League Rule Changes for 2014 AAG40 2.0

Domo arigato, Mr. Roboto

I believe Jenn and I agree that land bids will continue to be allowed on empty territories, while naval bids will require a unit present

I am in favor of 1 unit/territory/sea zone because I don’t believe bidding is supposed to be a strategic way to win the game.  I don’t think you should be able to surprise your opponent with a 15 bid by plopping 5 infantry on France or Yunnan or something.
I wouldn’t mandate 1 unit/area, but I think it should be set as the league default.  Like unless the players agree to something different.

The whole purpose of bidding is to make both players happy with their side and starting setup.

I’ll throw in my 2 cents to give the perspective of a new league player.

I agree that there should be a well-defined set of “default” bidding rules, but nothing needs to be disallowed completely. This allows us to bid in shorthand (“Allies +12”) and know what that means, but also be able to agree on other bidding rules if desired.

The current ranking system works, so I don’t see any reason to invest a lot of time coming up with something that (at best) would only be incrementally better. If anyone is seriously unhappy with the current system, there’s nothing stopping them from creating a new one and plugging all of the data into it. The majority of the discussion of alternate systems is speculation, so some hard evidence would help the cause of anyone promoting a change.

One suggestion I have for Gamerman is to just add a small “legend” table with the points awarded for wins/losses against the various tiers on the spreadsheet. Right now if you don’t know what they are, you either have to dig through a discussion thread or reverse engineer the awarded points to figure it out.

Low luck is kind of a tricky one, because it really depends on how different you consider the two games. Obviously if the games really are as different as football and soccer, it doesn’t make sense to mix the two in the same league, but this feels like a stretch to me. I’ve only experimented a little with LL, but I’d presume that the overall “game IQ” and tactics are similar, it’s really just the math of grinding the odds that is changing. So maybe it’s more akin to American football vs. Canadian football, to further abuse the professional sports analogy :)

Anyway, since it’s all speculation right now with no games actually played, I’d rather we just allow it initially and come back to it later if it starts becoming a “problem” in some way. Obviously it needs to be made clear that dice is the default, will likely be what is played in the playoffs, etc. If people are worried that a good LL player will rise to the top of the rankings without playing dice, there are ways of mitigating that - limits on what % of games may be LL, adding a ladder-type challenge system to the league so you can challenge players above you in the rankings, etc. But I really think we shouldn’t assume there will be problems with it until it actually happens.

Someone mentioned triplea bugs. Are there any bugs that are particularly annoying for the league? I’m not familiar with the code base but I do Java programming for a living, so I might be able to submit a patch or two.

Finally, thanks to everyone putting in the time to run the league, it’s very appreciated!

Thanks for the thoughtful input, nielsen
Yes there are a slew of Triple A bugs

Here is the link to my list
https://docs.google.com/document/d/1cS3hFwoBP0rWr3208jw1LAtCvHYa9Ji9uU8PwbCMIwM/edit?pli=1

Many of them are important.  We should include a link to this list with the league rules post at the beginning of the league year so that everyone reading the league rules will also be alerted to all of these issues.

should a loss against a tier 1 be worth 3 points (same as win against a tier 4?) and loss against tier 2 worth 2, loss against tier 3 worth 1, and loss against tier 4 worth 0?

triplea is not rigged, the numbers(dice) are random, it is just how we perceive luck.

I strongly disagree here.  Triplea, as great as it is, has a flawed random dice generator.  Yes over a decent number of rolls, you will get “random” like results with numbers rolled totaling about the same.  As a result, the statistics reported will appear as if everything is within the standard deviation.

However within those rolls, you will see sets of numbers rolled fairly consistently.  It could be all lows, or all highs rolled for say 10 dice being thrown.  Often it’s even the same number rolled.  Tell me the probability of 7 straight 1s being rolled?  I haven’t played many games and its not very often 7-10 dice are thrown in one roll, but I know that’s not statistically probable.  More over, I’ve seen these results repeat over and over again.

This is why you see some people complaining AA guys seem to hit too often.  Well on “average” they only hit the average throughout a game, for like 20-30 rolls.  But when they hit, they hit in bunches.  I’ve seen snake eyes or triple 1s way too often for this to be coincidence.  Of course this also means they can all miss even vs a dozen air.

Hopefully these sequences affect both players equally but it can throw off a lot of planning when it occurs in critical situations.

If people keep this in mind and pay attention to the actual dice thrown instead of the results, you’re probably see this too.

@Cmdr:

As for technology, in those games using technology (i.e. both players agree to use technology in their league game) should we allow technology tokens to be the default rule?

I think no tokens should be the default, because no tokens is what is in the rulebook.

@Boldfresh:

should a loss against a tier 1 be worth 3 points (same as win against a tier 4?) and loss against tier 2 worth 2, loss against tier 3 worth 1, and loss against tier 4 worth 0?

No, I think that’s too many points for losing

You need a lot of data and tests to back up those allegations/perceptions, Hobo

I do pay attention to individual dice and not just results - always - and I haven’t noticed “strings of repetitive numbers”
I did a little research a while back and learned that repetitive results are commonly perceived by people to be non-random, when that is not necessarily the case.

I don’t know how good of a job these dice servers do at approximating real dice, but it seems pretty good to me.
I seem to recall about 25 years ago reading that it is impossible for computers to create true randomness.  I’m sure this is oversimplifying it, but it has something to do with needing a “seed” to start.

If you’re going to point out a perceived problem, at least provide a suggested solution, else you’re just eroding faith in the dice server and that does not benefit anyone unless you have a better alternative.

