Conquer Club

One should know that it's very hard to make a perfect random number generator, almost all have some hidden inconsistency. In fact, they sell books full of random numbers for $100 each.
http://www.rand.org/pubs/monograph_reports/MR1418/

danfrank wrote:as you state here OLD.. The new intensity cubes are not the old dice..

For everything related to the discussion in this thread they are the same though. And hence the "old" Dice Analyzer is also the "new" intensity cubes analyzer, just without the name change.

14 armies to conquer

2009-06-16 15:33:04 - RADAGA assaulted F4 from F1 and conquered it from NightReves
2009-06-16 15:33:08 - RADAGA assaulted F5 from F4 and conquered it from NightReves
2009-06-16 15:33:13 - RADAGA assaulted F6 from F5 and conquered it from NightReves
2009-06-16 15:33:18 - RADAGA assaulted F3 from F6 and conquered it from NightReves

All defended by 1 army each.

this is simply RIDICULOUS!

I dare ANYONE prove to me that this is a commonplace with real dice: To lose 11 3x1 to win 4

15 rolls, 4 won, 11 lost...

73,3% for the defence
24,7% for the attacker.

I don't have time to read this entire long thread (dating from 2006!), so apologies if this link has been referenced already. But everyone should read the following website, in full:

Dice Odds in the Board Game Risk

Simply put, the odds are not as great as some of you seem to think they should be. It seems to me that most people fail to recognize the sheer power of the defender's ability to win ties.

For example, the very best odds an attacker can have are 3 dice versus 1 die. Note: that is not 3v1, it's 4v1 or better! 3v1 is 2 dice versus 1 die, and I've seen this mistaken notion referenced many times on these boards already. Anyway, back to the example. 3 dice versus 1 die (4v1 or better) is only a 66% win chance for the attacker. That means, on average, you'll win about 6 or 7 out of every 10 rolls against 1 defender. That's pretty good, but it's far from a sure thing.

And that's only on average. In real life, random doesn't conform to average, it follows a bell curve, which means that sometimes you'll get "average" results, but you'll also get extremes at either end, including well above average and well below average. Play enough and roll enough dice, and you too will experience the 30v3 loss - the more you play, the more you approach certainty of experiencing such an event. Those of you who have played thousands of games should not be surprised by this at all.

There is also a lot of selective memory going on here. You tend to remember the bad rolls - they really suck and they stick out in your mind. You tend to forget all the average rolls, because they're simply not notable. And you might remember the great rolls, but probably not as well as you remember the pain of the bad rolls. So, there's that.

Note that 2 dice versus 1 die is very nearly 50/50, and 1 die versus 1 die is in favor of the defender. So all those 3v1 and 2v1 battles you keep losing? Don't complain - if you knew the odds were even or against you, maybe you'd choose to fight those less often.

And 3v3? 2 dice versus 2 dice? That's suicide, almost a 50% chance for the defender to win both rolls, and another 30% to win one. Not a recommended attack.

The most common situation is multiple armies versus multiple armies, so 3 dice versus 2 dice. The average stats? Only a 37% chance the attacker will win both rolls. The defender has nearly that good a chance of winning both himself, almost 30%. And there's a 33% chance you'll each win one. That means on 2 out of every 3 rolls, on average, the attacker will lose at least one army. A good rule of thumb, if you want to approach guaranteed victory in any conquest, is to take 3 times as many men as what you're attacking. Even that's no guarantee, but it hedges the odds about as much as possible. Don't expect to win many 12v12 battles - the cumulative odds of taking such battles are not much better than 50/50 (actually a little worse once you take into account what happens when you're reduced to 3 men and only 2 attacking dice, and then 2 men and only 1 attacking die).

