calling probability whizzes

[Rebar]

Mark's Astounding CyberEmpire O' Doom!

 

This gentleman (whom I have yet to contact) has an interesting alternate SPD system that my GM is considering testing for adoption:

 

SPD and actions

 

In a nutshell:

roll a D10 (or so)

on a 5, SPD 5 and up characters have an action

on a 1, SPD 1+ characters go (everyone)

etc.

 

Of note, - this system

- randomizes not only *when* everyone gets to go in relation to each other, it also adds an element of variance to the *number of times* everyone goes.

- can be made to cut back on "free" phases, thus I believe it could penalize high-SPD characters who stand to lose phases in which they could use held actions

 

But, to my question:

 

Is this system, though randomly affected, still balanced? Specifically, does it favour low-SPD chas vs. high-SPD chas or vice versa?

 

I know in the long-term it wil all balance out, but in the short-term, where random rolls have a bigger effect, I suspect that it may be more likely to benefit low-SPD characters. Can some math whiz iron this out?

[Jhamin]

I suspect this will overly reward people who have bought their speed up.

 

If I have a character with speed 4, I know I am slow but I also know that within most campaigns I will only go one or two times fewer than someone with a higher speed. Under this system I have a less than even chance of going at all on any given round.

 

This system will average out in the end, but as most champions battles last less than a turn you will only see this in a long game. In a single fight you have a very real chance that they will get 3 or 4 actions before you even get one, depending on how the dice roll.

 

That puts way more distance between the utility of a speed of 4 and 6 than currently exists. I also question the websites assertation that adding another die roll to each and every combat round is faster than just asking somebody to write down their phases & try to pay attention to what is going on.

[Netzilla]

Originally posted by Jhamin

f I have a character with speed 4, I know I am slow but I also know that within most campaigns I will only go one or two times fewer than someone with a higher speed. Under this system I have a less than even chance of going at all on any given round.

 

There's a variation on this system that eliminates the chance of losing any actions due to poor dice luck. Write the numbers 1 through 12 on scraps of paper and draw one out of a hat each phase. If your speed score is equal to or higher than the number, you get to act. Numbers do not go back into the hat until after all 12 have been used.

 

This also can work by using the number on the paper as the column of the Speed Chart that will be played that phase.

 

Deric Page

--

"There is no cause so right that one cannot find a fool following it." -- Larry Niven

[Mark Taylor]

Originally posted by Jhamin

If I have a character with speed 4, I know I am slow but I also know that within most campaigns I will only go one or two times fewer than someone with a higher speed. Under this system I have a less than even chance of going at all on any given round.

 

Each roll of the die should be considered a segment, not a round.

[Mark Taylor]

Re: calling probability whizzes

 

Originally posted by Rebar

Is this system, though randomly affected, still balanced? Specifically, does it favour low-SPD chas vs. high-SPD chas or vice versa?

 

Assuming the die rolled is a D12, the system as written up on Mark's site produces (on average) exactly the same balance as the existing SPD chart in HERO System.

[caris]

Re: Re: calling probability whizzes

 

Originally posted by Realms of Chaos

Assuming the die rolled is a D12, the system as written up on Mark's site produces (on average) exactly the same balance as the existing SPD chart in HERO System.

 

May, I quibble? While I would grant you that the system is probably equally as balanced, I wouldn’t go as far as to say “exactly the same balanceâ€. There are elements between a fixed and random system that always impact issues of balance, mostly in the area of predictability. Since it effects all players and characters equally, I would agree that it is equally as balanced, but how it affects players based on style is not the same. Which means that either choice can favor or penalize players in an uneven manner.

 

Going to Hero System’s fixed initiative from the steady diet of random initiatives that I had experienced prior, opened me up to whole new areas of tactical planning by removing a random factor. I had never really explored or attempted to do much with timing issues, because timing was always too far out of my control. On any given phase, a character could go at any point, and I had no real way of making a sufficiently accurate guess for me to feel comfortable planning around. Now that I had some better ability to predict when characters could go, I could plan around it. It still wasn’t a perfect ability, since characters can be played in such a way as to give false impressions, or simply hold action when you don’t expect them to.

