Buying Down OMCV to Zero

[Christopher R Taylor]

I'm not sure anyone knows the reasoning for that exact number, but the basic thrust was I think just to make it harder to hit with area effect attacks.  I've used 0 DCV for a hex if its standing and 3 if its horizontal for a long time now.

[Surrealone]

Reminder: The hex you're in as well as those adjacent to you are DCV 0.

 

So no, a hex isn't that hard to hit if you're in it or it's close to you.  But 6' away, yea it's tougher than it was when it was right next to you or the one you were in … but still not tough.

[Killer Shrike]

3 hours ago, Chris Goodwin said:

 

Under ordinary circumstances, there's no mathematical difference.  Except that we have conditions, maneuvers, Powers, and so on, that put a target at reduced (half or zero) OCV or DCV.  The point of starting the CV's at 3 was for sure to keep them in line with historical values, partly for ease in converting, partly because most of us have some kind of a feel for an OCV of 3 or a DCV of 7, and partly because starting them at 0 would have messed with the game mechanics that reduce them.  

 

1/2 DCV itself is a problem, mathematically. It does not scale well at all.

 

Just at the most obvious level of consideration, for a high DCV character it is much more impactful than it is for a low DCV character.

 

Imagine if things that put you at 1/2 DCV instead gave a flat penalty ranging from -1 to -whatever. It would be mathematically consistent and completely scalable.

 

Quote

Incidentally, I've long thought that CV's ought to have remained tied to their original parent stats; they weren't Figured Characteristics as much as they were equivalent to Characteristic Rolls.  Decoupling those has never really sat right with me.  

 

That just leads to stat inflation; characters with absurd levels of DEX primarily to boost CV is so common in pre-6e that it is far more likely for it to be considered problematic when a character doesn't have inflated DEX. 

 

Separating CV from primaries was sensible. Separating O and D was sensible. But the opportunity to go further and re-baseline the CV numbers across the board was not taken. Half measures often result in unintended side effects.

 

Thus you end up with characters that have OMCV by default and no mental attacks, which is problematic (or at least silly). And secondarily, EGO is now almost useless...compare and contrast +10 EGO vs 10 Mental Defense for instance...with only a few situations where it provides greater utility than Mental Defense.

[Surrealone]

On 3/4/2019 at 10:07 PM, Killer Shrike said:

compare and contrast +10 EGO vs 10 Mental Defense for instance...with only a few situations where it provides greater utility than Mental Defense.

Not to digress, but, honestly, I believe Mental Defense provides more utility (at least, defensively) than straight EGO, since MD will apply as a defense to the STUN done by a Mental Blast … whereas straight EGO won't.  Thus, the defensively-minded character is incented to buy Mental Defense and, if s/he also cares about EGO rolls, s/he spends 2CP per +1 Skill Level to EGO rolls … especially since these levels are usable when EGO is drained to 0 (i.e. different drain needed to also drain away the levels).

Put another way, I tend to see no situation where buying straight EGO is a better deal than buying Mental Defense … specifically because the one drawback (EGO rolls suffering) is so easily overcome with 2pt Skill Levels to a single characteristic roll (EGO).

[massey]

45 minutes ago, Killer Shrike said:

 

1/2 DCV itself is a problem, mathematically. It does not scale well at all.

 

Just at the most obvious level of consideration, for a high DCV character it is much more impactful than it is for a low DCV character.

 

Imagine if things that put you at 1/2 DCV instead gave a flat penalty ranging from -1 to -whatever. It would be mathematically consistent and completely scalable.

 

 

It's not mathematically consistent at all, because Hero uses a bell curve.  Applying a flat penalty (-1, -2, -5, or whatever) will affect characters far differently, based upon the number needed to hit them in the first place.  Characters who have extremely high or extremely low DCVs will be affected less than characters with average DCVs.

 

So does 1/2 DCV of course, but I don't think you're gaining any kind of mathematical purity by changing to a flat penalty.

[Killer Shrike]

13 hours ago, massey said:

It's not mathematically consistent at all, because Hero uses a bell curve. 