So it just seems to me that people are expecting too much from the dice if they're the attacker. Perhaps they've forgotten what it's like to play real Risk, where the same kinds of crappy rolls happened all the time (more selective memory, me thinks). And the defender's ability to win all ties is very, very powerful - it can change games. If you don't know all these odds already, I suggest you learn them by heart and play accordingly. For your convenience, I'll repost the chart from the above-referenced website right here:

Attacker: one die; Defender: one die:
Attacker wins 15 out of 36 (41.67 %)
Defender wins 21 out of 36 (58.33 %)

Attacker: two dice; Defender: one die:
Attacker wins 125 out of 216 (57.87 %)
Defender wins 91 out of 216 (42.13 %)

Attacker: three dice; Defender: one die:
Attacker wins 855 out of 1296 (65.97 %)
Defender wins 441 out of 1296 (34.03 %)

Attacker: one die; Defender: two dice:
Attacker wins 55 out of 216 (25.46 %)
Defender wins 161 out of 216 (74.54 %)

Attacker: two dice; Defender: two dice:
Attacker wins both: 295 out of 1296 (22.76 %)
Defender wins both: 581 out of 1296 (44.83 %)
Both win one: 420 out of 1296 (32.41 %)

Attacker: three dice; Defender: two dice:
Attacker wins both: 2890 out of 7776 (37.17 %)
Defender wins both: 2275 out of 7776 (29.26 %)
Both win one: 2611 out of 7776 (33.58 %)

Memorize this, it will help, I promise. Knowing the odds by heart has won me games, because I know which attacks are worth attempting and which ones are best avoided, and not knowing the odds has lost games for some of my opponents, as I've watched them sometimes make utterly ill-advised attacks and then act surprised when they lose. And read the website, it's very illuminating. (It's not my website, so this isn't a shameless plug.)

I've taken notes on over 100 games I've played in, and the dice I've rolled approximately conform to these averages. I would expect them to get even closer as I keep playing, and I've experienced a 25v4 loss as well as a 10v30 defensive win. I've read of other people doing similar analysis and coming to the same conclusion. My understanding is that CC's numbers come from random.org, which is about as truly random as you can get. (And for crying out loud, don't listen to Klobber - he simply doesn't know what he's talking about and needs to take some remedial math classes.)

So my recommendation: stop complaining and play the game. The dice are random - really!

RADAGA wrote:14 armies to conquer

2009-06-16 15:33:04 - RADAGA assaulted F4 from F1 and conquered it from NightReves
2009-06-16 15:33:08 - RADAGA assaulted F5 from F4 and conquered it from NightReves
2009-06-16 15:33:13 - RADAGA assaulted F6 from F5 and conquered it from NightReves
2009-06-16 15:33:18 - RADAGA assaulted F3 from F6 and conquered it from NightReves

All defended by 1 army each.

this is simply RIDICULOUS!

I dare ANYONE prove to me that this is a commonplace with real dice: To lose 11 3x1 to win 4

15 rolls, 4 won, 11 lost...

73,3% for the defence
24,7% for the attacker.

Is it commonplace? No, but it's not exactly rare either - on average, it'll happen 1 out of every 4 times. If you took a 4-sided die and rolled it 100 times, you'd expect to roll a 1 about 25% of the time. Would you really be that surprised if you rolled 3 1s in a row at any point? Would you even be surprised if you rolled 3 1s in a row, then rolled a 4, then 3 1s in a row again? How many times have you rolled 3v1 dice in CC? Are you really that surprised that, occasionally, you'll have bad streaks? See my post above - it's not a curse, it's just a bell curve. You just experienced the low extreme - welcome to the club. I promise it will happen again at some point.

Halmir wrote:My proposal: Forget bitching about it in the forums, very little actions results. Instead, one Admin is needed to lead a committee/private discussion forum (i.e. without 300 posts clogging it up)/call it what you will, of interested players to work out an alternative or range of alternatives. I'd volunteer to help out on that if you want people, and in fact I'm sure there'd be no shortage. There is still a need for the current dice option, but we could do with others (defined kill rate when attacking, mix of dice and defined kill rate, whatever). Then add the outcome(s) as a public poll and add the most popular as one or more tickable options for people creating the games.

