I guess Helllas will take a win with a score 24-0.

"I just made it more complicated and detailed."

so you wont remove it? no respect to the dead...

FIrst matchups were made according to seeding, and as for seeding I'd rather not disclose the details of how it was made. For simplicity let's just say it's 100% my subjective opinion (although it isn't tbh).

So happy you cleared that up.

@Rento

Are the individual player matchups between teams random?

Yes, individual player matchups are randomized. I created scripts for Rento to handle randomization and unique pairing.

Edited 10/16/2019 18:04:53

Player matchups are random as Justin explained.

As for Hellas, they got a bye. Simply imagine there's a team that's so bad that it loses all its games to anyone who's paired against them. That's the bye team.

I am a supporter of the bye team

I think it's boring not to get to play a single game in the round. How this bye system work through other rounds? Especially during the last round?

This looks fun, good luck to all teams. Nations, I mean.

Is the colour coding in the result sheet according to expected result?

Some formatting suggestions:
some borders would be nice
I like the colour coding, but green should not be the standard colour then

I think how to decide on seeding and matchmaking is interesting and would appreciate it if you explain it in detail.
The sheet says you used 1v1 ladder performance as an indicator, why the secrecy here?

I got another question, what happens if both teams get tied? Like both of them get 12:12?

If both teams tie in the end, they both get half a win. They will have 0.5 matchups won.

The results in the Results tab are always [seeded team] vs [unseeded team]. If there's an upset, I change the wording from "defeats" into "loses to" and change the color to make it visible at glance.

I will not explain the seeding mechanism in detail, sorry :P It's because the current algorithm can be easily gamed if you know how it works. Two main factors are indeed 1v1 ladder results and number of players in a team, that's all I'm going to say :P

Correct me if I am wrong but this is how I think the system works.

We have 39 teams.Lets assume that from the 19 pairs we have 16 winners and 3 pairs were tied.

The 16 winners plus the bye winner make a group of 17 and they are ranked based on their winning score.Since there are 17 teams it will be treated like a group of 18 therefore ranked 1 team will pair with ranked 10.Ranked 2 will pair with rank 11 and so on.The team ranked 9 will not have a pair so it should pair with the first of the next group.

The next group is the one with the 6 teams that tied.Since they have all of them equal scores the initial power rank will be applied.So the 9th team from the previous group of winners will pair with top of the tiers.The 5 rest will be treated like a group of 6 and we will get pairs 1st-4th, 2nd-5th.The 3rd team will pair with the first team of the next group that is the group of the teams that lost on the 1st round.

So 16 teams that lost will get ranked based on the points they scored on 1st round.The top of them will pair with the remain team of the previous group.The remaining 15 will create a group of 16.The pairs will be 1st-9th, 2nd-10th...7th-15th.The 8th team will get the bye.

I hope that helps (if it is correct).

You got the general idea right, but it's even more complicated than that. Because 2 teams from the same nation can't meet each other, so I added a way to avoid that. Also it's very likely that not all the matchups will be resolved when the next round games are created. I need to deal with that too.

So I don't think you have a way to accurately predict whom you're gonna meet. I think it adds to the fun :P

isn't that biasing that if you lose the first match up it's very unlikely you'll ever match up against a team in the top 10? you'll always get "easier" matches and still might end up 5-1 in the end

Edited 10/17/2019 18:27:54

Due to slide ranking favorites really shouldn't be losing the first matchup. Each team is facing a team 20 spots lower than it. For example ps, your Portugal team is seeded 14th and is facing my team the 34th seed USA D. I think as a whole we can agree your team is stronger than mine, and it has played out that way so far.

well seems that Rento algorythm working perfectly, didnt counted but seems like >80% wins are from nations which are considered favorites... :) so it makes sense.

Edited 10/17/2019 09:47:09

@Rento
For matching teams for next round in case of ongoing matches

you may use your judgement to assess the most likely outcome and assume points by consequence, which may be ok, but arbitrary and requiring some effort

or

if you like to simplify and avoid to use arbitrary judgement do something like:

- if the favourite is winning or has equal wins, then assume 1-0
- if the favourite is losing, then assume 0.5-0.5

Edited 10/17/2019 16:10:33

If the matchup is undecided when the next games are created, I'll assume that both teams score a win (for matchmaking purpose only! that win won't register in the ranking obviously).

So for the team that was going to win, absolutely nothing will change.
But the team that was gonna lose will get a harder opponent than if they lost the matchup in time.
This is necessary not only because it guarantees that no 2 teams will end the season with perfect records, never meeting each other. But it also deals with stalling losses.

Nice try :)

just want to say, America is by far the THICCest region.