Probabilities and dice rolls


Advice

1 to 50 of 70 << first < prev | 1 | 2 | next > last >>

Pathfinder Pawns, Starfinder Roleplaying Game, Starfinder Society Subscriber; Pathfinder Roleplaying Game Superscriber

I suck at math and need a quick tutor. I'm trying to figure out the probability that my character will get at least one crit against an equal-leveled enemy under relatively ideal circumstances.

The character has a +31 to hit normally. However, when possible the character casts heroism (6th) before a fight, so their attack roll would then be +33. The character also has haste, double slice, and two-weapon flurry. That means, he can potentially have up to five attacks, the first two with no penalty, and the latter three with a -10 penalty.

According to the GameMastery Guide a moderate AC of a 16th-level creature is 38.

That means I need at least a 48 to get a crit.

48 - 33 = 15, or 30% chance of a crit on each of the first and second attacks.

The -10 penalty is pretty crippling, making it so we can only crit on a natural 20 (or 5% chance).

So the chances of getting a crit on each die roll looks something like this:
roll 1: 30%
roll 2: 30%
roll 3: 05%
roll 4: 05%
roll 5: 05%

...And that's about as far as I've got. I've Googled math tutorials online, but they don't seem to cover this situation (it's always trying to multiply the percentages together). I've also tried online calculators, but the best one I've found so far only showed the conjunction of 4 events (I need 5) and didn't even show the work for me to deduce how it all came together.

So now I'm here, looking for a tutor to show me how it is done and maybe help me pick up some of the proper vocabulary terms for this type of math problem (so that future Google searches might yield more success).


4 people marked this as a favorite.

You multiply the probability of failure each time.

So 0.7 * 0.7 * 0.95 * 0.95 * 0.95 to find out a single round.

Then if you wish you can just take 0.42011375 and multiply it against it's self.

EG. 3 rounds would be 0.42011375^3, or 0.42011375 * 0.42011375 * 0.42011375. Around a 0.07% probability of not rolling a crit in three rounds of 5 attacks a round. (a 97% probability that there will be a crit in 3 rounds)

I will leave the vocabulary terms for others :)


30% chance of a crit on the first attack.
30% chance of a crit on the second attack, but we don't want to treat it as an independent event. We want to figure out the probability that we crit on the second attack after not critting on the first; since we're looking for "at least one," it doesn't matter if both are crits and we don't want to count the outcomes that include more than one critical hit multiple times. 30% x 70% = 21% chance that the second attack will crit after the first one did not. 51% total chance that at least one of the two will crit.
Then 5% x 49% = 2.45%. 53.45% total.
5% x 46.55% = ...I need a calculator. Less than 2.45%, but not by much. Okay, it's about 2.33% but I've had to round now. Total of 55.78%.
5% x 44.22% = 2.21%. Total of about 58% chance at least one will crit.

Oh. Somebody else posted already.
Yeah, figuring out the odds that none will crit, and then subtracting that from 100% is... probably easier. Heh.


If in doubt I use probability trees for easier problems because as far as I am concerned they provide good visuals for any probability problem:

Coin toss example

Simply note the probability of each event on each branch (which easily can be more than 2 branches in case of multiple outcomes). Each individual probability is simply achieved by multiplying the probabilities of each individual branch over each number of events.

The overall total of the probabilities for any chain of events is always 1 (or 100%; that is the "Add" part of the example).

Chances of something not happening even once simply uses the product of all complementary probabilities (or failure probabilities) like @The Gleeful Grognard already explained. If you toss a coin and heads is the desired result this would be the events probability (50% in case of a single toss) and subsequently tails would be the complementary probability (also 50% in case of a single toss).

Note that using a probability tree will give you the same (failure) result of 0.42011375 for a single round, just visualized. It will also provide chances for exactly 1 crit, 2 crits etc if you follow the branches (multiply) and add up each branch with the same result. Also note that probability trees get out of hand really fast for cases with multiple outcomes and/or multiple events as they then grow in size rapidly, so no one would use a probability tree for like 526 coin tosses.

However I am just a dabbler in maths too so lets see what @Mathmuse has to add. ;)


13 people marked this as a favorite.

As a math professor, I'll throw my hat in here.

Your initial setup is correct, Ravingdork. The Gleeful Grognard has the right approach to complete the problem, but made a couple of arithmetic errors at the end. I'll work through the problem and try to fill in a bit of the language. I'll assume three rounds for the first case.

This is an example of a Classical Probability problem. (There are other types such as Empirical Probability.)

You've done the first part correctly.

For each round:
1st attack: 30% chance of a crit
2nd attack: 30% chance of a crit
3rd attack: 5% chance of a crit
4th attack: 5% chance of a crit
5th attack: 5% chance of a crit

Now if you multiplied those together like the online calculators were trying to do then you would have the probabilty that all five attacks resulted in crits. But that's not what we want here. We want the probability that there is at least one crit and that could mean 1, 2, 3, 4, or 5 crits.