I hope you're joking. An alternative to what? Random dice, like it's supposed to be? What possible alternative would anyone who is not a total wuss want to play? Defined kill rate - seriously? That would just make the game a fancy version of rock/paper/scissors.

The game is Risk with different maps and sometimes alternate map rules. But the game is still Risk. Risk involves dice, and dice involve luck - that's why it's called "Risk"! Risk without random dice is not Risk, it's something else entirely. Think about it - let's call a "defined kill rate" what it is, which is a guaranteed kill rate. Not that there's necessarily anything wrong with that - the Advance Wars series on Nintendo handhelds is a gleaming example of such a genre - but it's an entirely different game that requires a diverse set of unique unit abilities to be any fun. In Risk, all units are the same. I would certainly welcome an Advance Wars-like option to play in Conquer Club, but it would require massive amounts of programming, because you couldn't just plug in a defined kill rate system and leave everything else the same - it wouldn't work in terms of game design. Games would be predictable and dull, since every battle would have identical results with no chance to take advantage and change the results by pitting superior units against inferior units, which is how Advance Wars is played.

And who wants an alternative to random dice anyway? Not to be too harsh, but my take is this: if you can't take the stress of an uncertain outcome, then Risk nee Conquer Club is not the game for you. Period.

Anyway, not trying to bash you. I'm just incredulous that disillusion with the dice would lead to calling for wholesale changes to a gaming system that already works exactly as intended. There is no merit in such a suggestion, and I would be opposed to any changes of that nature. The dice are the cornerstone of this game - the uncertain outcomes are what make the games interesting and different every time. Even the best players can lose due to crappy rolls. That's how the game was designed; that's how it's supposed to be.

(By the way, if you don't know what Advance Wars is, I highly recommend it to all Risk fans - it's like Risk on steroids. I really would welcome such an option here, but it would require programming an entirely new game, so I wouldn't expect it, and I'd expect to pay a lot more to play if it were available.)

RADAGA wrote:More info to be ignored:

The dice distribuition is even with 20.000 rolls. One can almost believe they are truly random.

1s █████████████████████████ 3347 / 20204 (16.57%) █████████████████████████ 1838 / 10839 (16.96%)
2s █████████████████████████ 3355 / 20204 (16.61%) █████████████████████████ 1783 / 10839 (16.45%)
3s █████████████████████████ 3425 / 20204 (16.95%) █████████████████████████ 1806 / 10839 (16.66%)
4s █████████████████████████ 3354 / 20204 (16.6%) █████████████████████████ 1779 / 10839 (16.41%)
5s █████████████████████████ 3411 / 20204 (16.88%) █████████████████████████ 1835 / 10839 (16.93%)
6s █████████████████████████ 3312 / 20204 (16.39%) █████████████████████████ 1798 / 10839 (16.59%)

BUT then you see that the defence still get more victories than it should. both in 3x2 and in 3x1.

Battle Outcomes Actual Stats Ideal Stats
3v2 █████████████████████████ 1424 / 1239 / 1114 (37.7% / 32.8% / 29.49%) (37.17% / 33.58% / 29.26%)
3v1 █████████████████████████ 1753 / 962 (64.57% / 35.43%) (65.97% / 34.03%)

That´s yet another strong evidence that the dice come in streaks.

How so? These numbers conform almost exactly to the Risk dice odds I posted above. See the Risk dice odds website I reference in that post for more.