 

Since Hero is currently the only system that anyone, that I game with, plays that uses a non-random initiative system, I would be very resistant to switching one for Hero. If for no other reason, I like the change of pace. The fact that the controlled flexibility favors my style of play in combat, I will admit is also a factor. I’m fairly good at exploiting the speed chart, I admit. So I favor it, as an option that let’s me get an advantage in an area where I’m usually handicapped in other games. Just as there are players in other games that I'm in that feel they are aided by the random nature of the initiative, and would resist any attempts by me to switch the games to a fixed initiative system.

[Mark Taylor]

Re: Re: Re: calling probability whizzes

 

Originally posted by caris

May, I quibble? While I would grant you that the system is probably equally as balanced, I wouldn’t go as far as to say “exactly the same balanceâ€. There are elements between a fixed and random system that always impact issues of balance, mostly in the area of predictability. Since it effects all players and characters equally, I would agree that it is equally as balanced, but how it affects players based on style is not the same. Which means that either choice can favor or penalize players in an uneven manner.

 

Going to Hero System’s fixed initiative from the steady diet of random initiatives that I had experienced prior, opened me up to whole new areas of tactical planning by removing a random factor. I had never really explored or attempted to do much with timing issues, because timing was always too far out of my control. On any given phase, a character could go at any point, and I had no real way of making a sufficiently accurate guess for me to feel comfortable planning around. Now that I had some better ability to predict when characters could go, I could plan around it. It still wasn’t a perfect ability, since characters can be played in such a way as to give false impressions, or simply hold action when you don’t expect them to.

 

Since Hero is currently the only system that anyone, that I game with, plays that uses a non-random initiative system, I would be very resistant to switching one for Hero. If for no other reason, I like the change of pace. The fact that the controlled flexibility favors my style of play in combat, I will admit is also a factor. I’m fairly good at exploiting the speed chart, I admit. So I favor it, as an option that let’s me get an advantage in an area where I’m usually handicapped in other games. Just as there are players in other games that I'm in that feel they are aided by the random nature of the initiative, and would resist any attempts by me to switch the games to a fixed initiative system.

 

The "exactly the same balance" I was talking about is in purely statistical terms i.e. a SPD 4 character will on average get 4 phases out of every 12 segments, a SPD 5 character will on average get 5, etc. If anybody's unsure why that's the case I'm quite happy to post the maths which is really pretty basic.

 

Read what it says on Mark's site. To randomise combat and take away that element of predictability and planning is the whole purpose of his system! In other words he's aware of the issue you pointed out, and counts it as one of the two major reasons he prefers his system.

 

Personally, as a GM, I wouldn't say for sure I will use it but I will give it a try. IMO there is a reason most initiative systems have a random element - combat is chaotic and unpredictable, and that random element helps simulate it's chaotic and unpredictable nature.

 

I've always felt that combat should as be random and unpredictable as possible. The first time I went paintballing I realised it was even more random and unpredictable than even I had previously thought. Most of the time I couldn't even pass a PER roll to see who the hell just shot me, let alone predict what they were going to do next or when they were going to do it. Even very experienced paintballers who go on a regular basis find the same to be true.

[Rebar]

In a single fight you have a very real chance that they will get 3 or 4 actions before you even get one, depending on how the dice roll.

 

This strikes at the heart of my question.

 

I think I"ve got it:

 

A SPD 4 character and a SPD 6 character. They play 8 segments, rolling a D10.

 

When the dice do not fall perfectly, they will fall to one side of exactly average (let's say we're using a D10).

 

8 rolls, low average: 1, 2, 3, 4, 5, 6, 7, 8 (avg 4.5)

SPD 6 goes in 6 segments, SPD 4 goes in 4 segments.