 

A bell curve has nothing to do with mathematical consistency one way or the other. You are confusing consistency with distribution. Each step up or down (or left to right, depending on your visualization) along the standard normal distribution has a potentially differing probability density / i.e. is of different relative likelihood and are thus not of a constant value but that is not what consistency refers to in the context of mathematics. 

 

If Character A and Character B attempt the same action and suffer the same bonus or penalty value, it is consistent and also the guassian curve / standard normal distribution applies. 

 

But if Character A and Character B attempt the same action and suffer a different bonus or penalty to each other, then that is inconsistent, and the bell curve doesn't PREVENT that inconsistency...though it does MEASURE the inconsistency relative to the normal distribution. 

 

In other words, if Character A and Character B both take the same 1/2 DCV action, and Character A suffers a -3 penalty for doing so while Character B suffers a -1 penalty for doing so, a bell curve does not prevent that and nor does it somehow make that "consistent".

 

Lets have some fun with numbers...

 

3d6 Bell Curve
	3d6
	Pct	Odds
3	0.5%	215 : 1
4	1.9%	53 : 1
5	4.6%	20.6 : 1
6	9.3%	9.8 : 1
7	16.2%	5.2 : 1
8	25.9%	2.9 : 1
9	37.5%	1.7 : 1
10	50.0%	1 : 1
11	62.5%	1 : 1.7
12	74.1%	1 : 2.9
13	83.8%	1 : 5.2
14	90.7%	1 : 9.8
15	95.4%	1 : 20.6
16	98.1%	1 : 53
17	99.5%	1 : 215
18	100%	1 : inf

 

Characters A & B.

 

Character A has 6 DCV, Character B has 3 DCV.

 

Character A takes a 1/2 DCV action and drops to 3 DCV. 

 

Character B takes a 1/2 DCV action and drops to 2 DCV.

 

Assuming a neutral 11- 62.5% to hit, Character A suffers 21.3% of probability protection vs getting hit while Character B only suffers 11.6%...in other words, on neutral odds Character B is basically half as impacted as the character who paid for twice as much DCV. This is inconsistent between the two characters (they are affected in a disproportionate way, in real numbers), and further is also inconsistent to the system's meta concept of "pay more points to get more benefit" in that the character that paid more to be good at not getting hit suffers a bigger penalty when using the 1/2 DCV action. Thus there is an entire category of abilities and situations (those resulting in 1/2 DCV) that are biased against the character that spent points to raise their DCV and are biased towards the character that did not. 

 

If the action imposed a flat -2 DCV rather than 1/2 DCV, then Character A drops to DCV 4 and Character B drops to DCV 1. That works out to a 11.6% loss for Character A and a 12.5% loss for Character B. The character who paid less is affected slightly more, which better befits "get what you pay for", and the difference between the two outcomes is much closer. The range is smaller, the variance is less, the impact is linear, the penalty value is the same (-2 vs -?), and thus it produces results that are both more consistent and more normative.

 

It doesn't matter which values you plug in, as long as you avoid break points. Every 2 steps of DCV results in a deepening of the impact.

 

Put a different way. A 1/2 DCV maneuver is a -1 DCV penalty maneuver for DCV 3 Guy and a -3 DCV penalty maneuver for DCV 6 Guy...for no viable reason. This is particularly odd in that penalties in the Hero System are generally used as consistent negative modifiers for the purposes of reducing % chance of success to represent the relative difficulty of whatever is being attempted by a character. They are not relative to the character attempting the task, they are fixed per the task being attempted. For example, modifiers based upon relative ease of a given task are:

 

 

Thus, a character with 12 DCV taking a 1/2 DCV action is the equivalent of a "Sheer Folly" difficulty (-6), while for the character with 3 DCV to take the same action its the equivalent of a "Difficult" difficulty (-1). This doesn't make logical sense to me, as the two characters are attempting the same action.

 

But I hear you say "well, those are task modifiers, dummy. DCV penalties are different" (somehow , even though a 3D6 roll under attack and a 3D6 roll under skill are actually on the same curve, but whatever).

