| 2 people marked this as a favorite. |
I have done some research and calculations based on the information in the rule set and online, and have discovered that there is a series of skill checks that scale in difficulty in such a way as to make them impractical at higher levels. I will outline the skills and how I came to these conclusions. As a disclaimer, Paizo has addressed some of these for Starship Combat however this has not been done for the general role playing rules.
The skills and associated checks are as follows:
Acrobatics
Escape vs restained by rope/bindings 20 + 1-1/2 * CR
Tumble
Through (threatened) 15 + 1-1/2 * CR
Through (direct) 20 + 1-1/2 * CR
Bluff
Feint 15 + 1-1/2 * CR
Diplomacy
Change Attitude 15 + 1-1/2 * CR
Engineering
Disable Equipment 15 + 1-1/2 * CR
Identify Tech
Common 5 + 1-1/2 * CR
Less Common 10 + 1-1/2 * CR
Rare 15 + 1-1/2 * CR
Repair Item 15 + 1-1/2 * CR
Intimidate
Bully 15 + 1-1/2 * CR
Demoralize 15 + 1-1/2 * CR
Mysticism
Identify Magic Item 15 + 1-1/2 * CR
Repair Item 15 + 1-1/2 * CR
Survival
Handle an animal 15 + 1-1/2 * CR
Rear a wild animal 15 + 1-1/2 * CR
These skills begin to scale to a point where it becomes unfeasible to expect success as a result. I define a reasonable chance of success as a 50% chance to succeed.
I have a set of numbers below demonstrating this line. There are a few things to remember for these numbers.
First off, regardless of difficulty there is a 5% chance to succeed due to a critical roll (natural 20).
Secondly, these 2 sets of numbers are based off of a linear equation. An increase of 1 point will always increase the chance of success by 5% up to a 95% chance due to a critical failure (natural 1).
Challenge Rating
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
DC Check per Challenge Rating
1 3 4 6 7 9 10 12 13 15 16 18 19 21 22 24 25 27 28 30
Class Skill (no ability bonus) 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Success Chance (20 + 1*1/2 * CR) 20% 15% 15% 10% 10% 5% 5% 0% 0% -5% -5% -10% -10% -15% -15% -20% -20% -25% -25% -30%
Success Chance (15 + 1*1/2 * CR) 45% 40% 40% 35% 35% 30% 30% 25% 25% 20% 20% 15% 15% 10% 10% 5% 5% 0% 0% -5%
Class Skill (ability score + 3) 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Success Chance (20 + 1*1/2 * CR) 35% 30% 30% 25% 25% 20% 20% 15% 15% 10% 10% 5% 5% 0% 0% -5% -5% -10% -10% -15%
Success Chance (15 + 1*1/2 * CR) 60% 55% 55% 50% 50% 45% 45% 40% 40% 35% 35% 30% 30% 25% 25% 20% 20% 15% 15% 10%
These are a bit hard to read due to the formatting, however the numbers should be accurate. It's easy to gauge the effect of feats and items which increase a skill. Every point increases the chance of success by 5%. By levels 7-10 things begin to fall off as keeping any sense of pace requires investment in feats and gear which specifically target the associated skill or ability score. Once you factor in the variety of skills and associated abilities making these checks at higher levels become improbable.
Unfortunately this makes some of the skills offered very unattractive.
A good example is Intimidate.
This is a class skill for the Envoy, Mystic, Operative, Solarian, and Soldier. It's primary ability score is Charisma. Unfortunately, outside of skill checks, Charisma has little combat relevance. For any class other than the Envoy and Solarian, it is unlikely that Charisma will be an ability score on which a player focuses with the possible exception of a socially oriented game campaign.
By level 6 this skill begins to fall off from a standard chance of success and requires the player devote resources in either character choice or itemization to keep it relevant.
To maintain even a 50% chance of success post lvl 6 you must find a way to increase Charisma to 18, an item to increase the skill itself, or an ability from an ally can increase your success chance.
I hope that Paizo will take a closer look at some of these skill checks and their difficulties and consider reviewing them. Many of the skills that these checks are associated with are defined by these checks.
Thank you for your time
Tumble for me is the big one I noticed things on. You reference Intimidate as well which is another noticable one.
