Ability Roll > Point Buy?

106 posts / 0 new
Last post
Somehow I find it pretty odd, that you can roll for abilities of 18 but are only allowed to buy up to stats of 15. That means a "rolled" human can go for one ability of 21 value, while with point buy it's not possible.

I don't see any reason in this, especially as no rule forbids you to reroll your abilities if you don't like the result.
If people can roll for 18s + racial, class, etc bonus, why shouldn't they buy them as well?
Huh... If no rule forbids it that has got to be the most house ruled thing in D&D...

I think the idea is that rolling gives a nice bell curve. Point buy sits just below the top of the bell curve as a safe but not optimal choice where as rolling gives the ability to get lucky and get above average stats but also the ability to end up messing yourself up and having very poor stats. 
What you say even holds true also with the previous point buy models. So I do not understand, why they even further screw the point buy by denying access to ability scores higher than 15.

Further the point buy system has always had one advantage over the roll. In a 6-player party there will usually be at least one guy being jealous over other players if he rolls worse. The point buy eliminates that problem.

And you do not need a house rule to reroll ALL your ability rolls completely if you don't like them.

Point buy has always been a more stable, but subpar option to rolling. But the amount point buy allows is all based what the "average" stats they expect a character to have. A character in 5e is considered to have lower stats on average than a 4e character, so the point buy has been adjusted accordingly.

Also on your comment about balance, this is 100% correct and why I use point buy almost every time. But characters should not be able to use point buy and get optimal stats. That makes rolling a completely inferior option and no one would ever choose it. You can't have your cake and eat it too I'm afraid. 
My two copper.
If so, then at least the methods of rolling should be a little more comfortable, like they did with Pathfinder.
There you got similar options as in D&D 3.5 and you also have got the option to throw 24d6, remove the lowest 6 dices and with the remaining 18 you determine your ability scores. You can combine them how you like, but you can only combine 3 dices per ability score and each dice out of those 18 can only be used used once.

For Epic Fantasy they even suggest rolling 28 dices, so you can remove the lowest 10 before determining your ability scores.
Anybody know the average to 4D6, drop the lowest for scores?  What if the average for rolling is significantly higher than the array or point buy systems?
Anybody know the average to 4D6, drop the lowest for scores?  What if the average for rolling is significantly higher than the array or point buy systems?



I seem to recall they were equal in 4e.. .see 5e in a fit of nostalgia wants to encourage random goo.
  Creative Character Build Collection and The Magic of King's and Heros  also Can Martial Characters Fly? 

Improvisation in 4e: Fave 4E Improvisations - also Wrecans Guides to improvisation beyond page 42
The Non-combatant Adventurer (aka Princess build Warlord or LazyLord)
Reality is unrealistic - and even monkeys protest unfairness
Reflavoring the Fighter : The Wizard : The Swordmage - Creative Character Collection: Bloodwright (Darksun Character) 

At full hit points and still wounded to incapacitation? you are playing 1e.
By virtue of being a player your characters are the protagonists in a heroic fantasy game even at level one
"Wizards and Warriors need abilities with explicit effects for opposite reasons. With the wizard its because you need to create artificial limits on them, they have no natural ones and for the Warrior you need to grant permission to do awesome."

 

Anybody know the average to 4D6, drop the lowest for scores?  What if the average for rolling is significantly higher than the array or point buy systems?

The average of 4d6 drop 1 is ~12.24. (3d6 is 10.5).

Dwarves invented beer so they could toast to their axes. Dwarves invented axes to kill people and take their beer. Swanmay Syndrome: Despite the percentages given in the Monster Manual, in reality 100% of groups of swans contain a Swanmay, because otherwise the DM would not have put any swans in the game.
Anybody know the average to 4D6, drop the lowest for scores?  What if the average for rolling is significantly higher than the array or point buy systems?


I did not mean the "6x4d6, drop the lowest of scores" variant
I meant the "24/28x6, drop the lowest of scores and then combine the dices as you see fit" variant.