 

8 rolls, high average: 3, 4, 5, 6, 7, 8, 9, 10 (avg 6.5)

SPD 6 goes in 4 segments, SPD 4 goes in 2 segments.

 

As the rolls get more extreme, worst case for high SPD chas is that low SPD chas will go as often as they do.

But worst case for low SPD chas is that they might not get to go at all.

 

Yes, those are extreme, but the results fall on a bell curve, which means averages nearer the middle will have a less extreme effect - but still significant to favour high SPD characters.

[Mark Taylor]

Originally posted by Rebar

Yes, those are extreme, but the results fall on a bell curve, which means averages nearer the middle will have a less extreme effect - but still significant to favour high SPD characters.

 

I'm not sure I understand what you mean by this last paragraph. A single die does not generate a bell curve.

[Tom McCarthy]

I'm not a whiz, but I can fake it

 

You pose an interesting question, so I quickly had my computer run 12000 turns of cambat, and note how many phases the SPD 8 speedster got over the SPD 4 brick.

 

It looked like this...

 

0 extra phases 92 (0f 12000, less than 1%)

1 557

2 1523

3 2588

4 2862 (shockingly low ?)

5 2330

6 1246

7 582

8 190

9 27

10 1

11 2

 

Then I tried to slice it mathematically/analytically, where I'd expect:

 

0 92.5/12000

1 555/12000

2 1526/12000

3 2543/12000

4 2861/12000 (less than 24%)

5 2289/12000

6 1335/12000

7 572/12000

8 179/12000

9 40/12000

10 6/12000

11 .5/12000

12 .02/12000

 

So, nearly 40% of the time, SPD 8 doesn't get 4 more phases than SPD 4. 24% of the time he does. About 36% of the time, he gets more than 4 extra phases.

 

But, holding phases is a questionable tactic, since a held phase can be suddenly lost at any time.

 

By the way, I did it again for SPD 3 agents against SPD 6 martial artists, and got:

 

0 409 of 12000

1 1540

2 2732

3 3192

4 2293

5 1176

6 472

7 148

8 33

9 5

 

so, just more than 1/4 of the time do the phases stack up right by difference.

 

And for the stat geeks, by ratio, SPD 4 and 8,

 

Extremes, and expected

0:1 (SPD 4 got no actions, SPD 8 got unknown number; worst case for SPD 4) 95 of 12000

1:1 (SPD 4 and 8 got same number; best case for SPD 4) 96

1:2 1665 (expected ratio only occurs 14%)

 

Better for SPD 4

10:11 2

9:10 15

8:9 36

7:8 75

6:7 129

5:6 117

9:11 8

4:5 159

7:9 174

3:4 409

8:11 47

5:7 368

7:10 170

2:3 797

7:11 78

5:8 631

3:5 467

7:12 15

4:7 642

5:9 620

6:11 123

 

Worse for SPD 4

5:11 120

4:9 620

3:7 608

5:12 8

2:5 534

3:8 629

4:11 74

1:3 753

3:10 180

2:7 353

3:11 56

1:4 393

2:9 169

1:5 153

2:11 14

1:6 127

1:7 123

1:8 76

1:9 55

1:10 12

1:11 5

[Tom McCarthy]

That 0 ratio was bugging me. On a second run it broke down as:

 

0:2 2

0:3 3

0:4 14

0:5 12

0:6 24

0:7 12

0:8 15

0:9 2

0:10 2

 

Obviously, 0:10 is an extremely bad case for SPD 4.

[Mark Taylor]

Re: I'm not a whiz, but I can fake it

 

Originally posted by Tom McCarthy

*statistical stuff*

 

I don't have a problem with the numbers your computer spat out, personally. They're about what I would have expected. They're nowhere near extreme enough for me to write off this method as a bad idea. Although the numbers don't hit exactly the same values as the SPD chart that often, they cluster closely around them quite convincingly.

