ATD 2018 Rules Discussion

[rmartin65]

I'm surprised the round robin idea is getting any traction. That one round of voting will be very time consuming. You guys do realize that, right?

Perhaps, but there will also be a lot going on. Instead of waiting for one opponent to go back and forth with, there will be 3-4. Odds are, you will have at least 1 opponent interested in having a debate. So, while I think that it will add some time, it wont feel stagnant like some playoff rounds do.

 

[Theokritos]

I'm surprised the round robin idea is getting any traction. That one round of voting will be very time consuming. You guys do realize that, right?

For the record, my proposal does not include just one round of voting per division. I propose as many rounds of voting per division as there are divisional opponents: In a division of 4, every GM has 3 opponents, so there would be 3 rounds of voting.

I agree with all this, and I'd submit this for consideration: Since we're looking to speed up what's kind of a slogging playoff process, how about have divisions of 4, and we treat the round robin as a single, quick process where GMs vote on all games at once?
So let's say division A has the Penguins, the Senators, the Predators, and the Ducks. The volunteer in charge of creating the series thread would post all 4 rosters and specifically direct voters to vote on the following, and provide game scores:
Penguins vs. Senators
Predators vs. Ducks
Penguins vs. Predators
Senators vs. Ducks
Senators vs. Predators
Ducks vs. Penguins

As far as I'm concerned, that would defeat the main purpose of the round-robin proposal: Giving every GM several series in which he can argue his case against individual opponents. I think having a single vote wouldn't effectively encourage participants to look into the individual match-ups all too thoroughly. At least not as thoroughly as with separate votes on separate series.

 

[Theokritos]

Another point about the round robin is that as far as I understand it, you will play more opponents than you usually do, and maybe in less time too.That mkes it difficult to get to know your opponent's team better, which is essential in a playoff series.

Only if you vote on all series at once, which is not what I would suggest. Let's stick with the set-up you have suggested: 24 teams in 6 divisions with 4 teams per division. For example, divsion A has the following teams: Penguins, Senators, Predators, Ducks, Devils and Kings. Here's how the voting would work:

I) REGULAR SEASON

Round 1:

Division A:
Penguins vs Senators (8 games)
Predators vs Ducks (8 games)
Devils vs Kings (8 games)

Division B, C and D:
Same system: 3 individual match-ups per division.

Every GM has the opportunity to argue his case against the current opponent: Penguins' GM vs Senators' GM and vice versa, Predators' GM vs Ducks' GM and vice versa, and so on.

Vote 1:

You vote on the round 1 match-ups from each division. Your ballot will look something like this:
Penguins vs Senators 4-4
Predators vs Ducks 7-1
Devils vs Kings 6-2
etc

Round 2:

Division A:
Penguins vs Ducks (8 games)
Predators vs Kings (8 games)
Devils vs Senators (8 games)

Divisions B, C and D:
Same system: 3 individual match-ups per division.

Again, every GM has the opportunity to argue his case against the current opponent.

Vote 2:

You vote on the round 2 match-ups from each division. For example:
Penguins vs Ducks 5-3
Predators vs Kings 4-4
Devils vs Senators 3-5
etc

The process or arguing and voting on the individual divisional match-ups continues until all divisional opponents have faced each other. In our scenario (4 divisions with 6 teams each) it would take 5 rounds of discussion and voting to finalize the regular season. (Again: The main benefit is that every single participant gets to argue his case against several individual opponents. This is not the case in the current system where you're only guaranteed to get 1 vote on the regular season as a whole and then 1 vote on an individual playoff series, the latter one often being a rather hopeless case if you've been ranked low in the regular season.)

II) PLAYOFFS

The 2 top teams from each division advance to the playoffs. With 4 divisions, that means you've got 8 teams in the playoffs, all of them #1 or #2 seeds from the first stage. You will rarely get match-ups where the outcome is obvious in advance. And you will only have 3 playoff rounds overall (QF, SF, F) so that the overall length of the ATD shouldn't become too much. (5 rounds in the regular season + 3 playoff rounds = 8 overall rounds of discussion and voting.)

 

[BenchBrawl]

^

My problem with this structure is that it will take a lot of time, and the draft already ended too late last year.