Perhaps we should only roll dice live when the other player is there, and show it on a live video feed?

:lol: About the live video feed…

I prefer imperfect computer RNG’s to debating with Boldfresh about what is and is not a “legal” roll…  How much they need to be spun, “trick” rolling, landing on a corner, missing the box, and on and on…

Let us know if you find a better dice server that can be linked to Triple A.
Or provide reliable data/testing/results/conclusions using a proper scientific method if you’re going to allege that the dice are not random and spew out improper strings of repetitive numbers.

Everyone knows that if a dice is on greater than 0.45% tilt due to coming to rest partially on an object that the roll is not valid!  :-P

Dice on shaggy carpet?  Not valid.  Dice coming to rest atop other dice…. clearly not valid.

if I recall we solve this dilemma by rolling into a box. Simple enough :-D

remember when I rolled 57 twos and missed every single one of them against you with real dice?  What are the chances of that hobo?

Well the technology is designed totally wrong in my opinion.

At the current state, it is way too weak. USA is the only power who could at least try researching at all, and even that isn’t really cost effective.

In my opinion even with tokens, technology isn’t too strong. Maybe even still too weak.

So I am with Jennifer here. We barely have any tech games at the moment. With tokens, these might happen more often.

@Gamerman01:

You need a lot of data and tests to back up those allegations/perceptions, Hobo

I do pay attention to individual dice and not just results - always - and I haven’t noticed “strings of repetitive numbers”
I did a little research a while back and learned that repetitive results are commonly perceived by people to be non-random, when that is not necessarily the case.

I don’t know how good of a job these dice servers do at approximating real dice, but it seems pretty good to me.
I seem to recall about 25 years ago reading that it is impossible for computers to create true randomness.  I’m sure this is oversimplifying it, but it has something to do with needing a “seed” to start.

If you’re going to point out a perceived problem, at least provide a suggested solution, else you’re just eroding faith in the dice server and that does not benefit anyone unless you have a better alternative.

Perhaps we should only roll dice live when the other player is there, and show it on a live video feed?

The dice generator for forum dicing is pretty good.  I’ve played a lot more games using forum dicing than triplea, and never questioned the dice generator.

But after about 3 games of triplea, it was clearly apparent there was an issue.  How to fix it?  I can’t tell you without seeing how the program generates its random numbers.  Maybe changing the seed would do it.  Maybe linking the seed to a real time clock for each separate individual roll instead of rolling the dice for a particular round with the same seed.  You get the picture.

Like I said throughout a game, things should “average” out despite the dicing coming in clusters, but for any given battle it can change the results noticeably.

I didn’t even think of that last night….

So you could roll all of your dice on the forum and edit for the results.  Good luck finding opponents who are willing to comply…  Maybe Karl will  :-)

@Gamerman01:

I think no tokens should be the default, because no tokens is what is in the rulebook.

I knew I should have underlined certain words so I wouldn’t be misunderstood.

Tech tokens will continue to be allowed in league games.
No tokens being the DEFAULT means that if not specified, you are not playing with them.

Example:
“Hey Karl, wanna play a league game with tech?”
“Sure!”

If nothing else is said, this game will be without tokens.  The rule book should be the default.  If you want to play with tokens, go right ahead, but you’ll have to specify that and have mutual agreement.  Triple A allows you to play with tokens, just have the appropriate option check box ticked or unticked

@hobo:

@Gamerman01:

You need a lot of data and tests to back up those allegations/perceptions, Hobo

I do pay attention to individual dice and not just results - always - and I haven’t noticed “strings of repetitive numbers”
I did a little research a while back and learned that repetitive results are commonly perceived by people to be non-random, when that is not necessarily the case.

I don’t know how good of a job these dice servers do at approximating real dice, but it seems pretty good to me.
I seem to recall about 25 years ago reading that it is impossible for computers to create true randomness.� I’m sure this is oversimplifying it, but it has something to do with needing a “seed” to start.

If you’re going to point out a perceived problem, at least provide a suggested solution, else you’re just eroding faith in the dice server and that does not benefit anyone unless you have a better alternative.

Perhaps we should only roll dice live when the other player is there, and show it on a live video feed?

The dice generator for forum dicing is pretty good.� I’ve played a lot more games using forum dicing than triplea, and never questioned the dice generator.

But after about 3 games of triplea, it was clearly apparent there was an issue.� How to fix it?� I can’t tell you without seeing how the program generates its random numbers.� Maybe changing the seed would do it.� Maybe linking the seed to a real time clock for each separate individual roll instead of rolling the dice for a particular round with the same seed.� You get the picture.

Like I said throughout a game, things should “average” out despite the dicing coming in clusters, but for any given battle it can change the results noticeably.

Regarding the dice. In my first game I played against MrRoboto. I attacked a minor IC twice in a row and both rounds a 1 was rolled, which means my bombers are gone. Now fortunately that game was already lost to me a long time before that. However bad luck happens. You can re-roll as much as you want until you get satisfying casualties, but at that point you might as well not roll at all and simply assume what is going to die and what is going to live.

I have played tech games without tokens and I have hit several techs.  I disagree that only America should buy tech when there are no tokens, but this isn’t the place to go into it.  It’s been discussed exhaustively in other threads before, and does not need to be discussed here, because tech is allowed, tech tokens are allowed, there’s nothing to discuss about league rule changes in 2014 regarding tech.

Triple A is not harsh on bombers when it comes to SBR.  I have seen many, many misses, usually when my opponent is flying over me.  I think I get less than 1/6 on AA hits, so I am balancing you out, Jenn.  Small sample sizes.  Perception.  Etc.