Posted by hurst47 on 8/13/2019 7:10:00 PM (view original):

Posted by ozomatli on 8/13/2019 6:59:00 PM (view original):
hurst47, the difference in average rank of owners by league is minimal (the only reason it isn't perfect is due to commissioner assignments). Therefore, it did not bias the randomly generated divisions, so the results are truly luck.

Again, last year the 3rd ranked owner received the most difficult divisional pairings. The 8th ranked owner was also in the top 10. It's just random, just like any other league on WIS.

Well it didn't happen this year. I'd even argue that aligning by owner rating like is done in some regular leagues would make the tournament more interesting. Put all the top ranked owners together and low ranked owners together. Those of us who enjoy the Sim but don't put in all the study required to be an expert would appreciate something like that.

But that goes against the spirit of the WIS Championship - to determine the best owners. If top owners are in divisions that are way harder than lower ranked owners, how do we know who exactly are the best owners? Did someone unexpected get into the Cage (Top 24) just because they were the better than the weak competition they played? Did an owner who clearly had top 24 level rosters get excluded because they had to play a whole bunch of other top 10 level rosters?

Posted by justinlee_24 on 8/13/2019 6:55:00 PM (view original):
That's the final word... those look like balanced divisions to me.

Some people can't tolerate any kind of dissent. I'm guessing you're one of those types.

Posted by justinlee_24 on 8/13/2019 7:14:00 PM (view original):

Posted by milest on 8/13/2019 7:08:00 PM (view original):
I'm ranked 15th and I have the 26th most difficult schedule. I'm feeling left out of the (perceived) schedule advantage for top ranked owners.... I did a quick glance and I think it is the 2nd worst draw when you consider owner ranking in correlation to difficulty of schedule.
I will admit that it is it is fairly disheartening, especially when I see the ease of the schedule for some of the past winners and favorites.
I'm certainly not claiming it was rigged or anything of that nature, but coming from my view, it's hard to see it as exactly a 'fair' process to all 96 owners involved.

The only way to fix that would be to forcibly make every division have the same 48.5 average and have ozomatli manually determine everyone's division and conference in every league. That would be the only way to remove people getting lucky or unlucky draws.

Manually creating the leagues and the manually creating the divisions based on the owner ranking is what should be done in my opinion. If commissioners don't want to manually input divisions, then I'll volunteer to do it. I've done it in the NBA Sim many times

Posted by justinlee_24 on 8/13/2019 7:14:00 PM (view original):

Posted by milest on 8/13/2019 7:08:00 PM (view original):
I'm ranked 15th and I have the 26th most difficult schedule. I'm feeling left out of the (perceived) schedule advantage for top ranked owners.... I did a quick glance and I think it is the 2nd worst draw when you consider owner ranking in correlation to difficulty of schedule.
I will admit that it is it is fairly disheartening, especially when I see the ease of the schedule for some of the past winners and favorites.
I'm certainly not claiming it was rigged or anything of that nature, but coming from my view, it's hard to see it as exactly a 'fair' process to all 96 owners involved.

The only way to fix that would be to forcibly make every division have the same 48.5 average and have ozomatli manually determine everyone's division and conference in every league. That would be the only way to remove people getting lucky or unlucky draws.

Actually, that wouldn't be more fair. Randomness is guaranteed to be fair, while that solution would only be as fair as my ranking system is in the first place. Which, I put my best effort into, sure but, as I mentioned, there are flaws (hence is not being a perfect prediction each year). Randomness is superior if the goal is to be fair.

Posted by hurst47 on 8/13/2019 7:10:00 PM (view original):

Posted by ozomatli on 8/13/2019 6:59:00 PM (view original):
hurst47, the difference in average rank of owners by league is minimal (the only reason it isn't perfect is due to commissioner assignments). Therefore, it did not bias the randomly generated divisions, so the results are truly luck.

Again, last year the 3rd ranked owner received the most difficult divisional pairings. The 8th ranked owner was also in the top 10. It's just random, just like any other league on WIS.

Well it didn't happen this year. I'd even argue that aligning by owner rating like is done in some regular leagues would make the tournament more interesting. Put all the top ranked owners together and low ranked owners together. Those of us who enjoy the Sim but don't put in all the study required to be an expert would appreciate something like that.

This goes against the spirit of the tournament, which is to find the best owners. Artificially forcing them to play each other in Round 1, again assuming that my ranking system is the absolute truth (which it isn't), would lower the quality of Round 2. Given that the themes are more difficult in Round 2, this would also create a wider disparity in the results of Round 2 (not to mention Round 1, where top owners would likely be artificially grouped together relative to everyone else) and make Round 2 less interesting, since we'd have clear expectations around who would win.