 

Uh huh. Ok, how about this then?  Poor footing is a -1 DCV. If it is also on a tight rope that's another -2 DCV. So, a character fighting on an oil slicked tightrope takes a flat -3 DCV, while a 8 DCV character taking a 1/2 DCV action such as Bracing suffers a -4 DCV. That may make sense to you, but it does not make sense to me.

 

IMO, if a find & replace were done across the rulebooks to change all references to 1/2 DCV to instead be a flat -2 DCV it would remove that sort of nonsense (similarly, though much less commonly a thing, 1/2 OCV to -2 OCV). There's some logical follow through permutations to that, but I'll punt on exploring them right now as I have other things to get to tonight.

[Toxxus]

16 hours ago, Surrealone said:

Sure there's a downside to doing it -- if you pick up some object that requires decent OMCV to properly target it, the character can make less effective use of that object than others.

 

I'd have to artificially insert something into a campaign that otherwise wouldn't have such items and the characters who sold off their MOCV would all turn to the mage and bard in the party who have actually raised their MOCVs and just hand it over.

 

There would be zero inconvenience unless I further contrived an encounter where the bard & mage get - say Webbed - where their spell casting is shut down and their low STR scores prevent them from escaping and then follow it up with making such an item only be within reach of the characters who sold it off and.. barf...  it would be so obvious I was targeting the character for the sell off that it would likely irritate the player.

 

And while I wouldn't characterize myself as lazy I do have a full time job and a family so the two tables a week I run are based on ready made campaigns (one Pathfinder the other D&D) that I convert on the fly.  I may sprinkle in some ad lib material here and there, but I lack the time for a fully home built campaign.

 

 

[Hugh Neilson]

 

15 hours ago, Surrealone said:

If everyone's OMCV and DMCV default to 1, how is there a balance issue, exactly … especially if the GM is putting genre-appropriate caps in place for OCV, DCV, OMCV, and DMCV?

 

I agree in general - if everyone starts at 1 and the Mentalist invests 12 points for OmCV 5, he hits just as easily than if everyone starts at 3 and he invests 12 points of OmCV 7.

 

It does, however, make attacks using alternative CV a bit more challenging, as it is even easier to target DmCV.  We can always adjust the cost to taste, though.

 

15 hours ago, Killer Shrike said:

 

I basically agree with this, except I go literally "1" further...OCV / DCV / OMCV / and DMCV should all have defaulted to 0 in 6e.

 

The 11- attack roll is balanced unto itself. If the O and D values are the same value by default, there is no mathematical difference between ((11 + x) -x)- and just 11-.

 

There was no intrinsic reason for the values to be pegged at 3 other than past practice.

 

It does mean no one is "worse than normal".  It seems like that should be possible, at least for physical CVs where everyone has thrown a ball/dart/etc. and dodged something (even in a dodgeball game).

 

13 hours ago, Christopher R Taylor said:

Yeah that's an old rule which I question greatly: is it really that hard to hit a hex??

 

62.5% chance, when running around in combat (standing and focusing is bracing and setting, so better chances to hit) seems reasonable to me.  Remembering that this is a hex that is at least 6' or so away - the adjacent ones are DCV 0.

 

 

[Killer Shrike]

1 minute ago, Hugh Neilson said:

It does mean no one is "worse than normal".  It seems like that should be possible, at least for physical CVs where everyone has thrown a ball/dart/etc. and dodged something (even in a dodgeball game).

 

Currently, rules as written, OCV can be negative while DCV floors at 0 DCV, which is an odd case of non-mirrored rules and the discrepancy is not really explained in the rules text.

 

Negative OCV basically converts over to a DCV bonus for the target. It seems logical that negative DCV could be allowed and work the same way, acting as an OCV bonus for an attacker. If DCV were defaulted to 0 it would be worth considering. 

[Hugh Neilson]

5 minutes ago, Killer Shrike said:

 

Currently, rules as written, OCV can be negative while DCV floors at 0 DCV, which is an odd case of non-mirrored rules and the discrepancy is not really explained in the rules text.

 

Negative OCV basically converts over to a DCV bonus for the target. It seems logical that negative DCV could be allowed and work the same way, acting as an OCV bonus for an attacker. If DCV were defaulted to 0 it would be worth considering. 

