Ranes
Adventurer
ROLF. ROLF, I tell you!
(This was going to be an XP comment but your XP button's rubber band appears to have snapped.)
-
The VOIDRUNNER'S CODEX is coming! Explore new worlds, fight oppressive empires, fend off fearsome aliens, and wield deadly psionics with this comprehensive boxed set expansion for 5E and A5E!
High and Low Stat discrepancy and opinion
- Thread starter Naoki00_
- Start date
ROLF. ROLF, I tell you!
(This was going to be an XP comment but your XP button's rubber band appears to have snapped.)
It's pretty ridiculous to get pissed off at someone for rolling well. It's not like you came into the game and said, "Now, with my mighty control over the nature of fate and circumstance, I shall intentionally pull together above average scores. View, puny DM, as I imbalance your game." If the DM wants certain scores, he should use a system that makes those scores happen, instead of a system that sometimes makes those scores happen. Folks like point buy for a lot of reasons, and stat balance is one of them.
As for your other question, how important stats are, the answer is that it depends on the character, and that class power level tends to be a lot more important. As an example of two classes that show both things simultaneously, consider the monk and the druid. Druids don't need anything in particular out of their base stats, easily capable of functioning with the minimum scores allowed that don't allow for rerolls, also capable of functioning with nothing but 10's and 11's, and finally easily capable of functioning with just 8's, if you use something like anthro bat.
By contrast, a monk needs good stats all around in order to find anything like success. You need high scores in both dexterity and wisdom in order to match the AC provided by armor, you need strength to come close to the accuracy and damage of a fighter (and you generally fail), you need constitution, again for the same reason, and intelligence and charisma are nice though not necessary for skill stuff. Any increase in stat allocation to a monk is thus important in a way it's not usually going to be for a druid.
The general outcome of all of that is that monks make better use of high stats than druid does, but that druid doesn't care, because druid is druid. In other words, a druid that has all 8's as base stats, an impossibly low result in most stat allocation methods, is going to be better at most levels than a monk with all 18's. It's an odd result, but it is also a generally accurate one, at least at reasonable optimization levels. Stats just don't mean that much over the full balance spread of the game, and they tend to mean more to low power characters than high power ones, so high stat allocation probably lands somewhere between value neutral and good in the scheme of things.
As for your other question, how important stats are, the answer is that it depends on the character, and that class power level tends to be a lot more important. As an example of two classes that show both things simultaneously, consider the monk and the druid. Druids don't need anything in particular out of their base stats, easily capable of functioning with the minimum scores allowed that don't allow for rerolls, also capable of functioning with nothing but 10's and 11's, and finally easily capable of functioning with just 8's, if you use something like anthro bat.
By contrast, a monk needs good stats all around in order to find anything like success. You need high scores in both dexterity and wisdom in order to match the AC provided by armor, you need strength to come close to the accuracy and damage of a fighter (and you generally fail), you need constitution, again for the same reason, and intelligence and charisma are nice though not necessary for skill stuff. Any increase in stat allocation to a monk is thus important in a way it's not usually going to be for a druid.
The general outcome of all of that is that monks make better use of high stats than druid does, but that druid doesn't care, because druid is druid. In other words, a druid that has all 8's as base stats, an impossibly low result in most stat allocation methods, is going to be better at most levels than a monk with all 18's. It's an odd result, but it is also a generally accurate one, at least at reasonable optimization levels. Stats just don't mean that much over the full balance spread of the game, and they tend to mean more to low power characters than high power ones, so high stat allocation probably lands somewhere between value neutral and good in the scheme of things.
Identical would also remove point buy - different players allocate their points differently, and the same player may allocate differently for different characters.
But we still want "random" - or so the choice to roll the dice seems to indicate. We have established a range of 3 - 18, and also that we want it skewed more to higher rolls than to lower rolls.
I find it outright illogical to roll dice for stats and then complain that the use of a random chance methodology generates results which are random, so I disagree with your conclusion that there is nothing outright illogical in the GM complaining that a choice of a random method of generating numbers carries random results. Think of it this way - will he stop the game at some point and require some change be made because Player #3 has rolled an excessive number of critical hits, or his hit points reflect above average results, or perhaps he rolled max damage 3 times in a row, so he is not allowed to roll max damage again?
Not setting the parameters up front is, to me, "wrongity-wrong". I am curious whether his response would have been similar had the player's luck skewed the other way and he ended up with stats considerably lower than the rest of the players (again, a chance we take when rolling the dice). Perhaps he would, but there is no way to know. From the OP's comments that the other players seemed excited that they got decent stats, it sounds like there have been past occasions where they were less fortunate, and were stuck with those "unlucky" stats.
To me, by deciding "we will roll stats using this method", the group accepted the possibility someone would roll exceptionally well, or exceptionally poorly, with the result likely being a disparity in starting character bonuses. 4d6, drop the lowest, in no way guarantees "at least one terrible score and only 1 or 2 at a +2 mod or more", nor do I perceive the 8's noted as being "terrible scores". "Woe is me, I take a -1 penalty on a dump stat."