I'm sorry, but your complaints are baseless. Play some real life board game Risk, and you will get the same results. Sometimes you get the bear, and sometimes the bear gets you - Lady Luck is truly a harsh mistress.

yeah, right. on real life Risk you go with a stack of 14 armies against 4 defended-by-one territories and you will end up without any army left 25% of the time.

you will lose 11 armies to conquer 4 singles, one every four tries... thats what I have read here, at least...

mkohary wrote:

Halmir wrote:My proposal: Forget bitching about it in the forums, very little actions results. Instead, one Admin is needed to lead a committee/private discussion forum (i.e. without 300 posts clogging it up)/call it what you will, of interested players to work out an alternative or range of alternatives. I'd volunteer to help out on that if you want people, and in fact I'm sure there'd be no shortage. There is still a need for the current dice option, but we could do with others (defined kill rate when attacking, mix of dice and defined kill rate, whatever). Then add the outcome(s) as a public poll and add the most popular as one or more tickable options for people creating the games.

I hope you're joking. An alternative to what? Random dice, like it's supposed to be? What possible alternative would anyone who is not a total wuss want to play? Defined kill rate - seriously? That would just make the game a fancy version of rock/paper/scissors.

The game is Risk with different maps and sometimes alternate map rules. But the game is still Risk. Risk involves dice, and dice involve luck - that's why it's called "Risk"! Risk without random dice is not Risk, it's something else entirely. Think about it - let's call a "defined kill rate" what it is, which is a guaranteed kill rate. Not that there's necessarily anything wrong with that - the Advance Wars series on Nintendo handhelds is a gleaming example of such a genre - but it's an entirely different game that requires a diverse set of unique unit abilities to be any fun. In Risk, all units are the same. I would certainly welcome an Advance Wars-like option to play in Conquer Club, but it would require massive amounts of programming, because you couldn't just plug in a defined kill rate system and leave everything else the same - it wouldn't work in terms of game design. Games would be predictable and dull, since every battle would have identical results with no chance to take advantage and change the results by pitting superior units against inferior units, which is how Advance Wars is played.

And who wants an alternative to random dice anyway? Not to be too harsh, but my take is this: if you can't take the stress of an uncertain outcome, then Risk nee Conquer Club is not the game for you. Period.

Anyway, not trying to bash you. I'm just incredulous that disillusion with the dice would lead to calling for wholesale changes to a gaming system that already works exactly as intended. There is no merit in such a suggestion, and I would be opposed to any changes of that nature. The dice are the cornerstone of this game - the uncertain outcomes are what make the games interesting and different every time. Even the best players can lose due to crappy rolls. That's how the game was designed; that's how it's supposed to be.

(By the way, if you don't know what Advance Wars is, I highly recommend it to all Risk fans - it's like Risk on steroids. I really would welcome such an option here, but it would require programming an entirely new game, so I wouldn't expect it, and I'd expect to pay a lot more to play if it were available.)

OK I'll accept that you're not trying to bash me (despite what I just read lol!). But in my couple of years on this site, it does seem to be the biggest bugbear that causes most complaints. The fix is simple, just add a game type that doesn't use the dice in the same way or a "Rock/paper/scissors" if u will. It doesn't have to displace the dice, and in fact I'd play both as I'd welcome the extra options. OK u feel u have the advantage with ur knowledge of dice stats, how lovely for u, however it's like moaning about a Fog of War gametype (and I'm sure people did before it was introduced). If there's enough interest then build it in and more people are happier, thus sustaining the membership.

This is the view I'd take to such an ad hoc forum tasked with reviewing the options - and it sounds like you'd be a useful member also. However, I have my doubts that thrashing it out here would have any point, I can't say I've seen much result from topics raised on Suggs and bugs, with thousands of people viewing then many diving in to reply it gets too busy and uncontrolled... hence my idea to get a group with an admin and thrash it out separately.

Hey ho.

Halmir wrote:OK I'll accept that you're not trying to bash me (despite what I just read lol!). But in my couple of years on this site, it does seem to be the biggest bugbear that causes most complaints. The fix is simple, just add a game type that doesn't use the dice in the same way or a "Rock/paper/scissors" if u will. It doesn't have to displace the dice, and in fact I'd play both as I'd welcome the extra options. OK u feel u have the advantage with ur knowledge of dice stats, how lovely for u, however it's like moaning about a Fog of War gametype (and I'm sure people did before it was introduced). If there's enough interest then build it in and more people are happier, thus sustaining the membership.