 

I'd agree with penalties being able to make DCV negative. 

 

Not sure I would want the stats themselves to be able to go negative.  "Well, I'll just sell my OmCV back to -1,000 - now I have lots of points!"  Sure, no GM will allow it, but then it becomes the OmCV Limbo - how low can you go?

[Hugh Neilson]

To the question of the drawback, they saved 6 points (or 9).  Issues as common and severe as those from a Complication worth 10 or 15 points seems quite sufficient.

 

It does raise the age-old question of whether sellbacks should reduce points spent, or be a Complication themselves, but that's another can of worms.

[Killer Shrike]

1 hour ago, Hugh Neilson said:

Not sure I would want the stats themselves to be able to go negative.  "Well, I'll just sell my OmCV back to -1,000 - now I have lots of points!"  Sure, no GM will allow it, but then it becomes the OmCV Limbo - how low can you go?

 

DCV the characteristic would floor to 0. 

 

Calculated DCV, using the DCV calculation rules, currently also floors to 0 after penalties. Removing the floor from the calculated DCV and allowing the calculated DCV to go below 0 in exactly the same was as characteristic OCV / calculated OCV currently works, is what is being discussed here.

 

The main takeaway being characteristic vs calculated.

[Christopher R Taylor]

Quote

It does raise the age-old question of whether sellbacks should reduce points spent, or be a Complication themselves, but that's another can of worms.

 

This is an interesting question. In Hero terms, it seems like it really ought to be in the complications side of the equation: things which make your character weaker or disadvantaged overall.  Doing so would definitely address some balance issues, since you only get credit for so many points in complications.

[massey]

11 hours ago, Killer Shrike said:

 

A bell curve has nothing to do with mathematical consistency one way or the other. You are confusing consistency with distribution. Each step up or down (or left to right, depending on your visualization) along the standard normal distribution has a potentially differing probability density / i.e. is of different relative likelihood and are thus not of a constant value but that is not what consistency refers to in the context of mathematics. 

 

If Character A and Character B attempt the same action and suffer the same bonus or penalty value, it is consistent and also the guassian curve / standard normal distribution applies. 

 

But if Character A and Character B attempt the same action and suffer a different bonus or penalty to each other, then that is inconsistent, and the bell curve doesn't PREVENT that inconsistency...though it does MEASURE the inconsistency relative to the normal distribution. 

 

In other words, if Character A and Character B both take the same 1/2 DCV action, and Character A suffers a -3 penalty for doing so while Character B suffers a -1 penalty for doing so, a bell curve does not prevent that and nor does it somehow make that "consistent".

 

(snip lots of charts)

 

Good chart with the 3D6 distribution.  I need to save that.

 

My point with regards to mathematical consistency is not that the bell curve gives it.  I know it doesn't.  It's that you only get the illusion of consistency by going to a set minus.  You can make the 11-, 12-, 13- look consistent, but the actual percentage odds of success will still vary wildly.  You can make one set of numbers look nice and even, but the others aren't going to match it.

 

 

 

Professor Executioner has a 12 OCV.  He targets Captain Amazo (normally DCV 8).  He needs a 15- to hit, meaning he should expect to succeed (looks at chart) 95.4% of the time.  But Captain Amazo is Stunned from the last time Professor Executioner shot him, so his DCV is halved and is only a 4.  The Professor need only avoid rolling an 18 to hit him.  His chance of success rises to a mighty 99.5%, but only a 4.1% increase.  What if we make it a set -3 DCV while Stunned, instead of 1/2?  Same exact odds, still a 99.5% chance of success.

 

Compare this to Viper Agent #3, who opens fire on Captain Amazo.  #3 has an OCV of 6, and he's shooting at a DCV of 8.  He needs a 9-,  for a 37.5% chance of success.  If Captain Amazo's DCV is halved, then he needs a 13-.  His odds go up to 83.8%, an enormous 46.3% swing.  That's over 10 times the benefit that Professor Executioner got.  If we make it a flat -3 penalty, it's still a 12-, or 74.1% chance of success.  Still a 36.6% improvement.