That really makes a difference as you can select the dices as you like. With the old variant you could have rolled for example 5 times a 6 within 4 different 4d6 rolls out of those 6x4d6, so they would be in 4 different abilites. With the other you could combine them and put them into only 2 abilities (3x6 (18) and 2x6+x (12+x)).

I think that would encourage people more to choose rolling as they still have some more influence over their abilities, yet they still need some luck.
Putting a whole bunch of different rolling methods into the playtest takes up valuable space I would prefer to be reserved for anything else.
Putting a whole bunch of different rolling methods into the playtest takes up valuable space I would prefer to be reserved for anything else.


If space is your only concern, I am quite happy with. A .pdf is enlarged quite easily and without a huge demand of memory.
Even where people aren't cheating, most groups that roll use fairly generous reroll methods. I think people aren't always aware of the effect that things like "reroll if your modifiers don't sum to a certain amount" have on the expected values of rolls. There's nothing wrong with wanting to use generous rolling methods, of course, but you don't have to do much before you push them to a point where they tend to generally produce better results than point buy pretty reliably.
Dwarves invented beer so they could toast to their axes. Dwarves invented axes to kill people and take their beer. Swanmay Syndrome: Despite the percentages given in the Monster Manual, in reality 100% of groups of swans contain a Swanmay, because otherwise the DM would not have put any swans in the game.
Anybody know the average to 4D6, drop the lowest for scores?  What if the average for rolling is significantly higher than the array or point buy systems?


I did not mean the "6x4d6, drop the lowest of scores" variant
I meant the "24/28x6, drop the lowest of scores and then combine the dices as you see fit" variant.

The average of the top 18 dice of 24d6 is 75.63, so the average attribute is 12.605.

For the top 18 of 28d6 the average is 81.23 so the average attribute is 13.538.

However the average of the top *6* dice is is 33.63 or 34.28, which means it's almost guaranteed that you can get at least one 18 and you have a nearly-even chance of getting two (assuming you want them).

On the other end, the average dump stat is 7.26 or 8.95. Where with 18d6 you have a barely better than even chance of getting 4 or higher on your 3 lowest dice.

Simply rolling 3d6 six times your dump stat will average 6.77, and your high stat will average 14.23 (I think - if I figured out how to do that right on anydice.com ).

"The world does not work the way you have been taught it does. We are not real as such; we exist within The Story. Unfortunately for you, you have inherited a condition from your mother known as Primary Protagonist Syndrome, which means The Story is interested in you. It will find you, and if you are not ready for the narrative strands it will throw at you..." - from Footloose
Huh... If no rule forbids it that has got to be the most house ruled thing in D&D...

I think the idea is that rolling gives a nice bell curve. Point buy sits just below the top of the bell curve as a safe but not optimal choice where as rolling gives the ability to get lucky and get above average stats but also the ability to end up messing yourself up and having very poor stats. 



Point-buy can do a bell curve if you assume the dump stat from the start and then allow no lowering of stats.  The closest approximation to the average rolled array is 9, 10, 12, 13, 14, 16.  But lets say the first thing we want to do is dump.  Anything less than 6 is absurd for an adventurer, so lets shift the array to that bottom end.  That gives us a before-points starting array of 6, 7, 9, 10, 11, 13 (and an 18-point difference that we need to make up).

Now, offer your players that array and this point-buying rule: "Spend two points raising a stat towards 12, and then a third point raising whatever you want.  Do this six times, but never raise any stat over 18."

This allows for plenty of flexibility, but even the extremes of this buying system are pushed into a bell curve.  Here are some examples of resulting arrays:


07, 12, 12, 12, 13, 18
08, 12, 12, 12, 12, 18
10, 11, 11, 11, 13, 18
11, 11, 11, 11, 12, 18
12, 12, 12, 12, 13, 13
08, 11, 12, 12, 15, 16
08, 12, 12, 12, 15, 15
10, 10, 12, 14, 14, 14
09, 10, 12, 14, 14, 15
12, 12, 12, 12, 13, 13

And of course...