 

[Theokritos]

My problem with this structure is that it will take a lot of time

The actual time will depend on the specific setup. For example, if you have divisions of 4 teams (instead of 6) the regular season will only require 3 rounds of voting (instead of 5). With 24 participants overall, you could have six divisions of 4 (instead of four divisions of 6, as in my prior scenario) and you could have the six division winners plus the two #2 teams with the best records advancing to the playoffs. That would give you 6 rounds of voting overall (3 regular season match-ups, 3 playoff rounds) – just like in the last couple of ATDs. Therefore, it shouldn't take any longer than usual.

and the draft already ended too late last year.

Looking back at the last couple of ATDs, it seems like the Finals threads were usually up by mid May. Any idea why it took three weeks longer in 2017? (Of course, I was the one tallying votes, but I have no insight into the prior ATDs so I cannot compare and figure out what the difference was.)

 

[BenchBrawl]

The actual time will depend on the specific setup. For example, if you have divisions of 4 teams (instead of 6) the regular season will only require 3 rounds of voting (instead of 5). With 24 participants overall, you could have six divisions of 4 (instead of four divisions of 6, as in my prior scenario) and you could have the six division winners plus the two #2 teams with the best records advancing to the playoffs. That would give you 6 rounds of voting overall (3 regular season match-ups, 3 playoff rounds) – just like in the last couple of ATDs. Therefore, it shouldn't take any longer than usual.

I don't want to sound like I'm strongly against it.I'm not.Something seems off though, but then again I'm so tired I have both eyes in the same hole.If we ignore the amount of rounds (I think last year was 5 by the way, but could be wrong), isn't the total amount of match-ups much larger with this round-robin structure? Because we also need to think about the voters.It's fun to read series, but there's a limit to how many you can read, but maybe this is a false problem.

Consider a simple case, 16 teams total, 4 divisions of 4 teams.If we go by the standard 16-teams playoffs bracket (say conferences of 8 teams instead of using divisions, but something similar would be true if we used divisions), we have a total of 15 playoffs series.

If we did the round-robin (with only the Top-1 team of each division making the playoffs) from how I understood it, we would have a total of 27 playoffs series (or match-ups if you prefer).That's almost twice as many, unless I made a big blunder in my thinking.If we let the Top-2 of each divisions, that's even more.

Again, I don't want to sound like I particularly hate it.I do think it's on the complicated side, and the more complicated the more potential for hidden problems that will pop up.But maybe it's worth trying.

Looking back at the last couple of ATDs, it seems like the Finals threads were usually up by mid May. Any idea why it took three weeks longer in 2017? (Of course, I was the one tallying votes, but I have no insight into the prior ATDs so I cannot compare and figure out what the difference was.)

I have no idea why ATD2017 took so long.It's puzzling, since from what I recall the draft went smoothly.MIght just be each round being delayed due to lack of voters or other reasons.I recall GMs asking for more time at least once too.

 

[BenchBrawl]

Another point against the round-robin structure, is that if we imagine a strong contender trapped in a weak division, he will waste his time for 3 rounds.For me this is why I dislike anything to do with divisions anyway, and always prefered the good old conference rankings, where you progressively face stronger and stronger opponents as you advance in the bracket.With the round-robin system, you could theoretically face weaker and weaker opponents for at least 3 rounds straight.

I know people complained that it was hard to rank a conference, but I never understood it.The teams that are close, say the middle-tier teams, they will be interchangeable in the ranking, probably finishing 4th-6th or whatever, then they will have the opportunity to face each other and make their case for every voters to see so they can re-adjust.The top teams will get voted high and the weak teams will get voted low, and yes, those series will be predictable.But so will many series inside the round-robin system.

OTOH, divisions add a dimension to the competition, where you have to keep an eye on your divisional rivals and their assets and sometimes respond accordingly, which is less true of a big conference.

 

[jarek]

If you look at my suggestion, it breaks down to the following (assuming the same 16 teams):

You compare each team to every other team in the same division (which you would be doing regardless for regular season rankings anyways). At most, this is 6 comparisons per division (1v2, 1v3, 1v4, 2v3, 2v4, 3v4). Because the stakes are a lot higher than usual, I suspect people would take more time to do these comparisons. Realistically however, maybe 1-2 teams in each division will actually stand out and you can largely discard the rankings of the bottom 2 teams.

This leaves you with 4 serious comparisons to think about in the first "round". Then you would get the usual division winners facing each other, and finally the finals. A total of 7 "series" to seriously consider. Maybe you can bump it to 10 if a couple of divisions are really tight. Either way, this would be way more efficient than how we've done things in the past and would also substantially reduce the amount of time the post-draft process takes.