The Gleeful Grognard has the right approach. Instead of using the probabilities of getting crits, we want to use the probabilities of not getting crits. Getting a crit and not getting a crit are Complementary Events. These are events that are mutually exclusive and exhaustive. You can't both get a crit and not get a crit on a single roll (mutually exclusive) and every roll will result in one of those two results (exhaustive).

Using complements is a really great trick in cases like this because there are many, many, many ways to get at least one crit on 15 attacks. That makes for a lot of calculations. But the alternative to getting at least one crit is to get no crits, and there is only one way for that to happen - they all must fail. I tell my students to look out for that phrase "at least one." It's a clear signal that you should try thinking in terms of the complement.

Because complementary events split the total probability between them, their probabilities must always add to 100%.

So we have the following:

1st attack: 30% chance of a crit -> 70% chance of not a crit
2nd attack: 30% chance of a crit -> 70% chance of not a crit
3rd attack: 5% chance of a crit -> 95% chance of not a crit
4th attack: 5% chance of a crit -> 95% chance of not a crit
5th attack: 5% chance of a crit -> 95% chance of not a crit

Now we are assuming that these probabilities aren't affected by the outcomes of previous attacks and that makes them Independent Events. To calculate the probability that multiple independent events occur, we multiply their individual probabilities. That's what The Gleeful Grognard did.

So the probability that all of the attacks fail to crit is:
0.7 * 0.7 * 0.95 * 0.95 * 0.95 = 0.42011375

This is the Multiplication Rule for Independent Events: If events A and B are independent of each other, then the probability that both A and B occur is the probability that A occurs times the probability that B occurs. Or P(A and B) = P(A) * P(B)

(As a counter-example, if a crit on an attack were to change the opponent's AC by leaving them flat-footed, then later attack probabilities would depend on whether previous attacks succeeded or failed. In that case the attacks would be dependent events and would require more complicated formulas involving Conditional Probabilities.)

We are also treating each round as independent from the others so, as The Gleeful Grognard did, we apply the Multiplication Rule again. Since the probabilities are the same each round we can use exponents as a shortcut for repeated multiplication.

So for three rounds you would get (0.42011375)^3 for the probability that you got no crits in three rounds. Here the The Gleeful Grognard made an error in misplacing the decimal point. (0.42011375)^3 = 0.0741482128 = 7.4%, not 0.07%

So there is a 7.4% chance that none of the 15 attacks will succeed. Using the concept of complement again, that means that there is a 100% - 7.4% = 92.6% chance that at least one attack will crit.

If you want to calculate the probability of at least one crit in n rounds you would just use the formula:

1 - (0.42011375)^n

and then multiply by 100 if you want the result as a percent.

Example:

The probability that you would have at least one crit in 5 rounds would be

1 - (0.42011375)^5 = 0.986913 = 98.7%

I hope that helps.


I was basically going to explain what Gisher did, but since they are a math professor I will just say that was a very good ELI5 explanation on basic probabilities.

But it also reminds me I need to brush up on my conditional probabilities.


1 person marked this as a favorite.

I will note that most enemies (of ALL levels) have High AC, not moderate. Its very very close to 2/3rds are High, a couple have extreme, and the rest are scattered at Moderate, Low, and Even Somehow Worse.

Of the 8 level 16 critters in Bestiary 1, five of them have an AC of 39. Only one has 38 and the other two are 37 (I don't have a convenient way of searching Nethys for Bestiary 2 critters of a particular level).


Draco18s wrote:

I will note that most enemies (of ALL levels) have High AC, not moderate. Its very very close to 2/3rds are High, a couple have extreme, and the rest are scattered at Moderate, Low, and Even Somehow Worse.

Of the 8 level 16 critters in Bestiary 1, five of them have an AC of 39. Only one has 38 and the other two are 37 (I don't have a convenient way of searching Nethys for Bestiary 2 critters of a particular level).

Interesting, thanks for sharing.


Draco18s wrote:
(I don't have a convenient way of searching Nethys for Bestiary 2 critters of a particular level).

Why not? Just use the filter?

Set source to Bestiary 2 and the creature level to an approprate range and it will list all monsters of that range.

For example, Bestiary 2 and level 19 achieved the following results:

Ancient Cloud Dragon
Grendel
Sard
Star Archon
Taiga Linnorm
Vrolikai (Death Demon)


Ubertron_X wrote:
Draco18s wrote:
(I don't have a convenient way of searching Nethys for Bestiary 2 critters of a particular level).
Why not? Just use the filter?

Mostly because I can never get AoN's search to work right. Mostly me not understanding how to use it, I suppose, but I just generally don't bother with it any more.

Anyway, Bestiary 2:
38 AC: 1
39 AC: 3
40 AC: 1
42 AC: 1

So, even higher (and the average is over a full point!) than Bestiary 1.


It's kind of funny to me the mean and the mode of AC is what Paizo wrote as the high AC instead of the moderate AC values.

To me this indicates that individual designers are doing what they think feels right for the monster, but in so doing are artificially inflating the stats of monsters by accident.