09, 10, 12, 13, 14, 16


(You could alternatively start with 5, 6, 8, 9, 10, 12; use 24 points; and thereby make a result of 6, 12, 12, 13, 13, 18 possible.  But that presents a lot more opportunity for human calculation error.  To mitigate error even further and force a tighter range of possibilities, you might use 7, 8, 10, 11, 12, 13 and spread 12 points; but that leaves only a single choice of array to get an 18.  So I think the solution I presented above is the happy medium for this type of point-buy system.)

(The simplest alternative of all would be a starting array of 8, 9, 11, 12, 13, 14 and 6 points, with no limits on how to spend them other than a max stat of 18.  But that doesn't so much incentivize a bell curve as put us in a tight range and incentivize leaving it but restricting how much one can.  It's an opposite approach altogether, then, but still produces fairly reasonable results.)
Could be they're making sure everything works with 15s.
As I understand it, the argument for rolling is that with point buy, you can fully optimize, and therefore "all the fighters feel the same". The complaint about rolling is that since you only roll once, rolling can lead to one person feeling permanently gimped. It feels like the first one could be fixed with interesting trade offs so there's no optimal solution.

Given the choice, I'd design the system do that point buy is always better. You generally don't want to permanently punish a newby for feeling lucky, so have the better default be dice. Also, you want your system to discourage retool through suicide, and if your system is balanced to average roll, rolling a char and killing him if he isn't average becomes tempting.

However, it doesn't much matter since those who like balance will demand everyone point buy, so they don't Care what the average roll is. At least, if I dm, that'll be my plan.
The average roll matters even for a point-buy system because that's what the rest of the game will have been optimally balanced off of.  There should be plenty of leeway, but the game should be most balanced and most easily balanced for GMs when everyone is right at the average roll.  A point-buy system that is not constructed towards the average roll will therefore be inherently unbalanced.
I don't care if the point buy is on average sub-par compared to rolling. My group will still use point buy because we'd rather everybody suck equally and the DM have an easy time adjusting minorly to that than half the party be awesome and the other half suck horrifically and the DM have a more difficult time adjusting to that.

Why, yes, as a matter of fact I am the Unfailing Arbiter of All That Is Good Design (Even More So Than The Actual Developers) TM Speaking of things that were badly designed, please check out this thread for my Minotaur fix. What have the critics said, you ask? "If any of my players ask to play a Minotaur, I'm definitely offering this as an alternative to the official version." - EmpactWB "If I ever feel like playing a Minotaur I'll know where to look!" - Undrave "WoTC if you are reading this - please take this guy's advice." - Ferol_Debtor_of_Torm "Really full of win. A minotaur that is actually attractive for more than just melee classes." - Cpt_Micha Also, check out my recent GENASI variant! If you've ever wished that your Fire Genasi could actually set stuff on fire, your Water Genasi could actually swim, or your Wind Genasi could at least glide, then look no further. Finally, check out my OPTIONS FOR EVERYONE article, an effort to give unique support to the races that WotC keeps forgetting about. Includes new racial feature options for the Changeling, Deva, Githzerai, Gnoll, Gnome, Goliath, Half-Orc, Kalashtar, Minotaur, Shadar-Kai, Thri-Kreen, Warforged and more!
With just a choice, I'd say point buy is fairer than rolling.  However, one could allow a player to roll all stats using the 4d6 drop lowest method "X" number of times  With the sets the player rolled, they would choose their best set and compare it to what would be offered by the appropriate point buy method for the campaign(for high powered, difficult campaigns, the DM may assign a higher value than is regarded as standard for point buy, for instance).  They can either stick with their best set or opt for the point buy. 

The drawback is that someone may still roll really good and outclass the point buy.  Still, it eliminates the greatest potential disparity of someone rolling low and someone rolling high on stats.  It also allows players who want to roll stats to do so and have a safety net that they can fall into.  I am a fan of higher powered campaigns though(especially since I like to keep magic items very rare).