 

In essence, there is no practical difference between a character (of whatever SPD) missing out on a few phases, and a character getting those phases but fluffing the actions he takes in them. So essentially, the effect you are seeing is that the more dice rolls you add, the more characters of any SPD will be effected by pure chance, that is all.

 

I can see how the possibility of an extreme could cause a problem a superhero game in which a fight tends to last less than a full round, but in any more realistic genre I don't think it's a problem.

[Rebar]

so I quickly had my computer run 12000 turns of cambat, and note how many phases the SPD 8 speedster got over the SPD 4 brick.

 

1] I'm having trouble interpreting your numbers. Too much data, not enough labelling.

 

2] Yes, in the long run (12000 is a bit excessive), it will balance out. But that's not going to help the SPD 4 in a combat that's only 1 or 2 turns long.

 

3] When all is said and done, it's the balancing that doesn't balance. When it goes badly for low SPD chas, it goes

worse(ly) than when it goes badly for high SPD chas.

 

 

 

... the results fall on a bell curve... I'm not sure I understand what you mean by this last paragraph. A single die does not generate a bell curve.:

 

 

Make 10 rolls of a D10, record the average on a graph. It will likely be about 5.5.

Repeat that 10 times, recording each *average*. Most will fall around 5.5; some will fall high, some will fall low. Very few will fall at the extremes. Bell curve.

 

i.e. most combats will be mostly OK, some combats wil be biased (hi or low), a few will be very biased (hi or low).

[Rebar]

In essence, there is no practical difference between a character (of whatever SPD) missing out on a few phases, and a character getting those phases but fluffing the actions he takes in them..

 

There certainly is. One is player action. One is an unfair mechanic.

 

 

So essentially, the effect you are seeing is that the more dice rolls you add, the more characters of any SPD will be effected by pure chance, that is all

 

The problem is that it has a greater effect on a certain group than on others - namely lower SPD characters.

[Rebar]

A qualifier to those of you who may be wondering why this is being treated as a big deal.

 

Granted, nobody said everything balances out in the Hero System. Nobody said that putting 60 points into SPD gets you precisely 1.5x the effectiveness that only 40 points does. Thus, there is no fixed scale by which we can say that a system is biased against some construction. Of course it is.

 

The only thing we can say is that this system produces different results than the existing system - (incidentally, the one that has been play-tested).

 

My point in arguing this is that, to use this system, gamers should be aware how it changes things.

 

And there are few areas in the Hero System where subtlties are under more pressure than in the shear number of times you get to go in a turn. (That's why SPD costs 10 - count em - 10 points.)

[Agent X]

I've just skimmed the last few posts so if this has been said, I apologize.

 

The problem with this system is that things are more likely to go bad for the low speed character. On the bell curve, the low speed character is more likely to have wilder swings in a turn than a higher speed character, at least that is what I intuit.

[Mark Taylor]

Originally posted by Rebar

There certainly is. One is player action. One is an unfair mechanic.

 

The problem is that it has a greater effect on a certain group than on others - namely lower SPD characters.

 

It only has a greater effect on lower SPD characters if the dice rolls are particularly poor. It's no more unfair a mechanic that any other dice roll which may not go your way. What makes this particular dice roll more unfair to your mind?

 

Now, I'm not saying you have to like this mechanic, I would not presume to do so, I am saying that on average it produces essentially the same balance of results as the SPD chart, and the maths and figures bear that out.

 

Personally I actually think the picking the numbers out of the hat method mentioned by Netzilla is better because it ensures that everybody will get the correct number of actions, it just randomises the order.

[Mark Taylor]

Originally posted by Rebar

Make 10 rolls of a D10, record the average on a graph. It will likely be about 5.5.

Repeat that 10 times, recording each *average*. Most will fall around 5.5; some will fall high, some will fall low. Very few will fall at the extremes. Bell curve.

 

i.e. most combats will be mostly OK, some combats wil be biased (hi or low), a few will be very biased (hi or low).