Pathfinder Pawns, Starfinder Roleplaying Game, Starfinder Society Subscriber; Pathfinder Roleplaying Game Superscriber

Sounds to me like some errata is in order.


Ravingdork wrote:
Sounds to me like some errata is in order.

I hate to say that about changing monsters, but it feels necessary.

1 or 2 points of AC is significant in PF2, and if the average (mean/mode) is above the published guidelines for what should be "average" for the CR then it feels like it's wrong.

And I already had problems with to-hit in PF2, and this is just sort of further proof about why it's a problem.

The further we explore the system the less and less I like tying critical success/failure to the DC +/- 10 system.

I definitely prefer Starfinders flat system of nat 20/1 rolls only as crit failure/success with almost no ways to expand the crit range (I only know of one thing and doesn't make it a real crit, just allows you to apply critical effects but doesn't give double damage).


2 people marked this as a favorite.
Claxon wrote:

It's kind of funny to me the mean and the mode of AC is what Paizo wrote as the high AC instead of the moderate AC values.

To me this indicates that individual designers are doing what they think feels right for the monster, but in so doing are artificially inflating the stats of monsters by accident.

It's more that their choice of naming pattern doesn't mesh with intuitive reading because they are very clear that what they named "high" is their default.

People read "high" and assume it means "higher than normal" rather than "normal."

Just like how some video games have a difficult setting but instead of labeling one of them "normal" and basing the other names off that, they have "hard" and "easy" and "nightmare" or some other evocative name and they treat "hard" as the default - but people intuit "hard" to mean, well, hard so they might think they want to play on easy because what they are really looking for is "normal."


Claxon wrote:
Ravingdork wrote:
Sounds to me like some errata is in order.

I hate to say that about changing monsters, but it feels necessary.

1 or 2 points of AC is significant in PF2, and if the average (mean/mode) is above the published guidelines for what should be "average" for the CR then it feels like it's wrong.

And I already had problems with to-hit in PF2, and this is just sort of further proof about why it's a problem.

The further we explore the system the less and less I like tying critical success/failure to the DC +/- 10 system.

I definitely prefer Starfinders flat system of nat 20/1 rolls only as crit failure/success with almost no ways to expand the crit range (I only know of one thing and doesn't make it a real crit, just allows you to apply critical effects but doesn't give double damage).

At the very least, I would say that more care may need to be taken by GMs when designing encounters. The examples given are obviously for higher level content (which may well deserve to have a wider range for a variety of reasons) but if it turns out (as is likely) that this problem persists throughout the entire level range, then its a bigger problem and makes you wonder why it wasn't discovered during playtesting. Now the flip side of the coin is to also look at expectant to hit bonuses per level as well. If, for instance, monsters have higher AC than anticipated, but also lower to hit, then the issue may well cancel out to an extent -- though at the cost of longer, more drawn out battles. If the to hit is also high though then it really is setting the PCs up to fail AND more or less requiring that they min/max just to stay viable.

Now, that said, part of the underlying assumptions may well be that it is presumed that many battles will feature multiple, lower level, enemies rather than just a single enemy. That too would change the math somewhat. Less effective monsters being balanced by having more actions per round, more attacks, and presumably more HP too (in total, maybe). But, if that's the case, then it should be made more explicitly clear. Don't get me wrong, as a player I don't mind the game being difficult per se. That brings its own sense of accomplishment. But if the stated expectations of difficulty (i.e. Defenses/to-hit/etc) are not accurate then its going to be a problem for a lot of groups as I would imagine most GMs would assume a given monster is appropriate for its level.

Design Manager

6 people marked this as a favorite.
thenobledrake wrote:
Claxon wrote:

It's kind of funny to me the mean and the mode of AC is what Paizo wrote as the high AC instead of the moderate AC values.

To me this indicates that individual designers are doing what they think feels right for the monster, but in so doing are artificially inflating the stats of monsters by accident.

It's more that their choice of naming pattern doesn't mesh with intuitive reading because they are very clear that what they named "high" is their default.

People read "high" and assume it means "higher than normal" rather than "normal."

Just like how some video games have a difficult setting but instead of labeling one of them "normal" and basing the other names off that, they have "hard" and "easy" and "nightmare" or some other evocative name and they treat "hard" as the default - but people intuit "hard" to mean, well, hard so they might think they want to play on easy because what they are really looking for is "normal."

The problem we found is that the default attack/AC values for building your simple run of the mill combatant (high attack and high AC as they are called right now) are fairly high. The high AC isn't anything to write home about compared to what the best PCs can pull off but is still quite good, and the attack bonus is around a fighter's and very good. So if you call them "moderate" that seems off to players even more so because they are in fact pretty high. So the best thing was to call them high and then make it very clear that high was the default in those statistics.


Gisher wrote:
Your initial setup is correct, Ravingdork. The Gleeful Grognard has the right approach to complete the problem, but made a couple of arithmetic errors at the end.

Yeah, I dislike how I cannot go back and edit my post -laughs-; I noticed my mistake myself as I went to fix a grammatical error that was bothering me ("it's self").