Mainly, I am an Envoy, will a heavy Intimidate focus. Our Soldier is a big boy, and it should make sense he can be scary. You're right though, he's pretty much never going to hit the numbers without heavy Cha investment, where I in the early levels have been hard pressed to fail an Intimidate without rolling a natural 1.
For reference, my Intimidate is average +15 at level 3.
3 ranks + 3 class skill + 4 ability mod + 2 item bonus + 3 insight bonus = 15
I have Skill focus and Expertise, but they don't stack. So for now essentially the expertise die gets rolled and if it's less than 3 it means nothing. But it means on the high end I can be +18.
I would label myself as ver heavy investment though. By the end of the adventure at level 13, I should be something like +26 plus 1d8 for Expertise. Ballpark. That puts me for that level around...
Aiming for DC 34, above 60% success? The Expertise die isn't 5% per number as a D8 so it gets complicated. You can't really get more invested than my character though short of a racial modifier, so really 60% for pretty much full investment isn't very good.
Additional notes: those numbers, 15+(1.5×CR) and 20+(1.5×CR) are the numbers on Starship DCs after their adjustment, so the percentages directly parallel Starship combat difficulty. Made worse in one regard: Gunnery. The skill but #notaskill, which will always be BAB or Piloting ranks flat + Dex mod, giving it a significantly lower max cap.
Mm thoughtful and crunchy
Additional notes: those numbers, 15+(1.5×CR) and 20+(1.5×CR) are the numbers on Starship DCs after their adjustment, so the percentages directly parallel Starship combat difficulty. Made worse in one regard: Gunnery. The skill but #notaskill, which will always be BAB or Piloting ranks flat + Dex mod, giving it a significantly lower max cap.
Thank you for your response and affirmation. I was drawn to this research as I have created a Vesk melee oriented soldier and at lvl 7 found that I hadn't been using a skill though I had invested in it every level. When I realized that regardless of my investment I only had a 30% success chance against an equal level CR I was vexed that the system allowed this. This isn't the correct forum but I am also writing an analysis for Paizo to investigate melee combat in their game. (I wanted to play a melee class with it's high risk, high reward, play style however have found that the game's rule set is heavily focused on driving players to ranged combat.)
I also wanted to post the link to the official faq for Starfinder in case you have not had a chance to review the changes to Starship Combat DC's. It gives me hope that Paizo is listening although their response seems sluggish.
http://paizo.com/starfinder/faq
Happy gaming and again, thank you for your response.
I have a set of numbers below demonstrating this line. There are a few things to remember for these numbers.
First off, regardless of difficulty there is a 5% chance to succeed due to a critical roll (natural 20).
Secondly, these 2 sets of numbers are based off of a linear equation. An increase of 1 point will always increase the chance of success by 5% up to a 95% chance due to a critical failure (natural 1).
I'd like to point out this is not correct. There is no "critical" for a skill check. A 1 is not an automatic failure, nor is 20 an automatic success. Its quite possible to have a 100% or 0% chance of success.
Now to address your main comment, its a bit more complicated than simply saying a character with full ranks but no other bonuses should have a 50% success rate.
Lets consider the worst offender, level 20.
A level 20 character could realistically encounter anywhere from 16 CR 11's (An easy APL-1 encounter with 16 enemies, which gets 12+8), all the way up to a CR 23 (an epic APL+3 encounter with a solo boss).
For a 15 + 1.5xCR scaling, that represents DCs from 31 to 49.
Even if you GM never runs 16 enemies, they are likely to run groups of 4, making 4 CR 16s likely, which bumps the minimum DC from 31 to 39. Thats still a 10 point DC spread.
On the other hand, you could have a character with an 8 in stat they've never raised, non-class skill, but full ranks with a +19, all the way to optimized Diplomacy Envoy with 20 ranks + 3 Class + 1d8+4 (8.5 avg) insight + 2 Racial + 1 theme + 9 Charisma = 43.5 avg. That is a minimum of 21 point spread (24.5 on average).
Its impossible to not have one of those characters be at 100% or 0% chances on the same task.
Or a Hacker Operative with a Computers skill of 20 ranks + 3 Class + 6 insight + 2 racial + 1 theme + 9 Int = 41 and can take 10, for 51 every single time.
