Statistical Analysis of the Classes - popularity vs. power

[ichabod]

Yes, that's the failing; the second list is purely subjective. But, I figured that given enough responses, if everyone truly thought that all the classes were balanced, then the average scores for all the classes would come out to be 6.0 (the average of 1 to 11). In theory, a 6.0 for a class (which the Paladin got) is a perfectly balanced class. Anything above that is powerful, and anything below that is weak.

Given that, statistically speaking, cleric, fighter, wizard are all too powerful, and monk, ranger, and bard are all too weak. And, apparently, Bard is as weak as a cleric is strong (both are ~4 off from midpoint).

As a statistician in training, I'm going to have to weigh in here. The problem isn't really that the second list is purely subjective, people's opinions are measured all the time. You just have to be clear that you are measuring opinions, not facts. You did this at the begining, but seem to be losing sight of it. The main problem is that you haven't shown that either of these surveys is from a representative sample of the overall population of 3E players, or even 3E online players, or even members of these message boards. This casts serious doubts about any conclusions.

The other big problem is the way you phrased the question about power. You did it as a ranking instead of a rating. To show how this is a problem, take two people, A and B. A thinks there are some slight variations in power, but all the classes are pretty balanced. B thinks some classes are uberpowerful, while others are a waste of paper. A and B could still give the exact same answers to a ranking question, because you have imposed a metric upon them that is not necessarily reflected in their opinions. However, if you give a rating question, where each class gets a power rating from 1 to 10. The the difference between A and B's answers would become apparent (A having a lot of 4s, 5s, and 6s; B having a lot of 10s and 1s).

I don't mean to be overly negative, it's just that I don't get to take the Sample Surveys class until next semester, so I don't know enough to suggest useful alternatives.

 

[Ridley's Cohort]

What Ichabod said.

The data is confusing ordinal rankings with cardinal values of power. The average of an ordinal value is not very meaningful, the median would probably be better.

We also have the problem of level. Monks are pathetically weak at low level, adequate at medium levels, quite cool at high levels. Rangers and Barbarians are better than Fighters at low levels.

 

[Plane Sailing]

Hi, I'd just like to turn up and take a bow since I did the original survey of "who plays what classes" about a year ago.. I specifically decided to collect information on what people were actually *using*, rather than on opinions, in order to reduce the subjectivity slightly.

I was very pleased to have had so many respondants (about 100 board members provided information about the parties that the ran/played in).

I ought to clarify that the popularity isn't "number of characters with this class", it is "number of class levels taken". This was done to simplify my gathering of information.

I did consider doing another survey (perhaps I should?) which showed a grid of how many characters of each level there are. I did collect some information that way, and I can tell you that looking at Rangers there was 8x1st, 4x2nd, 2x3rd, 3x4th, 1x5th, 4x6th, 2x7th, 1x9th. On the other hand, Barbarians had 7x1st, 2x2nd, 2x3rd, 1x4th, 1x5th, 2x6th and 1x11th. It seems like there were not too many people who were taking 1 level of Ranger for the TWF amongst those who responded!

In the original survey I collected information about prestige classes too. Most popular was Shadowdancer (with 23 levels reported), then homebrews with 16, Arcane Archer with 15, Order of the Bow with 13, Templar with 11 and so forth.

Do you think it would be interesting to run this survey again, and create a class/level grid?

Cheers

 

[Frostmarrow]

I'm convinced that the designers made some classes better than others on purpose. I think their reasoning went something like this:

Fighters are weak. Most "fighters" nowadays are elven rangers because those are a lot more efficient in combat than a human fighter. Still, we want the human fighter to be the most common character in the game. Face it, what is the point of being an exotic ranger if everybody is an "exotic" ranger. Let's make human fighters rock most!

Clerics are unpopular because playing a cleric is a chore. All you do is run around healing the others in the party. We need to make clerics more popular or there will be too many parties without one. How do we make the cleric popular? We make them better in melee and we up there other abilities such as spell-casting and turning!

Wait! The ranger is still too popular. The ranger is a very attractive concept and far too many wants to play the silent, deadly, cool guy. We must make the ranger weaker or rangers will not be very exotic.