One thing for random rolls and arrays - I see a lot more odd numbers using those methods! Although it's nice to have an odd number once every four levels or so...
This made my day lol, sometimes I think one of the DM's we had viewed our optimizers like this
In the direst cases, it is "wrongity-wrong, with wrong sauce"
Aside from how those are apples and oranges, you mean?
If he finds that a character build is too powerful for his game, or is otherwise out-performing everyone else at the table, then yes, I expect him to eventually speak with the player about it. Not in the middle of a combat, but I expect adjustments to be made, in general. Adjustments before the character sees even a moment of play would be, in my mind, even better than having to rebuild the character after several levels of advancement.
As I said, we don't *JUST* want random. It is random, with parameters. As I have also said, it is not illogical to desire parameters that differ from what the basic "4d6, drop lowest" produces. And not everyone is a game designer or statistician that can easily make up the hard rules that will mechanistically produce the desired effect. Basically, "random, with DM oversight" seems as reasonable as "point buy, with GM oversight". The base desire to not have what dice will produce more than a couple standard deviations out is no stranger than not wanting to have a super-optimized point-buy build. His execution seems to have left something to be desired, but the general idea isn't problematic.
Have you never made a mistake? Done everything perfectly the first time out? Never been surprised by an unexpected result? Maybe you should cut some slack - at leastin the name of givign *constructive* criticism. Maybe he hasn't seen this sort of thing happen often enough for him to have to have thought ahead. At least he recognized that it would be an issue before play began, rather than after 5 levels had passed and the player was really attached to the character.
Was he perfect? No. When I use random generation, I generally also tell the players that, should a character come up too weak or too powerful (either compared to the other characters, or my expectations for the particular campaign, depending on the situation) that we may have to discuss some edits. He should have used some similar disclaimer, so expectations were set properly.
delericho
Legend
I've been surprised by how common that is. That, coupled with system mastery, is one reason that point-buy isn't actually as balanced as is commonly thought - some players will optimise their stats to the Nth degree while others cannot; and some are determined to avoid a penalty at all costs, even if that means weakening their characters in areas that are actually far more important.
(All of which is of course fine, if that's what they want, but it does impact on the gameplay, since the optimised PC is ahead of the curve in exactly the same way as the rolled-well PC in a random system - the underlying maths don't care why a character has a +3 instead of a +4, only that he does.)
In both cases (char gen an combat), the random nature of the dice are producing results which are statistically abnormal, which is resulting in a character who the GM believes is having too easy a time with the game. Not identical, I agree, but not wholly dissimilar.
Here, I do not disagree. However, it seems like this GM has been fine in the past accepting characters who are underpowered (the other players were thrilled they had "decent stats" this time). And, again, by choosing a random stat system, he chose to allow the possibility of a character with exceptional stats, just as the players appear to have accepted the possibility (even the reality in prior games) of sub-par stats. I also note it was the lucky player who first proposed some leveling of the results by allowing the poorer stats to be improved.
The fact is that a lot of players like the random roll methods, and even accept sub-par characters as they arise, because of the possibility the random rolls will generate a similarly above standard character on occasion. That is what random rolls do and, to me, is a big reason that other generation methods, such as stat arrays and point buy, have developed from the hobby's early days when stats were entirely random.
By choosing to roll, I submit that we have chosen some level of random. The parameters chosen are what 4d6, drop the lowest has the potential to generate. We did not choose, for example, 3d6 6 times (lower averages, still with the potential for substandard or over-standard results) or roll 3d6 12 times and keep the best six (reduced odds of something at the very low end). To me, if your desired parameters differ from what the basic 4d6, drop the lowest methodology produces, then choosing to use 4d6, drop the lowest is not logical.
We had a game at one time where the rules would say 3d6, in order, for stats. We took that approach, when one modification was suggested - you could buy "insurance". I don't want to be stuck with a '3' in this stat. OK, we decided, you can declare this up front, on any stat - rolls of 3 or 18 are both re-rolls. Have to have at least a 6? Reroll 16-18, as well as 3-5. IOW, whatever you are protected from at the low end costs you access to equivalent numbers at the high end.
Ultimately, my biggest issue is that the ground rules were not laid out beforehand. I would also think it inequitable to structure the system on the basis that being unlucky means you get a substandard character, and being lucky means you get to re-roll, as we play this game with only average to substandard characters. If that's spelled out in advance, the players know the ground rules going in. Here, the ground rules seem to be getting changed (maybe I am just reading that in, but the delight of the other players at getting 'decent' stats suggests they have rolled poorly and played sub-standard characters before). Ultimately, if we want the range to be more confined, use a generation system that is more confined.
Fair enough - but I can also see the view of a player (not the specific one in this instance) who rolled poorly a couple of times, average a few times, and always accepted that this is the way the game works - the vagaries of chance are part of random rolls. Then, the rolls go in his favour, he's about to get the payoff for being the "average to weak" character in prior games, and the GM says "nope, you get the downside only, not the upside". It seems like a player having that history could easily be "really attached" to the lucky rolls that finally came his way.