TY, great breakdown.


1 person marked this as a favorite.
Mark Seifter wrote:
The problem we found is that the default attack/AC values for building your simple run of the mill combatant (high attack and high AC as they are called right now) are fairly high. The high AC isn't anything to write home about compared to what the best PCs can pull off but is still quite good, and the attack bonus is around a fighter's and very good. So if you call them "moderate" that seems off to players even more so because they are in fact pretty high. So the best thing was to call them high and then make it very clear that high was the default in those statistics.

Thanks for the explanation, Mark. I hope it helps folks that are struggling with their own expectations right now by thinking that because AC throughout the Bestiaries trends to what is labeled as "high" means that ACs are higher than intended.


Mark Seifter wrote:
The problem we found is that the default attack/AC values for building your simple run of the mill combatant (high attack and high AC as they are called right now) are fairly high. The high AC isn't anything to write home about compared to what the best PCs can pull off but is still quite good, and the attack bonus is around a fighter's and very good. So if you call them "moderate" that seems off to players even more so because they are in fact pretty high. So the best thing was to call them high and then make it very clear that high was the default in those statistics.

Ooh, you're so smart Marky-poo! We're so blessed to have someone with such foresight, and who knows how to treat us right!

XOXO


2 people marked this as a favorite.
Pathfinder Roleplaying Game Superscriber

In my experience, monster AC basically falls into four categories: Stupid high, high, kinda bad, and immune to crits.

...Seriously, my players' disappointment when fighting oozes is always palpable. "I crit on a 5! ...But it's immune to crits..."

Good to have official confirmation that "high" is definitely the baseline. That matches what I've been doing with my homebrew monsters, mostly.


Pathfinder Pawns, Starfinder Roleplaying Game, Starfinder Society Subscriber; Pathfinder Roleplaying Game Superscriber

Is that just with AC/Attack, or is it true with other values as well, Mark?


Mark Seifter wrote:
The problem we found is that the default attack/AC values for building your simple run of the mill combatant (high attack and high AC as they are called right now) are fairly high. The high AC isn't anything to write home about compared to what the best PCs can pull off but is still quite good, and the attack bonus is around a fighter's and very good. So if you call them "moderate" that seems off to players even more so because they are in fact pretty high. So the best thing was to call them high and then make it very clear that high was the default in those statistics.

Yeah, I guess it would be a psychological problem to call the standard monster combatant - which comes equipped with "Fighter" to-hit and "Champion" AC (figure of speak) - moderate, because everybody that is not playing a Fighter or Champion (and even those) would be: Ok, if this is moderate, what are we...? ;)

Design Manager

1 person marked this as a favorite.
Ubertron_X wrote:
Mark Seifter wrote:
The problem we found is that the default attack/AC values for building your simple run of the mill combatant (high attack and high AC as they are called right now) are fairly high. The high AC isn't anything to write home about compared to what the best PCs can pull off but is still quite good, and the attack bonus is around a fighter's and very good. So if you call them "moderate" that seems off to players even more so because they are in fact pretty high. So the best thing was to call them high and then make it very clear that high was the default in those statistics.

Yeah, I guess it would be a psychological problem to call the standard monster combatant - which comes equipped with "Fighter" to-hit and "Champion" AC (figure of speak) - moderate, because everybody that is not playing a Fighter or Champion (and even those) would be: Ok, if this is moderate, what are we...? ;)

Champion/dex monk AC is much more similar to extreme AC, high monster AC is more like a fighter's (though not always as good as a fighter's, monsters use a simple progression but PCs are more nuanced than that), but you're exactly on point for fighters and to-hit. It feels wrong for the fighter to-hit to be "moderate," it's a very high accuracy.

As for the use of other statistics, they each work as mentioned in the GMG monster creation guide.


Ravingdork wrote:
Is that just with AC/Attack, or is it true with other values as well, Mark?

Its hard to speak about saves, as every monster has 3 and they tend to feature both a High and a Low. Skills can range all over as well, for the same reasons.

Perception is Moderate normative, though.

The save-DC of monster effects is High normative again (though if a monster has more than one way to generate a saving throw, they're not always all the same, but its highest is usually a High DC).

More specific:

Spoiler:

Note: This is only about 95% complete (only Bestiary 1). I'm about halfway through the level 1 creatures, and there are still some missing critters scattered about. Things like ponies, riding dogs, and metallic dragons.

"bnk" is "bonkers" and represents a value as much above Extreme as Extreme is above High. Ter(rible) and Worst are similarly lower-than-low.

Note also that these groupings do not all sum to the same number, as swarms do not have an attack bonus, lots of monsters don't have abilities, and stealth is not always a listed skill. Those all fall under a "none" category, which is not shown here.