Those numbers come just from assumed character build choices. That is not including equipment/magic items which provide enhancement or circumstance bonuses. Or even the lowly aid another.
In Starship combat, you've also got access to computer nodes, which might be +5/+5/+5, or even a +10/+10. Speaking of which, the highest starship you should encounter is APL+1, which actually shifts skills affecting the enemy ship downwards relative to typical encounters.
If you make it so simply investing 20 ranks into a skill (and not even putting one of your 4 +2 bonuses you get every level) to a 50% chance at equal CR, you've got anyone who actually invests into a skill hitting automatic success in virtually every encounter.
That includes Soldiers. Its trivial to get a secondary stat up to 24 (+7) on a Soldier by level 20 (start 14->16->18->19->20 + 4 personal upgrade). Add in a skill focus feat (because Soldiers have 20 feats by level 20), and its sitting at +33. If +23 has 50% chance of success, that Soldier is hitting 100% chance.
So I see where you're coming from, but I turn the question around and ask, whats your suggestion? Simply changing the DC scaling is going to run into a lot of automatic successes for well built characters. Do we need to compress both the range of DCs that a party can face as well as possible achievable skill bonuses? Halve all insight bonuses? Apply only half of your stat modifier to skills?
A range of 10 on DCs (i.e. 39 to 49) for enemies and a range of 20 (i.e. +23 to +43.5), results in making it hard to make everyone happy.
In Starship combat, you've also got access to computer nodes, which might be +5/+5/+5, or even a +10/+10. Speaking of which, the highest starship you should encounter is APL+1, which actually shifts skills affecting the enemy ship downwards relative to typical encounters.
You are correct on your points. I would like to point one bit out for Starship combat though.
Starship combat removes item bonuses and other bonuses that do not specify specifically working in Starship combat, so your nodes in many cases will be making up your differences. Still good, but not quite as good as it seems.
Many of the Starship DCs also aren't based on your enemy's tier, but your own. All of those checks are going off a DC based on exactly your APL (unless your GM is keeping your ship under tier until you get to a port or some other reason) instead of your enemy's.
First of all, let me say thank you to everyone who has posted. I am grateful for all responses.
For Hiruma, I believe you are missing the point. You did correctly state that there are no critical successes or failures for skill checks. That being said, this actually increases the DC of each check by 5% across the board as I had calculated a flat 5% increase for a critical success but no critical failure.
Your post makes many assumptions but doesn't contradict my claims. You assume many variables that could account for success in these situations but let me state it in another way.
The Game mastering portion of the game says that a DC 15 + 1-1/2 CR check is considered "challenging". This means that for the subset of skills I listed almost every single check is challenging.
The problem lies in that these checks don't scale with personal growth for a character. It's easy to reverse engineer an example. Let's start with a lvl 20 challenging check (we don't need to suppose or assume anything about our characters ie.. class, gear, encounter level, theme etc..).
To succeed on an ability check with a CR of 20, that is considered a chellenge, you would need to roll a result of 45 (impossible with a D20). If you have no contributing factors to your skill other than the single point you have placed in to the class skill per level, you have a -10% chance of success. Every single point you contribute will increase the chance of success by whatever means you acquire it. If you use a skill with an ability score that has been raised to 24 (+7) you can increase your chance by 35% (+7 x 5). This makes the success possible but not probable, netting you a mere 25% success chance (75% failure chance). Now, for a "challenging" roll, that seems a bit too challenging IMO. Further, it requires a player to focus on that particular attribute to achieve a probability of success(25%).
This seems to me rather inane. If someone is a soldier who has devoted their time and effort (represented by skill points) in to a skill for 20 levels you would expect that they would have a better than average chance of success without the need for additional items to boost them to that chance. If my soldier expects to succeed at an Intimidate check, then he requires a source (theme, feats, gear, bonuses from allies etc..) to give him at least 12 more points of skill in Intimidate to make it even a statistical even chance(50%). To guarantee success we have to add a whopping 22 points of that skill to the soldier in some fashion. This is ridiculous. We could assume X and Y and build Z and so many other things to get him to that point, but the question is, should that be a consideration? Is the system built from the back end around the highest attainable results? Or is it built to suitably scale?