That said, I don't really feel that there is anything wrong with using only point buy
As I understand it, the argument for rolling is that with point buy, you can fully optimize, and therefore "all the fighters feel the same".


This just depends on how the game is designed. If the design leads to one distribution of stats to be clearly optimal, then the result will be cookie-cutter characters. But if the game is designed to reward a variety of builds, you will get variety. It's no different than any other aspect of the game. If there's one spell/attack/skill/whatever that is clearly superior to all others, then everyone will use it. If everything is pretty much balanced, that won't be the case.

It seems to me that if people want some randomness, the best thing to do is just come up with a bunch of arrays using the point buy system and then roll to select one of them. That provides variety and interesting builds but makes sure that everyone is in the same ballpark.
I think they should go to a point-buy only system, and they should fix the point-buy to reflect the rest of the system's balance.  If a roll is balanced, then it can be replicated using a balanced point-buy system; therefore, the point-buy system encapsulates the roll system.  The opposite cannot be said.  So clearly point-buy should have precedence.
I think they should go to a point-buy only system, and they should fix the point-buy to reflect the rest of the system's balance.  If a roll is balanced, then it can be replicated using a balanced point-buy system; therefore, the point-buy system encapsulates the roll system.  The opposite cannot be said.  So clearly point-buy should have precedence.



Clearly no.  
Multiple options is how it's going to be and each gaming table decides for themselves.

For anyone interested tons of sites have already done all the number crunching for years about probability of rolling dice.

here's one site.
 www.d20source.com/2010/05/what-are-the-o...
Also, you want your system to discourage retool through suicide

How?
The only obvious ways to do it are DM fiat or some sort of baked in inherited punishment, neither of which are going to any good to retain players who roll "Mr. Nines" on their first guys.

Quite frankly, any DM who insists I play an awful pile of numbers is a DM who isn't going to see me at his table next weekend.  Similarly, any RAW of a -1 or something each time Bob[number] jumps off a cliff so I can roll a new guy is an extra $40-120 in my pocket to invest in videogames that don't care if I spend three hours humping a RNG until I'm happy.

I rolled 30 sets of stats after looking at this thread. Of all of those groups I only got three that failed to get multiple 15s or a 16 or higher. Half of those generated an 8 or lower. (I had 2 7s)

I generally let players roll three sets of numbers which will guarantee that they will never have weak characters.

The only argument I can relate to where rolling is concerned is for those who prefer their player's characters not have really good scores.

This may not be scientific or remotely apply to the law of averages in any measurable way but it is consistent with my experience playing this game for a very long time. I won't argue with a player who wants to use the array or point buy, but I won't put up with him moaning because the guy next to him legitimately rolled two high numbers and has a much higher average score than he does.
A stat array of 1 1 1 1 1 1 would be stable, and aptly demonstrates that stability is not a benefit unless it is stability about the average.

ARA: 09 10 12 13 14 16
CPB: 09 10 12  12 14 15

Not only does the current point-buy take away flexibility, it also takes away your ability to get an array equivalent to the average.

The reason for this is pretty obvious, in that whoever designed Next's current point-buy system forgot to account for the variance of the roll.  They just assumed that if the point-buy can produce an array of 12 12 12 13 13 13, then it is hitting the average.  But that is not the true average array.  On average, a roll will be have more variance than that, and therefore will produce higher highs and lower lows, just like most players would prefer for optimization.  A well-designed point-buy system doesn't need to be able to hit 18, necessarily, but it does at least need to be able to construct the true average rolled array.  (Of course, an ideal point-buy system distribution would be a perfect reflection of the roll distribution; and so if it is possible to improve on that accuracy with simple rules, it would be silly not to.)
(Of course, an ideal point-buy system distribution would be a perfect reflection of the roll distribution; and so if it is possible to improve on that accuracy with simple rules, it would be silly not to.)

For one thing, it needs the ability to go below 8.