This is the view I'd take to such an ad hoc forum tasked with reviewing the options - and it sounds like you'd be a useful member also. However, I have my doubts that thrashing it out here would have any point, I can't say I've seen much result from topics raised on Suggs and bugs, with thousands of people viewing then many diving in to reply it gets too busy and uncontrolled... hence my idea to get a group with an admin and thrash it out separately.

Hey ho.

Sorry I was overly blunt in my first reply to you. I just think all the dice complaints sound so superstitious. Yeah, ok, I suppose they could implement some kind of guaranteed win rate as a new game type, and people who wanted to play that game type could, and those that didn't don't need to. (I just don't like the idea of mixing scores from that game type with scores from the normal game type.)

But even better would be if they simply put up a page detailing how the dice work, explaining that they ARE in fact random and why, and that complainers will be summarily ignored.

I just rolled the dice and I didn't win the roll. This site is rigged

mkohary wrote:I don't have time to read this entire long thread (dating from 2006!), so apologies if this link has been referenced already. But everyone should read the following website, in full:
Simply put, the odds are not as great as some of you seem to think they should be. It seems to me that most people fail to recognize the sheer power of the defender's ability to win ties.

Not sure that specific site has been referenced. All the numbers and odds and such though is common knowledge. Always good to have more people that knows the facts around though.

mkohary wrote:Note: that is not 3v1, it's 4v1 or better! 3v1 is 2 dice versus 1 die, and I've seen this mistaken notion referenced many times on these boards already. Anyway, back to the example. 3 dice versus 1 die (4v1 or better) is only a 66% win chance for the attacker.

Ah yes, usually we agree quite early in these threads that any dice that isn't rolled is irrelevant to the argument so a 3v1 battle really is a 3v1 battle just without the added "dice" after every number posted in a thread.

mkohary wrote:I've taken notes on over 100 games I've played in, and the dice I've rolled approximately conform to these averages.

This is a little add-on that does the counting for you, try it out (as anyone with dice complaints also should).

RADAGA wrote:I dare ANYONE prove to me that this is a commonplace with real dice: To lose 11 3x1 to win 4

I don't have to since it isn't commonplace here on CC either...

I don't have to since it isn't commonplace here on CC either...

then why it happened again, today?

deployed 13 just to lose 8 in a row for a 1.

Not common, happens once every 5 rounds or so, given I am a freemium.

RADAGA wrote:Not common, happens once every 5 rounds or so, given I am a freemium.

So you think that your freemium status affect your dice rolls???

Thezzaruz wrote:

RADAGA wrote:Not common, happens once every 5 rounds or so, given I am a freemium.

So you think that your freemium status affect your dice rolls???

Sounds like he should pay for premium eh

e_i_pi wrote:

Thezzaruz wrote:

RADAGA wrote:Not common, happens once every 5 rounds or so, given I am a freemium.

So you think that your freemium status affect your dice rolls???

Sounds like he should pay for premium eh

No, sounds like I should be more explicit.

the - HAPPENS EVERY 5 ROUNDS OR SO, admits I get 5 rounds a day, if taken with the fact that one occourence happened within 24 hour range from another. So, why 5 rounds? Simply because I am a freemium, so, usually, I get to play, in the average, 5 times every day. If I were not, I would not gte better rolls, but you could say: oh, you get those once every day because you play 100 speed games simultaneously at any given time.

Did I made myself clear? Or should I be more specific? I suppose I could draw it.

RADAGA wrote:I suppose I could draw it.

You could "draw" one with the Dice Analyzer.

Thezzaruz wrote:

RADAGA wrote:I suppose I could draw it.

You could "draw" one with the Dice Analyzer.