To me, the answer is, again, to assess those parameters. What range of stats is acceptable? Is the player protected from the downside, as well as being denied the upside? If we want all the characters to fall into a certain range, then temper the random rolls accordingly, or use a system with less potential for variation. While the one player in this instance got pretty lucky, it's well within the range of 4d6, drop the lowest.
Another peculiarity f this specific situation, of course, is that it was for a one shot game. Assessing how much impact his 18, 18, 16, 14, 12, 13 provided over the 15, 15, 12, 12, 10, 8 average of the other players was pretty risk-free with no long-term campaign impact.
What's also unknown is how the other players felt. We know the GM is concerned that this one character will be overpowered, and the player is pretty happy with his rolls and looking forward to playing this pretty good to great stat set. Are the other players congratulating and high fiving him for his great rolls, or grumbling over how they all feel like comic relief sidekicks compared to SuperStats? It's about the enjoyment of the entire group, not just one participant. Finally, there's a clear variation in expectations, as the high roller feels his stats are reasonable, maybe a bit lucky, while the other players view their much lower arrays as unusually decent.
Clearly, someone at the table isn't happy (or won't be happy). To me, that's a matter for group discussion. However, I come back to "random rolls have an element of chance, so don't use that method if that element of chance is not desired". If the intent were to cap the characters at, say, 25 point stat buy, then why not let them use 25 point stat buy? If the intent is some variance in stats, we could also have a "roll your stat points, then buy" model - maybe we get 23 - 28 stat points. The average player looks to have 26, while the one player blew the doors off.
ADDENDUM: I find I'm typically on the side of the GM in similar discussions. I'm not sure why this one strikes such a chord with me. The "player gets the downside but not the upside" aspect is giving me some tunnel vision, I suspect. Our groups have moved to point buy specifically to remove the random "someone does poorly and someone else does spectacularly well" aspect of random rolls. We call it "random", but the laws of chance make it inevitable that someone will roll well, and someone poorly, at some point in time.
Random roll incorporates an element of chance. The more chance is removed, the more system mastery comes to the forefront. However, there is always a strong element of chance in d20 due to the d20 itself. That extra +1 is a 5% improvement to the chance of success on a d20 roll. Compared to the other array, the lucky player here has a 10% advantage on his main 2 stats (18 vs 15), and the next two are a 10% enhancement and a 5% enhancement (16 and 14 vs 12s). 5% and 10% on the last two (and that last 10% will be in the "dump stats", which we typically assign because they won't come up much.
Going to fixed arrays reduces flexibility further, but there's still a mastery element in choosing where to assign each stat.
Optimization/mastery goes well beyond the stat scores, and once we get past stats (where the rules provide random and non-random choices), the only remaining random element is the hit point roll after first level, so the game certainly allows for any level of optimization. I find that a problem only where the level of optimization varies considerably between the players, such that someone is clearl overpowering or someone else is clearly not powerful enough.
Last edited:
delericho
Legend
I don't disagree with any of that. However it was noted up-thread that a character who dumps a stat could potentially go a full campaign without it ever mattering - his Fighter may never miss a Cha check by exactly 1. The same isn't true of your main stat, though - over the course of the campaign the Fighter is very likely to either miss or hit by exactly 1.
So all I'm noting is that point-buy isn't a guarantee of balance within the party. Provided everyone is able and willing to apply the same level of optimisation, then it works great, but in a mixed group that's much less certain.
Plus, of course, in 3e the point-buy systems really hurt MAD classes. Speaking of which, given the stats he rolled, perhaps the OP should take this opportunity to play that Monk he's always wanted.
... okay, maybe not.
I don't think that's necessarily true. Yes, the main alternative, rolling, can produce a bunch of high stats, as it did here, but it can also produce mediocre and craptacular stats, cause randomness. Really, the best stat allocation system for a MAD class is point buy, but high point buy. At a certain level of stats granted, the SAD tier one classes stop particularly caring about increases to point buy, while the MAD classes really never stop doing so, and it all adds up to a marginal increase in balance. Alternatively, you could always run a tier biased point buy, as suggested in the tier system, granting tier 1's something like 24, or even 15, and tier 5's maybe 40. It doesn't do all that much to hit overall balance, but it's definitely the most balanced allocation system.
delericho
Legend
True. Which is why, with a random roll system, you should probably roll, then pick race/class, and then assign the stats. And only go for the MAD class if you have the stats to power it.
I was assuming a rough equivalence between the random and point-buy systems in use. 4d6-drop-lowest is roughly equivalent to 28-point buy (in 3.5e). If you're going to use a much higher point value, you need to compare it with a more generous random method, or you're not comparing like with like.
Incidentally, the best stat allocation method for MAD classes is "just pick whatever stats you want". Although I suppose that's indistinguishable from a sufficiently high point buy.
Last edited:
Similar Threads