I personally don't really like the round robin idea. It essentially becomes a more convoluted version of what I've just outlined above.

 

[Theokritos]

isn't the total amount of match-ups much larger with this round-robin structure?

(...)

Consider a simple case, 16 teams total, 4 divisions of 4 teams.If we go by the standard 16-teams playoffs bracket (say conferences of 8 teams instead of using divisions, but something similar would be true if we used divisions), we have a total of 15 playoffs series.

If we did the round-robin (with only the Top-1 team of each division making the playoffs) from how I understood it, we would have a total of 27 playoffs series (or match-ups if you prefer).That's almost twice as many, unless I made a big blunder in my thinking.

Of course. If you want to give all involved GMs at least a few competitive series, it's unavoidable to have a larger amount of match-ups.

Because we also need to think about the voters.It's fun to read series, but there's a limit to how many you can read, but maybe this is a false problem.

ATD participants can correct me if they disagree, but it seems there rather is a limit to how many series you can follow at the same time than a specific overall limit. Of course I wouldn't expect voters to keep track of 27 match-ups at once. If we stick with the latest scenario I've lined out (6 divisions of 4 teams each, 24 teams overall), there are 6x2=12 match-ups voters have to follow in any given regular season round. For comparison: In the 2017 ATD, voters also had to follow 12 match-ups (8 in the playoffs, 4 in the consolation round) in the first head-to-head round.

Match-up per round in the round-robin scenario:
Round 1 (RS): 12 match-ups (2 in your own division, 10 in the others)
Round 2 (RS): "
Round 3 (RS): "
Round 4 (QF): 4 match-ups
Round 5 (SF): 2 match-ups
Round 6 (F): 1 match-up

If we ignore the amount of rounds (I think last year was 5 by the way, but could be wrong)

No, you are right. My bad. Looking back at the ATD 2017 threads, the time table was this:

January 13: First picks made.
March 13: Last picks made.
Vote 1 → April 10: Threads posted for first round of head-to-head match-ups.
Vote 2 → April 27: Threads posted for second round of head-to-head match-ups.
Vote 3 → May 19: Threads posted for third round of head-to-head match-ups.
Vote 4 → June 6th: Thread posted for fourth and final round.
Vote 5 → June 25th: Winner announced.

Another point against the round-robin structure, is that if we imagine a strong contender trapped in a weak division, he will waste his time for 3 rounds. For me this is why I dislike anything to do with divisions anyway, and always prefered the good old conference rankings, where you progressively face stronger and stronger opponents as you advance in the bracket.With the round-robin system, you could theoretically face weaker and weaker opponents for at least 3 rounds straight.

Well, that of course is the perspective of someone who has been a contender and even an ATD winner a few times. Try to see the draft from the perspective of the participants who haven't made it as far (yet): They dedicate two months to pick their team, another two or three weeks to get the line-up straight and argue the overall case for their team – and after one vote they're effectively eliminated from the competition.

 

[BenchBrawl]

ATD participants can correct me if they disagree, but it seems there rather is a limit to how many series you can follow at the same time than a specific overall limit. Of course I wouldn't expect voters to keep track of 27 match-ups at once. If we stick with the latest scenario I've lined out (6 divisions of 4 teams each, 24 teams overall), there are 6x2=12 match-ups voters have to follow in any given regular season round. For comparison: In the 2017 ATD, voters also had to follow 12 match-ups (8 in the playoffs, 4 in the consolation round) in the first head-to-head round.

Match-up per round in the round-robin scenario:
Round 1 (RS): 12 match-ups (2 in your own division, 10 in the others)
Round 2 (RS): "
Round 3 (RS): "
Round 4 (QF): 4 match-ups
Round 5 (SF): 2 match-ups
Round 6 (F): 1 match-up

Fair enough.That's still an extra 2 weeks in the schedule though.

No, you are right. My bad. Looking back at the ATD 2017 threads, the time table was this:

January 13: First picks made.
March 13: Last picks made.
Vote 1 → April 10: Threads posted for first round of head-to-head match-ups.
Vote 2 → April 27: Threads posted for second round of head-to-head match-ups.
Vote 3 → May 19: Threads posted for third round of head-to-head match-ups.
Vote 4 → June 6th: Thread posted for fourth and final round.
Vote 5 → June 25th: Winner announced.

This is really weird.If we kept to a strict 2 weeks schedule for each round, the draft would have ended June 5th.Still late, but better.