I can do that too. But I won't because this is a thread for complaining about the dice.

Thezzaruz wrote:

RADAGA wrote:I suppose I could draw it.

You could "draw" one with the Dice Analyzer.

Cute. Someone like to bite and try to make people pass as stupid, but when someone gets bitten, suddently it is not that funny anymore.

Don´t worry, this happens in all famlies.

RADAGA wrote:Someone like to bite and try to make people pass as stupid

Never said you were stupid. But I do think you need to prove your statements a bit better than you have.

Just for reference....

For every 3 vs 2 attack, 5400 defenders should die for every 4600 attackers that die. (Java computer program)

I've also heard that 3 vs 1 ratio is about 76%, but haven't tested that one.

mako007 wrote:For every 3 vs 2 attack, 5400 defenders should die for every 4600 attackers that die. (Java computer program)

Or 37.17%/33.58%/29.26% to be a bit more exact. And it's 65.97% to win a 3v1 battle.
We know the odds, that's not the discussion.

Thezzaruz wrote:

mako007 wrote:For every 3 vs 2 attack, 5400 defenders should die for every 4600 attackers that die. (Java computer program)

Or 37.17%/33.58%/29.26% to be a bit more exact. And it's 65.97% to win a 3v1 battle.
We know the odds, that's not the discussion.

My beef here is not about statistics, they are fine, just as I stated several times ago.

The main problem I have been observing for a long time is that when you lose a 3x1, odss are y ou will lose another 2 or 3 before it wields.

Those who play here turn a blind eye, but they /know/ that the autoattack button is a certain demise to use. One of them even stated: "I have noticed when you can spare the armies and fac a lost battle (in the case a 9 versus 20) the best chance you have is to autoattack. Sometimes you will win and even still have armies to transfer." I have been using this myself, and yes, it works.

Get a pile of armies, hit autoattack and see. you will see fine statistics, but the wins and losses will most likely be groupped.

So the dice here are correct the the same way a 100 coins flipped landing 50 times on each face are consistent with a 50% ratio. But this coin in particular lands 25 times in a row on each side, twice.

RADAGA wrote:

I don't have to since it isn't commonplace here on CC either...

then why it happened again, today?

deployed 13 just to lose 8 in a row for a 1.

Not common, happens once every 5 rounds or so, given I am a freemium.

I'm sure your used to being insulted RADAGA (judging by your posts) and I really don't want to insult you now, but consider two propositions.

1. Your poor rank is based on a conspiracy to punish freemiums by giving them poor roles.

Or

2. You just aren't that good at the game.

Mr Changsha wrote:

RADAGA wrote:

I don't have to since it isn't commonplace here on CC either...

then why it happened again, today?

deployed 13 just to lose 8 in a row for a 1.

Not common, happens once every 5 rounds or so, given I am a freemium.

I'm sure your used to being insulted RADAGA (judging by your posts) and I really don't want to insult you now, but consider two propositions.

1. Your poor rank is based on a conspiracy to punish freemiums by giving them poor roles.

Or

2. You just aren't that good at the game.

I would accept it, if I have not observed the following: I have never changed the kind of games I play:

Freestyle, 3-6 players, flat rate and no spoils

For several MONTHS I usually win 30% of the games I play

then out of the blue I start to lose them ALL. there were a time where I got NOT A SINGLE VICTORY in 3 months in a row. That, just before winning 12 games on the previous month.

Considering the opponents are pretty much the same, that I dont change game type, and that whose win/losses are cyclical, so there is not "getting better" or "getting worse" at the game, I must assume there is something else affecting the overall win/loss ratio.

If I were not that good, how can you explain the medals and the 22% win rate, overall?

Yours are better because of the doubles and triples you play, rigged games with the same buddies, on the same maps, to trap newbies and uncoordinated opponents.

I challenge you for 20 1x1, random map, standard turns, no spoil games. Lets see how well you perform.