If it's basically going to be random anyway, then having rankings just makes it hard for people who feel like they are getting a bad draw in comparison to those rankings. I feel that if the rankings exist, they should be utilized. While your rankings may not 100% accurately reflect the owners' acumen, it does a better job than random.
Obviously, it's just my opinion, but I would do 1 of 2 things: Either scrap the rankings all together and make the tournament draws 100% random or follow the ranking as closely as possibly to set the tournament up based on seeds.

There is another factor everyone is forgetting. Not everyone plays the same at every cap level. As an example I am ranked 37th which I think is pretty fair BUT what my track record means is if I am in any league with a cap of 80 mil or lower you want me in your division because I haven't historically done well. But you definitively don't want me in your division if the cap is greater than 110 million as historically I usually make the playoffs. So what do you do, rank me in the bottom quarter for the 70 million cap and the upper quarter in the 130 million cap.

Also there are only 2 or 3 owners I look as possibly unbeatable if they are in my division. A couple of tournaments ago I was caught in a 100 million cap "division of death" with 3 owners ranked in the top twenty and me. It was actually a lot of fun competing. If I remember right we all finished with an 84-88 wins with a tiebreaker to move on. I have also been in divisions with 3 of other guys in the 11 - 65 range. We weren't considered a division of death at the start of the year but at the end it was considered one of the most competitive divisions in the field because we all had records in the 87 - 92 range.

I could probably name 20 - 25 guys ranked in the 11 - 65 range that the top ten don't want in their divisions. You can throw a net over us and use different criteria to sort us. Most of us only miss the 2nd round because we screw up one team or two but the other 4 or 5 are pretty good.

You have to assign leagues ahead of time, in order to get people's rosters entered. The random division alignment is the fairest way to allocate within the league.

Oh, and for the mathematically challenged... anybody seeded in the top 48 is a favorite to have an easier than average division. Why? Not because the tournament is rigged. It's because that's how the math works. The better the seeding, the more likely that person will have an easier division because their seeding number isn't included in the measurement of *his* division strength.

I think we've wasted enough time trying to help hurst47 understand this. He clearly chooses not to get it.

Posted by milest on 8/13/2019 7:30:00 PM (view original):
If it's basically going to be random anyway, then having rankings just makes it hard for people who feel like they are getting a bad draw in comparison to those rankings. I feel that if the rankings exist, they should be utilized. While your rankings may not 100% accurately reflect the owners' acumen, it does a better job than random.
Obviously, it's just my opinion, but I would do 1 of 2 things: Either scrap the rankings all together and make the tournament draws 100% random or follow the ranking as closely as possibly to set the tournament up based on seeds.

I do see this viewpoint and if this was my job, would agree with it, but this feels like a case of taking something too seriously to me.

The rankings are helpful but, more than that, they're fun, imperfect and shouldn't be doubled down on. If we did that, to mllama's point, we'd have to consider a whole bunch of additional features that would considerably increase the complexity of doing rankings and take the fun out of this for me. It is time consuming enough to run this tournament as it is. I do it because I love competing, the other owners and the game itself, but probably can't afford to add more time to it than I already do.

As for doing randomization across the board, that would lead to less balance than we currently have at no additional benefit, and would also get messed up by dropouts anyway. So, that isn't a viable solution here. What we have is really a happy medium that has worked well for 15 years and is pretty well engineered at this point -- if people aren't satisfied with that then (and I hate to be defensive here) I offer them to try running this themselves.

At this time this isn't something I'm willing to reconsider, but do appreciate the debate and feedback.

Posted by schwarze on 8/13/2019 7:57:00 PM (view original):
You have to assign leagues ahead of time, in order to get people's rosters entered. The random division alignment is the fairest way to allocate within the league.

Oh, and for the mathematically challenged... anybody seeded in the top 48 is a favorite to have an easier than average division. Why? Not because the tournament is rigged. It's because that's how the math works. The better the seeding, the more likely that person will have an easier division because their seeding number isn't included in the measurement of *his* division strength.

I think we've wasted enough time trying to help hurst47 understand this. He clearly chooses not to get it.

Oh, I get it. The double randomization process the way its done skews the results to give the higher ranked owners an easier draw. I have yet to see anything to convince me otherwise. You're bordering on a scam if you fail to realize this. You top dogs who pay for this thing with a few measly site credits from your obscenely large stash just don't understand the huge advantage you already have just by being experts in how WIS configures their simulation algorithm and through repetition have identified all the undervalued outliers within the system. I guess in light of all this maybe you should issue a warning to the uninformed masses that this is a tournament for experts only and set a minimum owner rating for participation otherwise the more casual participant's tournament is over before it even begins and we're unknowingly scammed out 75 bucks for nothing.