ATK
wrst: 0
ter: 0
low: 7
mod: 96
high: 237
ext: 34
bnk: 6

AC
wrst: 1
ter: 2
low: 33
mod: 81
high: 236
ext: 32
bnk: 0

ABILITY_DC
wrst: 0
ter: 2
low: 1
mod: 27
high: 218
ext: 28
bnk: 17

PERCEPT
wrst: 1
ter: 18
low: 60
mod: 242
high: 60
ext: 3
bnk: 1

FORT
wrst: 0
ter: 7
low: 59
mod: 156
high: 145
ext: 17
bnk: 1

REFX
wrst: 2
ter: 25
low: 93
mod: 161
high: 96
ext: 5
bnk: 2

WILL
wrst: 1
ter: 19
low: 171
mod: 139
high: 50
ext: 4
bnk: 1

STEALTH
(The only skill I've tracked, due to being in opposition with PC Perception)
wrst: 3
ter: 11
low: 58
mod: 77
high: 86
ext: 27
bnk: 3


2 people marked this as a favorite.
thenobledrake wrote:
Claxon wrote:

It's kind of funny to me the mean and the mode of AC is what Paizo wrote as the high AC instead of the moderate AC values.

To me this indicates that individual designers are doing what they think feels right for the monster, but in so doing are artificially inflating the stats of monsters by accident.

It's more that their choice of naming pattern doesn't mesh with intuitive reading because they are very clear that what they named "high" is their default.

People read "high" and assume it means "higher than normal" rather than "normal."

Just like how some video games have a difficult setting but instead of labeling one of them "normal" and basing the other names off that, they have "hard" and "easy" and "nightmare" or some other evocative name and they treat "hard" as the default - but people intuit "hard" to mean, well, hard so they might think they want to play on easy because what they are really looking for is "normal."

Well, that's very misleading.


1 person marked this as a favorite.
Ubertron_X wrote:

Yeah, I guess it would be a psychological problem to call the standard monster combatant - which comes equipped with "Fighter" to-hit and "Champion" AC (figure of speak) - moderate, because everybody that is not playing a Fighter or Champion (and even those) would be: Ok, if this is moderate, what are we...? ;)

Well personally I'm of the opinion that we (the player characters) suck.

I know it's not a popular opinion to say but I personally wasn't have fun with PF2.

My group stopped playing (primarily because the GM who was running it just couldn't do it anymore due to life problems) but everyone in the group was happy to go back to Starfinder when I made the suggestion.


1 person marked this as a favorite.
Claxon wrote:
Well personally I'm of the opinion that we (the player characters) suck.

Someone took the "heroic" out of the "heroic fantasy."


Pathfinder Pawns, Starfinder Roleplaying Game, Starfinder Society Subscriber; Pathfinder Roleplaying Game Superscriber
Draco18s wrote:
Claxon wrote:
Well personally I'm of the opinion that we (the player characters) suck.
Someone took the "heroic" out of the "heroic fantasy."

Our group has played through Age of Ashes 1, the first third of Extinction Curse 1, and the Absalom Initiation, and we have yet to have a single encounter that didn't involve dropping one or more PCs.

I'm hoping higher level play will help add some of that heroism feel into the game.


2 people marked this as a favorite.
Claxon wrote:
Well, that's very misleading.

No, it isn't. The book (GMG) clearly states the necessary to set appropriate expectations - that's how I knew to say what I said in the post of mine you quoted.

Then Mark explained the thought process behind choosing the particular name for the label, which could have been anything and it still be completely clear which column on the table held the 'standard' value.

If anyone is "mislead" it is by their own mind not adjusting for context.

Like how it's not misleading that common coffee size names "tall" and "grande" both mean larger than normal in some other context because context has been properly established to inform that those are actually the smaller end of the size range.


thenobledrake wrote:
Like how it's not misleading that common coffee size names "tall" and "grande" both mean larger than normal in some other context because context has been properly established to inform that those are actually the smaller end of the size range.

Somewhere, years ago, I ran into this gem:

"...and I'd like a small soda."
"We don't have small sir, only Medium and Large."
"Yes you do: you have two sizes; one of them is smaller than the other."

(And by the way, for reference, Starbucks does have a small. Its called a "short" and is 8oz to the tall's 12)


1 person marked this as a favorite.
Ravingdork wrote:
Draco18s wrote:
Claxon wrote:
Well personally I'm of the opinion that we (the player characters) suck.
Someone took the "heroic" out of the "heroic fantasy."

Our group has played through Age of Ashes 1, the first third of Extinction Curse 1, and the Absalom Initiation, and we have yet to have a single encounter that didn't involve dropping one or more PCs.

I'm hoping higher level play will help add some of that heroism feel into the game.

What is your party composition? Age of Ashes 1 has some tough calls (I've run AoA 1-4 twice, and the whole thing once), and I'm only to the penultimate fight of Extinction curse 1, but you shouldn't have someone dropping every encounter.

Heck, Extinction Curse other than one fight early on has felt fairly heroic in general - a whole lot of smashing through most fights pretty easily.

I wonder if you're lacking a good composition for mitigating or healing damage, or something similar.