Here's a zero-sum pointbuy I math'd together a few months ago:

Start with the array of 11/11/11/10/10/10
±2 = ±1 point
±4 = ±2 points
±6 = ±5 points
±7 = ±9 points

Point buy should always be inferior to rolls. otherwise no one would ever roll, and it would be a punishment for dms to ask their players to roll. This way it's a trade off for a little power for predictability.
My two copper.
Point buy should always be inferior to rolls.

Rolls already have the potential to go well above a standard pointbuy (or well below).  No method should be better or worse than any other method, just because.
 
Anybody know the average to 4D6, drop the lowest for scores?  What if the average for rolling is significantly higher than the array or point buy systems?


It is significantly higher as most of my characters have demonstrated over the years. If you roll you tend to have higher and better scores across the board than if you use point buy. This is even more so true now that you can roll 3 points higher than you can point buy. 

 
If so, then at least the methods of rolling should be a little more comfortable, like they did with Pathfinder.
There you got similar options as in D&D 3.5 and you also have got the option to throw 24d6, remove the lowest 6 dices and with the remaining 18 you determine your ability scores. You can combine them how you like, but you can only combine 3 dices per ability score and each dice out of those 18 can only be used used once.

For Epic Fantasy they even suggest rolling 28 dices, so you can remove the lowest 10 before determining your ability scores.


Maybe I'm just stupid or crazy, but this just sounds like a point buy without any balance and with more die rolling. I don't really see the appeal?
A few guidelines for using the internet: 1. Mentally add "In my opinion" to the end of basically anything someone else says. Of course it's their opinion, they don't need to let you know. You're pretty smart. 2. Assume everyone means everything in the best manner they could mean it. Save yourself some stress and give people the benefit of the doubt. We'll all be happier if we type less emoticons. 3. Don't try to read people's minds. Sometimes people mean exactly what they say. You probably don't know them any better than they know themselves. 4. Let grammar slide. If you understood what they meant, you're good. It's better for your health. 5. Breath. It's just a dumb game.
I think they should go to a point-buy only system, and they should fix the point-buy to reflect the rest of the system's balance.  If a roll is balanced, then it can be replicated using a balanced point-buy system; therefore, the point-buy system encapsulates the roll system.  The opposite cannot be said.  So clearly point-buy should have precedence.



I'm not sure if this is where you are heading, but Point Buy Should be the defualt with rolling as an alternative.
Point buy should always be inferior to rolls.

Rolls already have the potential to go well above a standard pointbuy (or well below).  No method should be better or worse than any other method, just because.
 


Gambling should have a higher potential payout, because it is a gamble. If a more controlable, predictable alternative was just as good then "rational" human beings would never choose gambling. It's the reason 99% of people work steady jobs and don't spend all day at casinos. Even if you ask the people who do make a living out of gambling, you will see that they steadily "grind" poker tables for a certain amount per hour, and never do the huge "No limit" all in stuff. It's not really "gambling".

It's simple logic, not "Just because".
My two copper.
4d6 and drop the lowest isn't really a gamble worth calling a gamble. Now 3d6 is a gamble, 4d6 drop the lowest is statistically in your favor.
Yes, but a point buy that's even would be FAR more statistically in your favor :P
My two copper.
IMO,best is ability bounded by fixed number of total ability scores.

I.E. standard array of 15,14,13,12,10,8 is 72(but that's for ****) we want to be superheros so we say that ability total is 80.

then everybody roll 6×4d6, and if someone is above target then picks which score to reduce to get to 80, and the one below randomly increases scores to get to 80.

That way everyone get's the same score but still have some random thrill with rolling them.
Yes, but a point buy that's even would be FAR more statistically in your favor :P



Have you ever TRIED asking a bunch of people who claim to prefer rolled what their initial ability scores were, checked the point buys required, and averaged.

I have, several times. The average is ALWAYS over 40 points by 3.x point buy. EVERY TIME.

The 25 recommended by the DMG or the 28 needed to make the "average" set of legal rolls aren't even on the table.

No one plays 3d6 in order (what you should use if you actually want to play an "everyman" type with a "realistic" ability distribution).