 

Sure the averages fall onto a bell curve, however I'm failing to understand how these averages and the fact that they fall on a bell curve are significantly disadvantageous to low SPD characters. Each segment the die roll on a D12 is just as likely to be a 1 as a 12 or any other number. The bell curve that ensues from multiple rolls helps even things out, not throw them out of whack.

[Zeropoint]

Make 10 rolls of a D10, record the average on a graph. It will likely be about 5.5.

Repeat that 10 times, recording each *average*. Most will fall around 5.5; some will fall high, some will fall low. Very few will fall at the extremes. Bell curve.

 

Of *course* you get a bell curve--you just rolled 10d10 ten times.

 

Also, with ten dice, your bell curve will be severely clustered around the midpoint. It's been a while since my statistics class, so I can't say exactly how severely, but with ten d10's you'd only have a one in ten billion chance of a critical success (all ones).

 

Zeropoint

[Mark Taylor]

Originally posted by Agent X

I've just skimmed the last few posts so if this has been said, I apologize.

 

The problem with this system is that things are more likely to go bad for the low speed character. On the bell curve, the low speed character is more likely to have wilder swings in a turn than a higher speed character, at least that is what I intuit.

 

Yes the lower speed characters are more likely to be affected by wild swings, both downwards and upwards. In essence the lower SPD character has both an unfair disadvantage and an unfair advantage! As I said I can see it being a problem in superheroic campaigns where combat is short and sharp, but in a more realistic type of campaign it could work quite well. For less competent characters to experience wilder variations is what I would expect in a real situation.

[caris]

Re: Re: Re: Re: calling probability whizzes

 

Originally posted by Realms of Chaos

The "exactly the same balance" I was talking about is in purely statistical terms i.e. a SPD 4 character will on average get 4 phases out of every 12 segments, a SPD 5 character will on average get 5, etc. If anybody's unsure why that's the case I'm quite happy to post the maths which is really pretty basic.

 

Read what it says on Mark's site. To randomise combat and take away that element of predictability and planning is the whole purpose of his system! In other words he's aware of the issue you pointed out, and counts it as one of the two major reasons he prefers his system.

 

Personally, as a GM, I wouldn't say for sure I will use it but I will give it a try. IMO there is a reason most initiative systems have a random element - combat is chaotic and unpredictable, and that random element helps simulate it's chaotic and unpredictable nature.

 

I've always felt that combat should as be random and unpredictable as possible. The first time I went paintballing I realised it was even more random and unpredictable than even I had previously thought. Most of the time I couldn't even pass a PER roll to see who the hell just shot me, let alone predict what they were going to do next or when they were going to do it. Even very experienced paintballers who go on a regular basis find the same to be true.

I freely admitted it was a quibble, and I meant it. I just didn’t care for your choice of phrasing, and after considering it all day, I come to the conclusion that in and of itself it is not equally balanced without at least considering its impacts on other game elements like Haymaker, the value of certain levels of the Extra Time Limitation, and the cost of the Distance Shot maneuver. All of those items were cost and set up under an assumption of a fixed speed chart, change to a random speed chart is going to impact at least the perceived value of the elements. It would seem appropriate to allow characters who have purchased things like Combat Skill levels with Haymaker, or a power with the Limitation Extra Time: Delayed Phase the option to change the way those points are spent, if you are changing initiative significantly.

 

Of course, the purpose of the rule is to increase the random element, and unpredictability of combat. That is a given, but balance is about more than just statistical probability. How the change impacts the way the game plays, and interacts with the style of the players is important part of the balancing issue. While Mark is aware of the issue, it doesn’t meant that the GM who is considering adding it has considered the issue.

