B is rating of team 2
- C is home field advantage, which I estimated as C=175 based on internal consistency of several friendlies
- D is a denominator that scales the teams in the calculation, which I estimated as D=750

So I would be surprised if there are these huge swings amongst relatively close opponents in friendlies. But I haven't re-checked my model since the last time I tried to calculate it. The +22 / -22 of ESV and GUA doesn't jive with my model, so I'll be the first to say ... I dunno. 🤷‍♂️ Maybe the index is calculated every few months and there were more results?

 I think the rankings don't fall that far behind. I'll be at least a bit surprised if we don't have new rankings before October 1st, and quite surprised if we don't have them before the next window opens.

Back to the formula though. This is CONCACAF, so maybe they have been tweaking the P(win) formula. So it could be that yours was pretty accurate, but it has changed since then. Maybe they are changing it again now to make sure USA finishes top 2.

[Olympique_de_Marseille]

On 9/4/2024 at 4:28 PM, narduch said:

The real sickos

The CONCACAF sickos can find the October CNL schedule here:

https://www.concacaf.com/nations-league/news/concacaf-confirms-october-schedule-for-2024-25-concacaf-nations-league-group-stage-matches/

+

If @Kent , @maplebanana or some other astute mind can figure out the CONCACAF rankings, by the end of the night of our friendly against Panama in Toronto, we should be able to calculate the Top 4 rankings and therefore know our November opponent.

CONCACAF reconfirmed in the link above that October rankings are the ones that matter.

[Unnamed Trialist]

Considering how close the rankings are and how important, it seems like the wrong time to play against a non Fifa rival if we do a second game.

Find a lesser rival, play at home, ensure the Fifa points. And a much easier quarter final in Nations League. 

[Obinna]

55 minutes ago, Unnamed Trialist said:

Considering how close the rankings are and how important, it seems like the wrong time to play against a non Fifa rival if we do a second game.

Find a lesser rival, play at home, ensure the Fifa points. And a much easier quarter final in Nations League. 

The problem is there is almost literally nobody available. You can probably count on your hands and toes how many available teams there are amongst the +200 FIFA member nations. The best bets would (I think) come from the minnows of AFC.

[Mihairokov]

58 minutes ago, Obinna said:

The problem is there is almost literally nobody available. You can probably count on your hands and toes how many available teams there are amongst the +200 FIFA member nations. The best bets would (I think) come from the minnows of AFC.

I believe Laos, Pakistan, Myanmar, and Mongolia are available but we're talking less than a month for them to get organized, get flights, hotels, and cobble a team together. Seems incredibly unlikely.

If Canada Soccer were desperate for coin and forward-thinking they'd host Philippines in any of our largest cities and let the ticket sales do what they must.

Edited 56 minutes ago by Mihairokov