Well, that of course is the perspective of someone who has been a contender and even an ATD winner a few times. Try to see the draft from the perspective of the participants who haven't made it as far (yet): They dedicate two months to pick their team, another two or three weeks to get the line-up straight and argue the overall case for their team – and after one vote they're effectively eliminated from the competition.

OK, but I wonder if those GMs really want this opportunity.Will they use it, or will we have four cemetery threads instead of one? I think everyone should speak out now, so we know what the majority wants.I'm fine with any structure.

 

[Theokritos]

Fair enough.That's still an extra 2 weeks in the schedule though.


This is really weird.If we kept to a strict 2 weeks schedule for each round, the draft would have ended June 5th.Still late, but better.

I wonder whether the time between the last picks and the first vote could be reduced from three (or more) weeks to two weeks.

OK, but I wonder if those GMs really want this opportunity.Will they use it, or will we have four cemetery threads instead of one?

That indeed is the crucial question. It was my fundamental assumption that a lack of actual competitive match-ups for many GMs was what hurt activity and participation after the first round of voting. But of course I'm only making suggestions as an outsider here, it's most definitely up to you ATDers to figure out what you want and what you need as a group.

 

[ImporterExporter]

Honestly, anything that even presents a chance at prolonging the post draft process (which is dragged out to begin with) is a non starter for me. I get the idea behind the round robin style tournament but most people frankly won't invest that much time into it. I can 99% guarantee it.

People are going to determine the "best" teams once final rosters are presented. How many times have people's minds REALLY been changed on a message board?

To be quite frank, if I drafted a team and nothing else happened I wouldn't be disappointed. But I'll go along with whatever is decided and be active regardless. I simply urge people to break away from the drawn out style of playoffs we have now. If that means some teams don't make the playoffs, so be it.

 

[BenchBrawl]

Honestly, anything that even presents a chance at prolonging the post draft process (which is dragged out to begin with) is a non starter for me. I get the idea behind the round robin style tournament but most people frankly won't invest that much time into it. I can 99% guarantee it.

People are going to determine the "best" teams once final rosters are presented. How many times have people's minds REALLY been changed on a message board?

To be quite frank, if I drafted a team and nothing else happened I wouldn't be disappointed. But I'll go along with whatever is decided and be active regardless. I simply urge people to break away from the drawn out style of playoffs we have now. If that means some teams don't make the playoffs, so be it.

This can happen, but if one team is much better Idon't see myself changing my mind.I've changed my mind many times in close series due to the debates though.

 

[rmartin65]

Honestly, anything that even presents a chance at prolonging the post draft process (which is dragged out to begin with) is a non starter for me. I get the idea behind the round robin style tournament but most people frankly won't invest that much time into it. I can 99% guarantee it.

Which is why I like JE's idea on the previous page (I like my augments to it as well, obviously). It is like a round robin, but the round robin part is all done at once, so its not taking forever. Everybody gets their chance to argue, it doesnt add a bunch of time since all the division battles happen at the same time, and the playoffs are short and sweet (either the just the 1st seeds advance, or the 1st and 2nd seeds).

People are going to determine the "best" teams once final rosters are presented. How many times have people's minds REALLY been changed on a message board?

I have changed my mind on several playoff series due to arguments and match-ups, as have numerous others, I would imagine. Haven't you?

 

[ImporterExporter]

I have a few times, yes.

Honestly I think player biases are a bigger issue for some folks in all reality. I honestly am not sure I'd ever draft Crosby (even being a huge fan and if the value was there) because he's so polarizing for some people. But there is no way to truly know or limit how much positive/negative bias a player gets from any GM or voter.

 

[BenchBrawl]

^ I don't see Crosby as polarizing at all in the ATD.I can think of one GM who doesn't like him, but that's it.I don't recall any big splash when he was taken in the 1st round last year, and he was clearly a steal for me in 2016.

Problem with Crosby is that he's a center and stuck in a cluster of other centers with similar value (arguably,which is the point).

 

[BenchBrawl]

Which is why I like JE's idea on the previous page (I like my augments to it as well, obviously). It is like a round robin, but the round robin part is all done at once, so its not taking forever. Everybody gets their chance to argue, it doesnt add a bunch of time since all the division battles happen at the same time, and the playoffs are short and sweet (either the just the 1st seeds advance, or the 1st and 2nd seeds).