I can also more or less vouch for the fact that things become more forgiving as you level, and start getting reliable debuff's and action stealer's to consistently limit powerful boss enemies to 2 actions a round instead of a much more deadly 3 - options like concealment also will cut down on incoming damage to a huge degree. Both of my parties I've run for have gained a lot of ability to deal with challenges as they've leveled up.


I think there is another element relating to what heroic means to different people.

Succeeding against the odds is often how heroism is portrayed. Taking injuries and having tension is also very heroic, heck even dying is heroic (look to Boromir, or Ganalf vs the Balrog for classic heroic scenes. One against soldiers, one against a "boss" type encounter).

Although if someone is dropping every combat either luck isn't going your way, and or the tactics should probably change.

As said though, Extinction Curse is WAY more forgiving in the first book. The demon fights could be a real challenge if the GM doesn't run them as written (with heavy player facing advantages) or the party decides to just sit and trade blows instead of making use of their huge advantages.

I like it as a system introduction way more than Hellknight Hill.


thenobledrake wrote:
Claxon wrote:
Well, that's very misleading.

No, it isn't. The book (GMG) clearly states the necessary to set appropriate expectations - that's how I knew to say what I said in the post of mine you quoted.

Then Mark explained the thought process behind choosing the particular name for the label, which could have been anything and it still be completely clear which column on the table held the 'standard' value.

If anyone is "mislead" it is by their own mind not adjusting for context.

Like how it's not misleading that common coffee size names "tall" and "grande" both mean larger than normal in some other context because context has been properly established to inform that those are actually the smaller end of the size range.

A player shouldn't have to read the GMG to get this information and labeling something as high when it's actually the standard is to me the definition of misleading.

Personally I view the comparison to PC abilities not relevant because this is a table for NPCs/monsters.

Though the fact that AC and to-hit values for an equal CR monster is higher than PC values are quite illustrative to the overall poor personal experience I had with the game.


7 people marked this as a favorite.
Claxon wrote:
A player shouldn't have to read the GMG to get this information and labeling something as high when it's actually the standard is to me the definition of misleading.

Here's the thing though: if a player hasn't read the GMG, they don't know the default AC for monsters is labelled as "high" in the first place - so they won't have been mislead by that.

The only way to be "mislead" is by reading the GMG selectively - check the table, but not the accompanying "here's what's standard" - and then arbitrarily select one of the words from the table (none of which are "normal" "standard" "default" or anything synonymous with those) to believe is the default.

Which is, as I said earlier, anyone that is "mislead" having mislead them self - not been bamboozled by the clearly written text in the book.


High is higher than moderate, which is higher than low, all of which are higher than extreme. Note that literally none of these are named "normal" or "standard". As Mark noted, people felt that the default AC felt High... so they called it High.


"High" in this case is already explained - its High compared to the majority of player characters, who are not Fighters.

That is the "agreed upon set point" you refer to.

Yes, its not what is initially expected- which is why the documentation explains that High is the default standard.

But its totally valid in the context provided.


I understood it, after thenobledrake's post about it.

I just think it's wrong and shouldn't be labeled that way.

You shouldn't use player character values as a comparison point.

It's simply a point of disagreement, one on which I will not change my mind.

It is possible to understand someone's position but disagree with it.

Edit: To clarify, that applies to my position and to yours.

In my opinion the better way to have labeled the columns would have been something like: Terrible, Low, Standard, High. Because the labeling inherently specifies what's the expected value.

Calling it "Low, Moderate, High, Extreme" is basically done to make people feel better about the fact that the best a PC character can do is only equal to what an average monster can do.

It honestly feels like they went out of their way to avoid using the words average, standard, or default simply to avoid the direct comparison to player character values.


Claxon wrote:
Not true at all. The way I've learned about this stuff is people posting and talking about it here on the forums.

Then what has happened is that other people - not the book - have mislead you.

And those people, unless their intentionally misrepresenting what is in the book, have done the self-misleading selective reading I mentioned.

Yet at no point is the book misleading anyone, despite the names chosen for the categories not being whatever you would have preferred for them to be.

They can call it low, moderate, high, and extreme. They can call it like it, love it, and gotta have it. They can call it small, medium, and large. They could even call it medium, large, and extra large - but none of that is misleading unless they say which value is supposed to be the default assumed value and then it isn't that value.


It's less that you "disagree" and more that you're claiming Paizo is misleading people for... some reason. You care way too much about a single word's definition.


Setting aside the chosen nomenclature, the better question I guess would be how easy is it for a typical PC to hit an equal level fighter? (Since that is where your typical monster will be) Also, how easy is it for the typical fighter to hit an equal level PC?

Answering those questions and then placing them in context of action economy will go a long way toward figuring out if the system feels "right". I will agree that giving slightly higher values to the monster (compared to the typical PC) can be fine given the action economy (PCs don't need to hit as often as the monster because the PCs get more attacks/actions). But, in doing that, you also have to balance the damage and hit points. Obviously all the other abilities have to be considered too, but you get the gist. I won't pretend to have the answers to those questions yet though since I don't have enough miles in the system yet.