Hardly anyone plays 4d6 6 times and keep the result as long as it meets the minimums in the PHB. And those that do use reroll by suicide whenever they get anything vaguely CLOSE to 14, 10, 10, 10, 10, 8 (a legal array that you're supposed to play).

Gary and company advocated 3d6 in order, but if you look at the characters they actually played you see a bunch of 18s in key abilities and that they needed that extra % roll to differenciate all the 18 strength characters.

As for your claim that people don't like to gamble. WRONG. Las Vegas stays in business because people LOVE to gamble so much that they will do so even KNOWING in advance that the house takes a cut off the top and always wins.

People's willingness to take bad odds bets supports a multi-trillion dollar a year industry, you won't convince me it doesn't exist.
Jenks, from my experience point buy is never in your favor, it is in the dms favor however. Maybe I just play with power gamers too much. But rolling rarely produces stats below 10. Last point buy game I played I had two stats at 10. Rolling I've rarely had that.
Have you ever TRIED asking a bunch of people who claim to prefer rolled what their initial ability scores were, checked the point buys required, and averaged.

I have, several times. The average is ALWAYS over 40 points by 3.x point buy. EVERY TIME.

The 25 recommended by the DMG or the 28 needed to make the "average" set of legal rolls aren't even on the table.


I'm not sure what you are getting at here. Are you agreeing with me? Nightslasthero is suggesting that point buy be dead even in potential scores with rolling, that's what I'm arguing. Not that it has been worse/better in the past.

 Hardly anyone plays 4d6 6 times and keep the result as long as it meets the minimums in the PHB. And those that do use reroll by suicide whenever they get anything vaguely CLOSE to 14, 10, 10, 10, 10, 8 (a legal array that you're supposed to play).

 
You are correct, and that's an issue with rolling for sure. I said earlier in the thread that I prefer point buy because of this, but that doesn't change what I'm arguing. 

As for your claim that people don't like to gamble. WRONG. Las Vegas stays in business because people LOVE to gamble so much that they will do so even KNOWING in advance that the house takes a cut off the top and always wins.

People's willingness to take bad odds bets supports a multi-trillion dollar a year industry, you won't convince me it doesn't exist.

Yes, a lot of people do like to gamble, but that is not how people make logical decisions :P Most people gamble for the sake of gambling (Which might cause some people to still do rolls if they were even), but most people do not see gambling as a superior decision to stability in most aspects of life :P 
My two copper.
Jenks, from my experience point buy is never in your favor, it is in the dms favor however. Maybe I just play with power gamers too much. But rolling rarely produces stats below 10. Last point buy game I played I had two stats at 10. Rolling I've rarely had that.


This is 100% true, and it's why I use, almost exclusively, point buy or arrays when I can. I'm not arguing for using rolling all the time, just that if you want to still have it as an option it needs to have the potential to produce superior stats to point buy in order to maintain it's viability as an option.
My two copper.
...no rule forbids you to reroll your abilities if you don't like the result.


I feel I must point out that no rule forbids a fighter from shooting 10d6 lightning bolts out his arse all day either.  Since the rules do not say that you can re-roll your stats until you get a set you like, doing so would be equivalent to inventing a house rule to that effect.
Jenks - gambling already has a higher potential payout - you could get all 18s. What you're asking for is a higher average payout, which isn't "simple or obvious". As pointed out, rolling tends to result in higher scores already compared to the hypothetical average - because characters are a lot, roles are lost, and bad characters aren't resurrected. Since point buy never has those, if you want a party who has some rolling and some point buy to be fair, point buy has to average more. Otherwise, people who want the security of point buy will be pressured to roll.
Now, it's fair to say that if point buy is too much higher, people will be pressured to point buy. Which is why making them average the same, while giving a small statistical advantage to rolling, is still probably the best.

And, as dm, I wouldn't want my players to feel like they had to play a subpar char. Which is why I highly recommend point buy.
If you roll too high or too low, ill ask you to retire your character, if it leads to friction at the table. But, that's just me.