 

The point is, the average odds of success are going to vary wildly.  The odds of hitting aren't consistent, even when we're talking about one character with a set DCV.

[Killer Shrike]

4 hours ago, massey said:

Good chart with the 3D6 distribution.  I need to save that.

 

http://www.killershrike.com/GeneralHero/BellCurve.aspx

 

Quote

My point with regards to mathematical consistency is not that the bell curve gives it.  I know it doesn't.  It's that you only get the illusion of consistency by going to a set minus.  You can make the 11-, 12-, 13- look consistent, but the actual percentage odds of success will still vary wildly.  You can make one set of numbers look nice and even, but the others aren't going to match it.

 

I encourage you to please consider clicking the links for consistency and distribution and bone up on the subjects a bit, it probably won't kill you and it might even benefit you.

 

Quote

Professor Executioner has a 12 OCV.  He targets Captain Amazo (normally DCV 8).  He needs a 15- to hit, meaning he should expect to succeed (looks at chart) 95.4% of the time.  But Captain Amazo is Stunned from the last time Professor Executioner shot him, so his DCV is halved and is only a 4.  The Professor need only avoid rolling an 18 to hit him.  His chance of success rises to a mighty 99.5%, but only a 4.1% increase.  What if we make it a set -3 DCV while Stunned, instead of 1/2?  Same exact odds, still a 99.5% chance of success.

 

-2 is what I'm suggesting, actually. But that's irrelevant to the flaw in your logic here. Allow me to break it down.

 

In your scenario the attacker has a +4 net bonus over the target at full DCV. No amount of penalty to DCV will make much of a difference on a 3D6 curve when the before-penalty difference is 4 steps or more away from mean. That would be classed in the category of "working as intended". 

 

So lets instead try an example made of less straw.  

 

Captain A normally has a 12 DCV but is currently stunned. As written, Captain A drops to 6 DCV, suffering a -6 DCV penalty. As suggested Captain A drops to 10 DCV, suffering a -2 penalty. 

 

Also in the scene is Doctor Stranko who is normally 10 DCV, but is Concentrating to cast a spell. As written, he drops to 5 DCV, suffering a -5 DCV penalty. As suggested, Dr. S instead drops to 8 DCV suffering a flat -2 penalty.

 

Professor Ex is there with his 12 OCV. If he were to target staggered Captain A, as suggested he'd hit ~84% of the time vs 99.5% of the time as written. If he were to target concentrated Dr. S he'd hit 95.4% vs a 99.5% as written.

 

So breaking this down finely:

 

You seem to be under the impression that "consistent" refers to the differences between percentages at each step in the distribution. It does not.  An equation system is inconsistent if there is no set of values for the unknowns that satisfies all of the equations; and is consistent if the opposite is true. In this case the wildcard nature of 1/2 DCV resulting in a different value for the same situation (1/2 DCV) is not consistent. 

 

p = DCV penalty

 

If p is a fixed value, then the following could be written and would be true where p=2

 

C1(11 + 12) - (12-p) = 13-

C2(11 + 12) - (10-p) = 15-

 

12-10 = 15-13

 

This could be reexpressed as the more general function which is consistent for any fixed value of p.

 

C1(11 + x) - (y-p) = r1-

C2(11 + x) - (z-p) = r2-

 

where y>=z

y-z = r2-r1

 

You can then test that by plugging in whatever positive values you like and you should find that it is always consistent when p is a specific fixed value and is not consistent when p is indeterminate due to being based on a halving of some other variable and which may be a different value per character. Further statements could be made from there.

As an aside, removal of the multiplicative aspect of 1/2 DCV also simplifies the order of operations and would allow the notation to tighten up a bit and also allow them to be applied in any order:

 

C1(11 + x - y - p) = r1-

etc

 

But that's not relevant / doesn't change anything for purposes of this discussion per se. The main takeaway is that the value of the penalty itself as a fixed number is consistent (the value of the penalty itself is consistent in the colloquial sense and it is possible to formulate mathematically consistent expressions around it) vs the value of the penalty as a fraction of some other number (the value of the penalty is not consistent colloquially; it may or may not be consistent mathematically speaking...feel free to work it out if you are motivated to do so).