One problem I can see with this system is that there's so many match-ups to consider at the same time that it almost loses it's "match-up" quality, meaning the individual confrontation with all the details that comes with it.It would be hard to follow everything, and I fear people would just vote systematically, which is what they would do if they just ranked the divisions from the get go.

 

[BenchBrawl]

Regarding the trade issue: what if trading already drafted players (rather than just picks) was banned - either for good, or at least until every player is drafted. Would that satisfy Dreakmur's* problem with the way last draft went down while also satisfying those who still want the option of trading?

BTW, a bit late, but I'd be fine with this.I think it's a good idea.

 

[ResilientBeast]

BTW, a bit late, but I'd be fine with this.I think it's a good idea.

Agreed I think this solves most of the problems since, we have a standard when it comes to pick for pick

 

[rmartin65]

One problem I can see with this system is that there's so many match-ups to consider at the same time that it almost loses it's "match-up" quality, meaning the individual confrontation with all the details that comes with it.It would be hard to follow everything, and I fear people would just vote systematically, which is what they would do if they just ranked the divisions from the get go.

Sure, thats definitely a concern. No system is perfect though, what we are doing now is trying to decide which system is the best. Personally, I like this idea, as it is a hybrid of the regular style of playoffs and the round robin (which I do think will be too lengthy).

 

[BenchBrawl]

The thing with having 2-3 match-ups at the same time, is that, when I'm in a playoff series, I'm discovering my own team more in depth, but more importantly, I'm discovering the other team more in depth too.This will be impossible to do in this crazy round.Whether that's a real concern or not is up to others to decide.I'm not strongly against it, even if it looks like I am.I'm merely pointing out potential problems for every structure brought forward.

 

[BenchBrawl]

Right now we have the following structures (please correct me if you see any mistake in your structure, I did this quickly):

• Theokritos's round-robin proposal, where every team inside division X play against every other team inside division X, done round by round (so not simultaneously).Then the Top-2 teams of each division make the playoffs and we have a small bracket.

• jarek proposed that we just directly vote for the teams, and the Top-1 or Top-2 of each division makes the playoffs and we have a small bracket.So your chance to defend your team is before this actual voting gets done, inside a "divisional thread" where every team of division X make their case simultaneously, but without specific match-ups in mind.

• Johnny Engine proposed to have a first round where every team inside division X play against every other team inside division X, but simultaneously.Then the Top-1 or Top-2 of each division makes the playoffs and we have a small bracket.

• I propose that we just rank entire conferences and the Top-8 teams of each conference make the playoffs, regardless of how many teams we end up having in the draft.Then we have the usual bracket.Also open to have division leaders automatically get the top few spots.

• Another proposition of mine is that we rank the conferences and every team make the playoffs.If the number of teams is not neat, the top teams get byes.

Anything else?

 

[jarek]

That seems about right.

One of the main reasons that I really want a short playoff to be honest is that I'm not sure how much time I'll have on my hands. I am currently living in an apartment with my sister, and very soon I expect to be moving to a home. Life is a bit hectic right now, and I'm not sure if it's going to get better or worse.

 

[Theokritos]

Honestly, anything that even presents a chance at prolonging the post draft process (which is dragged out to begin with) is a non starter for me.

the round robin (which I do think will be too lengthy).

If the general sentiment is that shortening the entire process should be a priority then the round-robin system is definitely not the way to go. It would take as much time as the 2017 ATD or even a bit more.

Bench Brawl's proposal includes 4 rounds of voting: One regular season vote and then 3 playoff rounds & votes (QF, SF, F).
With the proposals by jarek and Johnny Engine the number of rounds depends on the specific setup. If you have 4 groups of 6 teams each and the top 2 teams from each division advance, you'd get 4 rounds of voting (just as in BenchBrawl's model). If only the division winners advance, you'd cut it down to just 3 rounds of voting overall.

For comparison: The 2017 ATD had 5 rounds of voting that stretched out from early April to late June 2017. Assuming a schedule of 2 weeks per playoff round, a 4-round-draft could be expected to end in the second half of May and a 3-round-draft in early May.

 

[ResilientBeast]

See I think we could still achieve the round robin

For each team in your division you guys "Play" five times
For each team in your conference not in your division you "Play" three times
For each cross conference team you "Play once"

It would require you to spend a little more time voting and thinking about your votes but that would truly mix things up IMO

We could accomplish this with a google poll type thing, that just requires you to tick the boxes and hit submit