My response for the changing of the scale so that well built characters automatically succeed is ... and?? isn't that the point of a well built character? If a player is willing to devote such excessive means to a specific skill then isn't it reasonable to expect that a challenging roll for that character would still succeed in almost all situations. For someone who has devoted few points or scattered those points across multiple skills, then it remains that a challenging roll would still be a challenging roll. In my opinion it seems that the rules should reward players who decide they want to use the rules to make even challenging checks, easy for their character.
My proposition was simple. I would like Paizo to review these particular skill checks for these skills and perhaps provide a more realistic measurement for these checks. Not every Intimidate or handle animal check should be challenging (and yes I concede that a DM can modify these checks however in many cases DM's tend to scale upwards more often than downwards using the rules as a baseline.)
I've looked at possible changes and it seems to me that a 15 + CR is a more meaningful measurement. It makes it much easier for DM's to calculate the numbers on the fly and then adjust the Base DC upwards or downwards to compensate.
Here is an example. Let's work with a lvl 1 character. At level 1, if you attempt a DC 15 + CR check this requires a roll of 16 or higher. Taking in to account only the skill ranks a player devotes (and nothing else, not even attribute bonus) then that player has a 40% chance of success. That seems challenging to me. You fail regardless of what you do 60% of the time even though you are skilled. If you aren't skilled, then your chance of success is even worse. 20% at best. Not impossible, but very improbable. If you are skilled and naturally adept (assume a +2 in your associated attribute bonus) you have only ended up even money. You have a 50% success chance... roll a 10 or more.. your are set.. anything less, you fail. That seems reasonable to me and rewards players who devote their resources to increasing a skill to be "100%" successful.
This same debate is happening in the Pathfinder Playtest.
Without going into too much detail, skills are bounded to the Proficiency system this resulted in two leading arguements.
On the one side, is the arguement that untrained skills are at the lowest possible level-3. People cannot be "bad" at a skill as high level characters have similar boosts in their worst stats to mid-level characters with max stats.
On the other side is the complaint that this binding extends to DCs as well. A completely optimized character hovers around the 50% success rate mark without item, circumstance, or other bonuses. Plus with four degrees of success and natty 1s and 20s return things get iffy.
Neither side can be quelled, but it is a proposed solution to the DC gap problem Starfinder suffers from, since the difference between the base and the max isn't as outrageous.
First of all, let me say thank you to everyone who has posted. I am grateful for all responses.
For Hiruma, I believe you are missing the point. You did correctly state that there are no critical successes or failures for skill checks. That being said, this actually increases the DC of each check by 5% across the board as I had calculated a flat 5% increase for a critical success but no critical failure.
Automatic success on a 20 only affects rolls which had no chance normally to succeed. If a 20 succeeds normally, it doesn't affect the odds. So its not really an across the board increase. And in any case, your charts were correct.
Your post makes many assumptions but doesn't contradict my claims. You assume many variables that could account for success in these situations but let me state it in another way.
I guess I was trying to frame it in the larger context of the rest of the skill parameter space. I agree, I made assumptions about possible builds. However, you're also making assumptions about possible builds and making choices about which of those builds to balance against. In this case, something like a Charisma 10 soldier (or maybe 16 Charisma at 20th) trying to Intimidate.
I totally understand what you're getting at, full skill ranks at high levels by themselves don't do anything against your typical challenging skill check, which according to the core rule book most of them would seem to be challenging. And that non-specialized characters effectively get worse relative to their NPC peers as they level up, rather than better.
This seems to me rather inane. If someone is a soldier who has devoted their time and effort (represented by skill points) in to a skill for 20 levels you would expect that they would have a better than average chance of success without the need for additional items to boost them to that chance. If my soldier expects to succeed at an Intimidate check, then he requires a source (theme, feats, gear, bonuses from allies etc..) to give him at least 12 more points of skill in Intimidate to make it even a statistical even chance(50%).