Let's tone down the paladin and the bard too. Those are supposed to be more uncommon than rogues and fighters anyway and as we know players will pick the most favorable choices for the most part we need to make fighter, rogue, wizard and cleric more favorable than the others.

That's my assessment. I'm pretty sure I'm 100% wrong as usual.

 

[Plane Sailing]

One of the things that astonished me was how poorely the Barbarian did - far less popular than I'd imagined.

 

[Celtavian]

Re

It really is too hard to rank class power.

Think about the following:

Would a fighter survive if he didn't have a cleric healing and wizard casting? Probably not. One bad will save and adios.

Would a cleric survive without a wizard casting or fighter brawling? They might depending on the creature, but one bad dispel or missed concentration check would cause them a world of hurt.

Would any party really want to go through a dungeon without a rogue? Advance scouting, lockpicking, and the big one, trap finding, are essential to successful dungeoneering.

Would a party really want to go without a wizard? No haste, no area of effect spells, no magic missile, no dispel magic, and several other useful spells.

The core classes are all fairly equal.

You can basically replace them with any of their subclasses, except you can't replace a low level monk with a fighter, though a high level monk might be able to replace a fighter.

I have played or had others play every single core class, and not one is useless. People even love the bard's battle songs and his spells come in useful as well.

I think each class is playable and fairly well-balanced if you dont' look at it as player vs. player. If I were to rate classes player vs. player, I would say a prepared wizard is the most powerful.

 

[candidus_cogitens]

I'm convinced that the designers made some classes better than others on purpose. I think their reasoning went something like this:

Fighters are weak. Most "fighters" nowadays are elven rangers because those are a lot more efficient in combat than a human fighter. Still, we want the human fighter to be the most common character in the game. Face it, what is the point of being an exotic ranger if everybody is an "exotic" ranger. Let's make human fighters rock most!

Clerics are unpopular because playing a cleric is a chore. All you do is run around healing the others in the party. We need to make clerics more popular or there will be too many parties without one. How do we make the cleric popular? We make them better in melee and we up there other abilities such as spell-casting and turning!

Wait! The ranger is still too popular. The ranger is a very attractive concept and far too many wants to play the silent, deadly, cool guy. We must make the ranger weaker or rangers will not be very exotic.

Let's tone down the paladin and the bard too. Those are supposed to be more uncommon than rogues and fighters anyway and as we know players will pick the most favorable choices for the most part we need to make fighter, rogue, wizard and cleric more favorable than the others.

That's my assessment. I'm pretty sure I'm 100% wrong as usual.

Wow! This really makes me think. I agree that rogues and fighters should be more common, for the sake of realism. On the other hand, couldn't there have been another way around the problem. For example, you could just make a rule stating that there cannot be more than one bard in a party? Ditto for rangers, druids, and monks.

 

[ForceUser]

Wow! This really makes me think. I agree that rogues and fighters should be more common, for the sake of realism. On the other hand, couldn't there have been another way around the problem. For example, you could just make a rule stating that there cannot be more than one bard in a party? Ditto for rangers, druids, and monks.

Yuck! No way. That smacks of Deus Ex Machina. It's bad enough when a DM does it in-game; it would really suck if a game designer did it. Not to mention that it would get rule-zeroed pretty quickly in quite a few campaigns anyway.

By the way, druids have been way undervalued by these results, which leads me to believe they are meaningless. If you want to know just how hardcore and scary a druid can be, read about Nwm the Druid from Sepulchrave's story hour.

 

[Crothian]

By the way, druids have been way undervalued by these results, which leads me to believe they are meaningless. If you want to know just how hardcore and scary a druid can be, read about Nwm the Druid from Sepulchrave's story hour.

Druids rock. We, I should say he, stopped an invasion by himslef. 4 landing craft of 20 barbarians each (all about 5th level or so). The rest of the party stood and watched as he used the elemetnas and nature to destroy them all. A Druid who has time to prepare the right spells for the job is scarey.

 

[JeffK1966]

Well, I recently played in an Epic level campaign, and my 25th level human ranger was probably the biggest "butt kicker" in a climactic combat, outdoing the human fighter/ranger and the elven fighter/mage bladedancer and the 26th level halfling rogue, as well as the epic monk... and, my ranger never got to cast any of his spells until the next session.

So, I don't know if I'd consider the ranger to be underpowered...