As to the feelings of the game being overly hard, I do have to admit that one of the bigger issues is simply that PF2 is much different than PF1 in terms of what is expected of the PCs/players. You need different tactics both in and out of combat. That's not an inherently bad thing, but it is the kind of thing that will create a learning curve after so many years of PF1 experience.

As a player in Extinction Curse, still very early on, I would say that my experience has been a bit mixed. The bard was insta-killed by the water mephit due to massive damage. Sure, it was a lucky crit, though there was a 15% chance to crit. Even a non crit on Acid Arrow though would, on average have put the bard the verge of unconscious with the average persistent damage knocking him out if not healed before it. Likewise, the cockatrice stoned the fighter. Some bad luck on the fighter's part as the fighter had a 45% chance to make the save each time. Other party members would have had an even lower chance. The ruffians in the first show knocked two PCs out as well. Again, there was some bad luck in there as diplomacy and demoralize rolls both failed (in order) and then there were some misses as the party was trying to do non-lethal. The first mini boss (druid I'm guessing?) had an incredibly high (by comparison) Will save which hurt the spellcasters, though she went down fairly easily to melee attacks. Granted, spell casters should be prepared to target all the defenses, but that's a lot harder to do at level 1.

All that said, there were also definitely some bad tactical choices made and certainly some bad die rolls as well. That happens. The druid was actually one of the easiest encounters because she did go down easy to melee attacks. The snakes were largely easy to deal with, though their poison did cause the bard problems (requiring the use of both of his 1st level spells to stay upright). I don't necessarily believe that the system is too difficult, but it does require a much different approach. That's not bad, but it will create problems early on for a lot of players. Just as an example, the fighter wanted to just swing three times in combat because he could! The idea of tripping or demoralizing never crossed his mind because of PF1 thinking where if you weren't really specced for it, you probably weren't going to do well AND you might take an AoO for your trouble. That's something you learn quickly of course, but early on it can create issues. Not the system's fault per se, but rewiring the brain after over a decade of PF1 does take some time.


thenobledrake wrote:
Claxon wrote:
Not true at all. The way I've learned about this stuff is people posting and talking about it here on the forums.

Then what has happened is that other people - not the book - have mislead you.

And those people, unless their intentionally misrepresenting what is in the book, have done the self-misleading selective reading I mentioned.

Yet at no point is the book misleading anyone, despite the names chosen for the categories not being whatever you would have preferred for them to be.

They can call it low, moderate, high, and extreme. They can call it like it, love it, and gotta have it. They can call it small, medium, and large. They could even call it medium, large, and extra large - but none of that is misleading unless they say which value is supposed to be the default assumed value and then it isn't that value.

You are correct in so far as I should have gone to the GMG section and looked at it myself, but I actually disagree with your conclusion regardless because I think the terms Paizo has used are misleading.

Words have meaning. The word high means greater than. In this case of monster attributes one would assume greater than average, and be incorrect unless they pay attention to the line that specifies otherwise.

Using the word high is inherently misleading when talking about "this is the most common/default/average value that monsters of this CR will have".


Pathfinder Roleplaying Game Superscriber

The reason it makes sense to me, at least, is because monsters and PCs are built using the same math. A monster with a "moderate" attack roll will have a very similar bonus to a PC with a "moderate" attack roll - a rogue, for example. And a monster with a "high" attack roll will have a similar bonus to a PC with a "high" attack roll, such as a fighter.

The reason high attack rolls are so common is because the majority of monsters fill the same combat role as the fighter. If you had a party of mostly fighters, then the "high" attack roll would be the most common among the party.


Claxon wrote:


Using the word high is inherently misleading when talking about "this is the most common/default/average value that monsters of this CR will have".

So if I say "Barbarians typically have high strength" I am inherently misleading you about what to expect of a strength score for a barbarian?

Because yes, "words have meaning" - and those meanings are determined by context.

Design Manager

5 people marked this as a favorite.
MaxAstro wrote:

The reason it makes sense to me, at least, is because monsters and PCs are built using the same math. A monster with a "moderate" attack roll will have a very similar bonus to a PC with a "moderate" attack roll - a rogue, for example. And a monster with a "high" attack roll will have a similar bonus to a PC with a "high" attack roll, such as a fighter.

The reason high attack rolls are so common is because the majority of monsters fill the same combat role as the fighter. If you had a party of mostly fighters, then the "high" attack roll would be the most common among the party.

Combat-focused monsters usually have either high attack / high damage like a fighter or moderate attack / extreme damage like a barbarian (this one is the brute roadmap, it's all there in the GMG) because those are both simpler than making something techy like a ranger that is harder for a GM to play in multiples. That said, swinginess goes against the PCs in the long run and reliability is in their favor, so it's usually safer to face a foe that hits more often for less damage, meaning high/high fighter style is more common than moderate/extreme barbarian style. There's also monsters who are mainly spellcasters who are worse than either of those combinations, and of course, property runes and special abilities might eventually put the PC damage higher than the monster damage that I mention here as being akin to their damage (for instance, high damage at level 20 is 44, but a fighter with a greatsword using three low level property runes could wind up with more like 51.5 damage per hit even before taking into account any other special abilities they have).