You're right, they'd need to get +12 to have 50/50. I'll note you can build such a Soldier while still maintaining maximum Strength (Gladiator Lashunta, skill focus Intimidate, +7 Charisma mod by starting Str 18/Cha 14, and putting in a +4 personal augmentation), gets you +13. Which, is your point I think, in that it requires a lot of specific choices to boost that skill on the part of the Soldier, rather than simply spending the skill ranks.
On the other hand, some classes whose entire point is skills can get even larger bonuses than +12 by 20th without even trying. Envoys are likely to have a +8 (starting 14 or 16) or +9 (starting 18) charisma modifier by 20th. They also get 1d8+4 insight bonuses to a bunch of skills (with Intimidate on that list). I figure a typical Envoy build with Intimidate as one of its expertise skills is going to be around +13 minimum (more like 17.5 on average due to 1d8). That is minimal investment for an Envoy. They need Charisma to power their abilities, so they're going to have it high. And you can't use skill expertise on anything but skills.
For better or for worse, the developers decided to write the core rule book scaling against the skillful classes (i.e. Envoy, Operative) and other classes in their niche (i.e. Mechanic with Computers/Engineering, Technomancer in Engineering, Mystic in Mysticism), resulting in a 1.5xCR scaling. Soldiers, who shine in combat, are bad at skills unless they make an effort to focus on skills.
If the developers did scale it to a no stat investment Soldier (i.e. using 15+CR), then adventuring parties as a whole would be succeeding on many challenging rolls 80-85% of the time (or more with aid another/circumstance/equipment or by rolling separately) by 5th or 7th level, and pretty much auto-succeeding by 11th.
Which of those choices results in the better overall situation for the game is going to be an opinion that varies from person to person, I think. Given the Starship DCs were shifted to mostly 1.5xCR scaling in the FAQ, I think that scaling factor is the one the developers are happiest with, so I don't think you'll see much official change on that front.
One of the other points I was trying to make before was a character isn't always going to be facing exactly equal CR or challenging rolls. I admit this is tangential to your main point, but I feel its worth noting in this discussion.
There are many uses for skills which are not just the enumerated ones in the core rulebook. Let me present the skill check DCs of one Starfinder scenerio from Paizo I happen to have:
Levels 1-2:
DC 10
DC 12 x6
DC 13
DC 14 x2
DC 15 x7
DC 16 x6
DC 18 x5
DC 20 x2
DC 21
Roughly half are DC 15 and easier, and roughly half are harder. So more than half the skill checks were easier than 15+1.5xCR standard.
Levels 3-4:
DC 10
DC 12
DC 14 x2
DC 15 x4
DC 16
DC 17
DC 18 x6
DC 19 x2
DC 20 x3
DC 21 x3
DC 22 x2
DC 24 x2
Roughly half the checks are DC 18 and easier, and half are harder. So again, more than half the skill checks were easier than 15+1.5xCR.
So in overall adventure design, there should be both easier and harder enemies, as well as easier and harder skill checks. On those easier checks, a minimum investment character can still contribute with 50/50 or better odds.
And there are also some skill checks there that even with a specialist, the team might have 50/50 odds, and thus aren't necessarily expected to succeed (but just make their lives harder if they fail).
I'll note the single most consequential skill check in that entire scenario was DC 13 (or DC 15 for high tier), and you could take 10 on it.
Thank you for the thoughtful and well written response!
I guess I was trying to frame it in the larger context of the rest of the skill parameter space. I agree, I made assumptions about possible builds. However, you're also making assumptions about possible builds and making choices about which of those builds to balance against. In this case, something like a Charisma 10 soldier (or maybe 16 Charisma at 20th) trying to Intimidate.
I totally understand what you're getting at, full skill ranks at high levels by themselves don't do anything against your typical challenging skill check, which according to the core rule book most of them would seem to be challenging. And that non-specialized characters effectively get worse relative to their NPC peers as they level up, rather than better.
This is exactly my concern and well spoken. It seems to me that this is a problem with the system. Almost as if it was engineered backwards to ensure that classes who are devoted to skill checks will always face some level of challenge and never quite hit that "perfect" success option.
That being said, as I read your post I thought about the class designs and it may be that classes are designed in a way that their natural inclinations lean them towards specific skills and they any other skills should require additional resources from the class. I don't know, since I haven't designed the game, but it's a reasonable thought.
Thanks for your response and post!