 

 

You also seem to be under the impression that the way to assess impact on a Gaussian curve is to do so at the edges. It isn't. The greatest impact is by definition closest to the mean. That is in fact the core concept. To assess the impact one looks most closely around the mean. I.e., on a 3D6 / 3-18 / 16 step curve between +/-1 to +/-4, with +/-3 being the point of collapsing probabilities. Ever wonder why 8- and 14- are "special" / show up in various contexts in the HS? That's why.

 

CV's from 4 to 5 (or +1 to +2 over baseline) would not be significantly affected by a change to -2 vs 1/2 DCV. It starts to matter a bit more around CV 6 (+3 over baseline), and then every 2 steps beyond that matter.

 

Similarly, if the attacker's OCV is more than 3 steps (+4 or better) beyond the defender's DCV before penalties, then the DCV penalty itself is so close to being irrelevant as to not matter as the attacker is already out on the edge of the distribution curve and are meant to be there. 

 

Quote

Compare this to Viper Agent #3, who opens fire on Captain Amazo.  #3 has an OCV of 6, and he's shooting at a DCV of 8.  He needs a 9-,  for a 37.5% chance of success.  If Captain Amazo's DCV is halved, then he needs a 13-.  His odds go up to 83.8%, an enormous 46.3% swing.  That's over 10 times the benefit that Professor Executioner got.  If we make it a flat -3 penalty, it's still a 12-, or 74.1% chance of success.  Still a 36.6% improvement.

 

As we are talking about a 1/2 DCV penalty vs a -2 DCV penalty the difference to be compared is that between the two penalties, and at different values of DCV. So poor Agent #3 is going to be busy attacking three different opponents at different DCV's calculated for 1/2 DCV and -2 DCV.

 

vs DCV 8 Guy example:

RAW 1/2 DCV penalty:  OCV 6 vs DCV 4: 13- / 83.8% == +46.3% diff

flat -2 penalty:                 OCV 6 vs DCV 6: 11- / 62.5% ==  +25% diff

 

As we are talking about the idea that two different characters with different DCV's being affected in a more consistent way by taking an action that is currently 1/2 DCV you also have to plug in a second target to compare and contrast. Thus:

 

vs DCV 6 Guy example:

RAW 1/2 DCV penalty:  OCV 6 vs DCV 3: 14- / 90.7% == +28.8% diff

flat -2 penalty:                 OCV 6 vs DCV 4: 13- / 83.8% ==  +21.3% diff

 

What we see here is the 1/2 DCV version swinging wildly even at a small difference in base DCV (17.5% jump from DCV 6 to DCV 7), while the flat -2 results does not exhibit that behavior relative to itself or its progression.

 

We can further show that by stepping up.

 

vs DCV 10 Guy example:

RAW 1/2 DCV penalty:  OCV 6 vs DCV 5: 12- / 74.1% == +57.9% diff

flat -2 penalty:                 OCV 6 vs DCV 8: 9- / 37.5% ==  +21.3% diff

 

Quote

The point is, the average odds of success are going to vary wildly.  The odds of hitting aren't consistent, even when we're talking about one character with a set DCV.

 

Again, you are misusing the term "consistent" when what you are talking about is probability density / different relative likelihoods along the steps of a normalized distribution curve. 

 

To summarize your argument: You seem to think I don't know that the step from say 11- to 12- represents a greater probability density than say the step from 14- to 15-. I assure you I am fully aware of this. You seem to think that the term "consistent" refers to the %-ile values across the steps of the curve. It doesn't. You seem to think that if there is minuscule difference at the edges of the curve between the two methods (which of course there is because that is how normalized distribution curves work) it means the flat penalty offers no benefit over the 1/2 DCV version. This is incorrect; the impact is felt towards the middle of the curve (which of course it would be because that is how normalized distribution curves work).

Edited March 5, 2019 by Killer Shrike
removed unnecessarily snarky statement. I don't mean to be so snarky, it's a character flaw and I'm working on it.