1 person marked this as a favorite.
Adventure Path Charter Subscriber; Pathfinder Adventure, Card Game, Companion, Lost Omens, Rulebook Subscriber
Claxon wrote:
... To me this indicates that individual designers are doing what they think feels right for the monster ...

I would MUCH rather this scenario than any other, and despite there being formulas, I hope the final decisions always rests upon the creative ideals of the experienced designer.


1 person marked this as a favorite.
Gargs454 wrote:
<stuff I can't adequately trim down to a concise point without inadvertently misrepresenting, but which boils down to "some bad luck, several times">

This is part of why I think the balance is off and/or flawed. The approach was taken that for any individual task (not goal!) a 50% probability of failure was applied (modified in certain respects, but the design intent was to make things tight, and roughly 50-50 before considering circumstances).

The more tasks you add in order to achieve the goal, (irrespective of how long each task takes) the closer and closer the overall probability that you succeed at the goal approaches the probability of succeeding at a given task.

That is:
With a 50% baseline chance of success on a task, a campaign has a 50% chance of succeeding and a 50% chance that everyone dies. Its not luck or strategy at that point, its statistics. Remember, strategy can only help so much and in only so many situtations. A failed stealth roll means an extra combat? That's an extra chance to outright kill a PC. It may be a low chance, but... Do you feel lucky, punk?

Are we OK with half the people that play Extinction Curse (or whatever module) losing completely?

You can see that effect happening in every single fight Gargs's posted about. Every fight at least one PC was dropped unconscious or otherwise removed from the game. Maybe not dead-forever, but certainly in the status of "player is no longer playing the game and has gone to get pizza" removed from game.

Its happened to my group as well. Every fight, roughly one player was dropped unconscious.

"But dropped unconscious isn't lost the campaign!" you say. No, it isn't. But its a lot harder to get details on the number of PCs that complete a campaign vs. the ones that actually-died over the course of it.

Is half the players losing completely "heroic"? Is it more heroic for those that won because the other half lost? I'd argue no in both cases.

I'll bring up Kingdom Death Monster and Gloomhaven as an example. In general you don't lose a scenario in Gloomhaven unless you underestimate a new monster, misunderstand a rule, or have god-awful luck (but even that's rare*).

Kingdom Death Monster? No, every fight you lose the first time. And losing a fight means you permanently lose population. Your population goes down to zero and you have to start all over from scratch. Its an intentional design decision that you have to play the game, from zero, multiple times in order to win.

Do we want Pathfinder to be like Gloomhaven, or like Kingdom Death Monster?

*As an example I tried to teach some players Sentinels of the Multiverse and grabbed a "more beneficial to the players than the villain" environment (none are only candy, but some contain no candy at all). The first card it popped out shut down EVERY hero completely. There aren't even that many heroes IN THE GAME that could have dealt with it on their own, as few heroes have the ability to deal damage without playing any cards. There's only about a 1 in 30 chance of that card coming out first.

Mark Seifter wrote:
There's also monsters who are mainly spellcasters who are worse than either of those combinations...

I'd say "not really, Mark" except that SOME do EXIST, so you're not technically wrong. The problem I have with the statement is that there are SOME spellcasters who ARE ALSO FIGHTERS with regards to their AC, melee hit rate, and damage. The lich comes to mind.

Design Manager

3 people marked this as a favorite.

The lich is a level 12 monster with +24 to hit (moderate accuracy) 18 damage (worse than low damage), and 31 AC (between moderate and low). It is exactly an example of a spellcaster who is worse than either of those combinations.


Mark Seifter wrote:
The lich is a level 12 monster with +24 to hit (moderate accuracy) 18 damage (worse than low damage), and 31 AC (between moderate and low). It is exactly an example of a spellcaster who is worse than either of those combinations.

I must be thinking of a different monster. I know I ran across a spellcaster that was definitely "above the norm."

Let me see if I can dig it up.

Ah, here we go.

Treerazer I'll give a pass, its level 25. Its still pretty bonkers for its level, though.
(Extreme AC, Extreme attack, spellcaster with Moderate DC, Moderate HP, High? damage)

Nilith: Extreme AC, High attack, spellcaster with High DC, Moderate HP, Low (melee) damage.
Ancient Red Dragon: Extreme AC, High attack, spellcaster with High DC, High HP, High (melee) damage.

Nearest comparison, trying to quantify "well its 1 point below Extreme, but 2 points above High..." is harder to make short and unambiguous.

Basically, something that can basically always hit you, its hard to hit back, even if it doesn't do "that much damage" is not a fun fight. By any stretch. Its worse when its described as "being a spellcaster" because its weakness is "its inability to do a lot of damage at a time, but can't be killed" is not a weakness.

1 to 50 of 70 << first < prev | 1 | 2 | next > last >>
Community / Forums / Pathfinder / Pathfinder Second Edition / Advice / Probabilities and dice rolls All Messageboards

Want to post a reply? Sign in.