Rare outcomes from dice rolls

@axis_roll +1

@blister911:

We use a mix of dice colors (white, red, black, dark blue, light blue, green, purple, orange, and grey).  Each one designated for its hit value, special attack, or unit differentiation (subs, super subs, shore bombardment, SBR, etc).  That way we can roll the entire attack or defense at once.  It gets interesting when rolling 15+ dice all at the same time.

That sounds pretty cool I’ll have to try that.

I think someone said something about rolling each set twice.  Anyway, here is another idea.

Any time there are more than X dice being rolled (you decide X for yourself, but I would say at least 4 dice [you can’t average 1 roll, and it would get absurd if you roll twice for 2 dice, perhaps absurd for 3]), then the person rolling them rolls them two times, then takes the Average number of hits between the two rolls.  In case of decimal, round in the direction of the first roll.
For example, I am attacking with 3 inf and 3 tanks against you with 3 inf.  I would roll twice, lets say the first roll I get 2 hits, and the second roll I get 5 hits.  The average of this would be 7/2= 3.5 hits.  Because the first roll was lower, you round down, so that is 3 hits now.  You would not get to roll twice because you are only rolling 3 dice. 
Or another example, I attack with 3 inf 2 art, against you with 3 inf.  I roll twice and get 2 hits the first time, and 1 hit the second time.  The average is 1.5, but you round up because the first roll was higher.  You again would not roll twice because you have less than X=4 dice to roll.  You get lucky and roll 3 hits.  Now it is 2 art against 1 inf.  Neither of us would roll twice because we are now both less than 4.

Or another way to do this, would be to use a certain number of “roll twice for an average” chips / tokens.  Lets say you get 5 total for each team, and are allowed to purchase more for 4 ipcs each.  You may use one of them up to do exactly what I just described above.

For f2f games, you need professional equipment, if better skills and experience is not enough, you need to be sure.

It’s even better than low luck  :-D

http://cardshark.us/museum_frs.shtml

Check out the dice cheat section.

Maybe I may do an error, because I do not play by forum but I have read as the forum dice roller works. The number after the @ indicates the attack value of the dice. Hypothetically the hit value should not influence the result of the “dice rolled”, which is always a d6. The problem is that rolling a d6 with an attack value of 13 it is impossible to have a hit confirmation from the dice roller. It is still possible to asses the hits with “human intervention”.

In such situation, the, the 2@13 are two d6 rolled for the two infantry and should be considered ok. If one of them score a a 1, it should be a hit. So the only thing to do is to roll for the 3@3, for having the fighters outcome.

As I said, however, I am not an expert of forum dice rolling, so a more qualified opinion is needed!  :-D

Ok to refute the argument on LL is less complex, um, there about the same.

With dice you do what you do praying the dice don’t take a dump on you, sending in more units than needed is still wasteful because the dice could just take a larger dump on you.  With LL I have to be more aware of strafes, as well as cascading dice failure is a huge issue in LL, something i regularly expose my opponents too.  In LL I have true risk management when i send either a DD and a fig at that sub, or a DD and a bomber.  Or 2 inf and a fig to trade, or 2 inf and a bomber against that 1 inf.  2 inf + 1 bomb vs 2 inf in LL can turn into me loosing both inf, I still have your ‘risk management’ to deal with.

With dice I just hope for the best, I have no true control over what the dice will do.  In LL i have to choose to send the extra units to ensure the result, or not to.  I cannot begin to tell you the number of battles that come down to that AA gun hitting, but they do exist.  Or the time when my DD+fig missed that sub and it hit, then sunk a tranny next turn.

They are different styles, and neither is the ‘correct’ one, how about we all stop arguing over it.  Its like watching a political debate.

@shifty:

What kind of “program” are you writing?  What equation/algorithm are you using?

It’s written in BlitzMax. The source below should (hopefully :)) not be too hard to understand - if you have questions, feel free to ask:

' Compares 2nd ed. A&A tech dev cost/rate with Anniv. ed. A&A ' 20 turns to test; record turn number that a total of 2 breakthroughs were achieved.  Test 1 & 2 rolls/turn. ' If two breakthroughs aren't achieved, "turn 0" is recorded (ignored for graphing & turn average) ' (When going for two breakthroughs, two aren't achieved in about 2% for both 2nd and Anniv.) Strict Local breakthroughs, totalCost, turn, totalTurns Local answer[21], prefix$, graph$ Local breakthroughLimit = 2 For Local rolls = 1 To 2 ' actual rolls for 2nd Ed, Researchers for Anniv. Print;Print; Print rolls + " Tech Roll(s)/Researcher(s) per Turn:" Print; Print "2nd Edition" totalCost = 0 totalTurns = 0 For Local c = 0 To 20 answer[c] = 0 Next For Local test = 1 To 100000 ' a hundred thousand tests for accuracy breakthroughs = 0 turn = 0 While breakthroughs < breakthroughLimit And turn <= 20 turn :+ 1 totalCost :+ rolls * 5 ' each roll costs 5 IPCs For Local c = 1 To rolls If Rand(6) = 6 Then breakthroughs :+ 1 Next Wend totalTurns :+ turn If turn > 20 Then answer[0] :+ 1 Else answer[turn] :+ 1 EndIf Next ' display results For Local c = 1 To 20 graph = "" ' divide results by 1000 so each asterisk represents one percent For Local cc = 1 To Int(answer[c]/1000.0 + .5) graph :+"*" Next If c < 10 Then prefix = " " Else prefix ="" ' extra space for single digits Print prefix + c + " " + graph Next Print "Avg turn = " + totalTurns / 100000.0 Print "Avg cost = " + totalCost / 100000.0 Print; Print "Anniversary Edition" totalCost = 0 totalTurns = 0 For Local c = 0 To 20 answer[c] = 0 Next Local researchers ' researchers accumulate from turn to turn For Local test = 1 To 100000 researchers = 0 breakthroughs = 0 turn = 0 While breakthroughs < breakthroughLimit And turn <= 20 turn :+ 1 totalCost :+ rolls * 5 ' each researcher costs 5 IPCs researchers :+ 1 * rolls ' accumulate 1 researcher per "roll" Local success For Local c = 1 To researchers If Rand(6) = 6 Then success = True ' only one regardless of # of sixes rolled Next If success Then researchers = 0 ' eliminate accumulated researchers when successful breakthroughs :+ 1 EndIf Wend totalTurns :+ turn If turn > 20 Then answer[0] :+ 1 Else answer[turn] :+ 1 EndIf Next ' display results For Local c = 1 To 20 graph = "" ' divide results by 1000 so each asterisk represents one percent For Local cc = 1 To Int(answer[c]/1000.0 + .5) graph :+"*" Next If c < 10 Then prefix = " " Else prefix ="" ' extra space for single digits Print prefix + c + " " + graph Next Print "Avg turn = " + totalTurns / 100000.0 Print "Avg cost = " + totalCost / 100000.0 Next

I release it into the public domain (modify it as desired/convert it to the language of your choice).  breakthroughLimit is initially set for 2, which generated the first set of graphs in this thread (I hand-rounded the accuracy to a single decimal place when copy/pasting results in the thread). Setting it to 1 (i.e. stopping after getting one breakthrough) produced the second set above. I also made another, slightly tweaked version to generate the “only buy Researchers on turn 1” version for wodan46.

Just roll them all at once.  As long as you use different colors for different attack powers (ie all 1’s roll white, all 2’s roll red, all 3’s roll blue and all 4’s roll yellow or something) or you just roll for each attack power separately, you are fine.  No need to roll each individual bomber since all the hits total up at the end anyway.