Posted by hurst47 on 8/13/2019 8:44:00 PM (view original):

Posted by schwarze on 8/13/2019 7:57:00 PM (view original):
You have to assign leagues ahead of time, in order to get people's rosters entered. The random division alignment is the fairest way to allocate within the league.

Oh, and for the mathematically challenged... anybody seeded in the top 48 is a favorite to have an easier than average division. Why? Not because the tournament is rigged. It's because that's how the math works. The better the seeding, the more likely that person will have an easier division because their seeding number isn't included in the measurement of *his* division strength.

I think we've wasted enough time trying to help hurst47 understand this. He clearly chooses not to get it.

Oh, I get it. The double randomization process the way its done skews the results to give the higher ranked owners an easier draw. I have yet to see anything to convince me otherwise. You're bordering on a scam if you fail to realize this. You top dogs who pay for this thing with a few measly site credits from your obscenely large stash just don't understand the huge advantage you already have just by being experts in how WIS configures their simulation algorithm and through repetition have identified all the undervalued outliers within the system. I guess in light of all this maybe you should issue a warning to the uninformed masses that this is a tournament for experts only and set a minimum owner rating for participation otherwise the more casual participant's tournament is over before it even begins and we're unknowingly scammed out 75 bucks for nothing.

So you would rather have the #12 ranked owner get scammed as he would have to play the top 11 owners in your ideal league alignment. Talk about rigged...

Posted by hurst47 on 8/13/2019 8:44:00 PM (view original):

Posted by schwarze on 8/13/2019 7:57:00 PM (view original):
You have to assign leagues ahead of time, in order to get people's rosters entered. The random division alignment is the fairest way to allocate within the league.

Oh, and for the mathematically challenged... anybody seeded in the top 48 is a favorite to have an easier than average division. Why? Not because the tournament is rigged. It's because that's how the math works. The better the seeding, the more likely that person will have an easier division because their seeding number isn't included in the measurement of *his* division strength.

I think we've wasted enough time trying to help hurst47 understand this. He clearly chooses not to get it.

Oh, I get it. The double randomization process the way its done skews the results to give the higher ranked owners an easier draw. I have yet to see anything to convince me otherwise. You're bordering on a scam if you fail to realize this. You top dogs who pay for this thing with a few measly site credits from your obscenely large stash just don't understand the huge advantage you already have just by being experts in how WIS configures their simulation algorithm and through repetition have identified all the undervalued outliers within the system. I guess in light of all this maybe you should issue a warning to the uninformed masses that this is a tournament for experts only and set a minimum owner rating for participation otherwise the more casual participant's tournament is over before it even begins and we're unknowingly scammed out 75 bucks for nothing.

The sign up post says "serious WIS owners only"

Exactly. I’d have expected this debate 2-3 months ago. I don’t understand ******** about the format AFTER you’ve paid and entered your team with full knowledge of how it works.

Posted by hurst47 on 8/13/2019 8:44:00 PM (view original):

Posted by schwarze on 8/13/2019 7:57:00 PM (view original):
You have to assign leagues ahead of time, in order to get people's rosters entered. The random division alignment is the fairest way to allocate within the league.

Oh, and for the mathematically challenged... anybody seeded in the top 48 is a favorite to have an easier than average division. Why? Not because the tournament is rigged. It's because that's how the math works. The better the seeding, the more likely that person will have an easier division because their seeding number isn't included in the measurement of *his* division strength.

I think we've wasted enough time trying to help hurst47 understand this. He clearly chooses not to get it.

Oh, I get it. The double randomization process the way its done skews the results to give the higher ranked owners an easier draw. I have yet to see anything to convince me otherwise. You're bordering on a scam if you fail to realize this. You top dogs who pay for this thing with a few measly site credits from your obscenely large stash just don't understand the huge advantage you already have just by being experts in how WIS configures their simulation algorithm and through repetition have identified all the undervalued outliers within the system. I guess in light of all this maybe you should issue a warning to the uninformed masses that this is a tournament for experts only and set a minimum owner rating for participation otherwise the more casual participant's tournament is over before it even begins and we're unknowingly scammed out 75 bucks for nothing.

I do feel like the evidence has been laid out fairly thoroughly and while I am sorry if it hasn’t been, I’m not sure what else I can show you.

However, there is no “double randomization process” so I do wonder if you aren’t misreading something.

Hurst47, if you still have concerns I’d encourage you to sitemail me directly. So far you have mostly been responding to other people on this thread.