 

Also what you are not addressing (edit- ok, you were not addressing when I wrote this) is the fact that the proposed rules are only statistically identical if you use a 12 sided die, but the rules are not for a 12 sided dice. The rules are only for a die that is larger than the highest speed in the game. If the GM uses a smaller die than a 12 (as it seems to be the case in this situation since a 10 sided dice is given as the example), higher speeds are favored more than lower speeds. Look at speeds 3, 6 and 9, on a 12 sided die and a 10 sided die their probabilities are:

 

Speed 3 - 25% on D12 – 30% on D10 – net change increase 5%

Speed 6 – 50% on D12 – 60% on D10 – net change increase 10%

Speed 9 – 75% on D12 – 90% on D10 – net change increase 15%

 

Now on the other hand the reverse is true, dice larger than 12 sided loose probability faster than lower speeds.

 

Speed 3 – 3% on D100 – net loss 22%

Speed 6 – 6% on D100 – net loss 44%

Speed 9 – 9% on D100 – net loss 66%

 

As a matter of style, I don’t want any of my combat to be as chaotic as real life, which is not unusual very few games truly want to have combat match real life exactly. In my case what I like is the ability to make meaningful decisions in combat at a tactical level, it is something I enjoy. Many games really have you make all the decisions in a more strategic fashion. Most of the meaningful decisions are made before combat begins, and the whole combat is making a series of rolls to determine how well your strategy worked out. Granted part of it is the GM, but all the descriptive narration of what those dice rolls mean do not increase my enjoyment as much as being able to make meaningful decisions.

 

Obviously, increasing the similarities to real life combat is a priority for you and Mark. There is nothing wrong with that, and I consider the two levels to be equally good ways to play. I just don’t consider them identical.

[Agent X]

Originally posted by Realms of Chaos

Yes the lower speed characters are more likely to be affected by wild swings, both downwards and upwards. In essence the lower SPD character has both an unfair disadvantage and an unfair advantage! As I said I can see it being a problem in superheroic campaigns where combat is short and sharp, but in a more realistic type of campaign it could work quite well. For less competent characters to experience wilder variations is what I would expect in a real situation.

Yeah, but if you have a 4 speed and you the first 4 segments and the 8 speed guy goes 3 of the 4 segments you got 1 segment on him. If he goes 4 of the first 4 segments and you go one 1 of the first 4 segments he's got 3 segments on you. Which is more likely to happen?

[Mark Taylor]

Re: Re: Re: Re: Re: calling probability whizzes

 

Originally posted by caris

Obviously, increasing the similarities to real life combat is a priority for you and Mark. There is nothing wrong with that, and I consider the two levels to be equally good ways to play. I just don’t consider them identical.

 

Actually, it depends on the genre of game I'm running. I run a bunch of different genres. Realism is a priority in some, not so in others.

 

To quibble with the quibbles various people are throwing my way - I've already said (twice now) that I can see it would create problems in superheroic campaigns.

[Mark Taylor]

Originally posted by Agent X

Yeah, but if you have a 4 speed and you the first 4 segments and the 8 speed guy goes 3 of the 4 segments you got 1 segment on him. If he goes 4 of the first 4 segments and you go one 1 of the first 4 segments he's got 3 segments on you. Which is more likely to happen?

 

Well the first won't happen at all, because if a SPD 4 character got the first four segments the SPD 8 character would have got them too. In fact, lower speed characters will NEVER get more phases than higher speed character, however they're still more likely to get more phases than their actual SPD score than higher SPD characters are (and more likely to get less).

 

In a superheroic campaign sure, it's a problem. In a realistic campain, tough, that's the breaks. Life isn't fair. Yes, it's possible for low SPD characters to get no actions in a turn at all, but it won't happen very often, and even when it does, it's perfectly normal in real combat for less competent individuals to blunder around not doing anything useful because things are simply moving too fast for them!

[Mark Taylor]

Re: Re: Re: Re: Re: calling probability whizzes

 

Originally posted by caris

Also what you are not addressing (edit- ok, you were not addressing when I wrote this) is the fact that the proposed rules are only statistically identical if you use a 12 sided die, but the rules are not for a 12 sided dice.

 

Well actually I addessed it well before you wrote that. The opening words of the first post of mime to which you responded were "Assuming the die rolled is a D12". How much more directly could I address it?