D&D 5E Nobody Is Playing High Level Characters

According to stats from D&D Beyond, above 5th level characters start to drop off sharply, and above 10th level, the figures are very low. The exception is level 20, which looks like it's probably people creating experimental 20th-level builds.

Some of them say 0%; this isn't strictly accurate, but levels 16-19 are used by an insignificant number of players. Interestingly, there are more 3rd-5th level characters than there are 1st-2nd level.

D&D Beyond has said before that under 10% of games make it past 10th level, but these figures show the break point as being bit lower than that. DDB used over 30 million characters to compile these stats.

 

[Anoth]

There is tons of character growth in D&D. The mistake is thinking character growth can only be expressed by mechanics options.

The myriad things your character will do in game over so many levels is where the real character growth lives. The things that happen in your game that brings real life to your characters and to the world. Picking feats isn’t real character growth... it’s just video game character building.

In my experience, excessive mechanical options get in the way. They force optimization and picking cool powers in order to be effective.

I’m pretty close to giving up on 5e because it only rewards mechanical optimization. Players pick options for all the powers they grant and the mechanical doo-dads they provide.

That’s not a system problem. That’s a player problem or a DM problem.

 

[Umbran]

Thanks Umbran, I did miss that! However, I would want to see the actual precise methodology before I believed that they'd done an even okay job there

You do realize that, unless you are a statistician or data scientist yourself, your ability to really critique the methodology is limited? I mean, maybe you are, and then, cool. But otherwise...

The stakes on this... aren't big, you know. Is it really so important that they'd lie about it? Do you figure this piece of misinformation is going to move the needle of sales... like, at all?

because it would imply the level 20 test build theory was wrong

The solution to that is simple - they can't go beyond 20, so they stack up there.

and is very hard to square with the peak in characters at L3-L5

Why? I don't see the problem - using either XP or milestones in common WotC modules, low-level play gets over and done with very quickly, but advancement slows down after that. So, characters don't stay at 1 or 2 for long, and linger at 3+...

Alternatively, the same filters for "active" characters might well not see level 1 or 2 characters very well - by the time they have been active enough to be reliably active... they are 3rd level. I mean, really, it can take like 2 sessions, especially if they are milestone leveling.

which are precisely the levels of the vast majority of created but unplayed characters that I've seen

Which is... a statistically relevant sample? If you are going to question their methodology, your own must also be reviewed, no? Or are we not going to be even-handed about that?

If they have done good job, then that implies to me that a lot of people are starting at 3 or higher and maybe that people are playing L20 games a bit (something I have seen here and on reddit occasionally).

See above.

 

[dave2008]

Thanks Umbran, I did miss that! However, I would want to see the actual precise methodology before I believed that they'd done an even okay job there, because it would imply the level 20 test build theory was wrong...

Why is that? When we do a 20th level test, we actually play the characters. We don't make characters to look at them. I have never had a campaign go to 20, but I have run to 5e adventures at lvl 20.

 

[Ruin Explorer]

You do realize that, unless you are a statistician or data scientist yourself, your ability to really critique the methodology is limited? I mean, maybe you are, and then, cool. But otherwise...

The stakes on this... aren't big, you know. Is it really so important that they'd lie about it? Do you figure this piece of misinformation is going to move the needle of sales... like, at all?



The solution to that is simple - they can't go beyond 20, so they stack up there.



Why? I don't see the problem - using either XP or milestones in common WotC modules, low-level play gets over and done with very quickly, but advancement slows down after that. So, characters don't stay at 1 or 2 for long, and linger at 3+...

Alternatively, the same filters for "active" characters might well not see level 1 or 2 characters very well - by the time they have been active enough to be reliably active... they are 3rd level. I mean, really, it can take like 2 sessions, especially if they are milestone leveling.



Which is... a statistically relevant sample? If you are going to question their methodology, your own must also be reviewed, no? Or are we not going to be even-handed about that?



See above.

You're reading an awful lot in that isn't there, Umbran.

I mean "lying" for "sales" where is that coming from? Not from me. It's nonsensical - this doesn't drive sales, it's a fluff piece. Lying requires malice or intent, whereas the issue I see is potentially the opposite - low-grade work.

You definitely do not need to be a statistician or data scientist to criticise the basic methodology here. The issue is down to what they're choosing to say means a character is played or not played. That is something that can be explained. As an aside, I was a researcher until recently and am now working in knowledge automation so I don't think I'm exactly totally unqualified to say anything.

As for the stakes, they're low, i agree, and the main issue I have is just two things:

1) The credulity of a lot of people re figures like these in general.

2) How seriously people take them, despite them being fluff, and how they will, based on previous experience, be repeated as if they are unquestionable scientific fact/gospel truth (delete as appropriate) for literally years, and used to make all sorts of dubious claims and arguments.

If the general position was one of amused skepticism, ie that the figures are fun and interesting but not necessarily accurate, I wouldn't be interested in the methods, but my experience, based on previous statistics stuff the Beyond team have put out is that this is not taken that way at all. YMMV. Hopefully I'm wrong and we don't see various people and threads in the next few years making claims based on this stuff but...

 

[ChaosOS]

There's actually an interesting point about characters "stacking up" at 20 - in Hearthstone (and a few other games), they have a draft format that's kinda like a 4e skill challenge - go until X wins or Y losses, whichever comes first. In Hearthstone that's 12 wins or 3 losses. Well, as it turns out, finishing with 12 wins is more likely than 11 wins, as the 12 win runs are the sum of all the runs that could've done 12 wins or better - it's got 12-0 runs, 12-1 runs, and 12-2 runs, while the only way to finish with 11 wins is to go 11-3. Point being, if there's a hard cap, it's perfectly natural for that to end up grabbing a larger percent than all the lead up to it.

 

[Ruin Explorer]

Why is that? When we do a 20th level test, we actually play the characters. We don't make characters to look at them. I have never had a campaign go to 20, but I have run to 5e adventures at lvl 20.

That is, by definition, not merely a test character.

 

[Anoth]

Personally, I’m going to go save the princess.

 

[dave2008]

That is, by definition, not merely a test character.

I gave my group a stack of pre-generated 20th level characters to choose from. Once they selected their characters I ran them through an adventure so we could test how 20th lvl 5e worked. We had no other purpose than to test 20th level play. If that is not what you consider a test character, what is?

 

[Michael Silverbane]

I suspect that there is much less opportunity to play high level characters, because no one is running games for them.

 

[Ruin Explorer]

I gave my group a stack of pre-generated 20th level characters to choose from. Once they selected their characters I ran them through an adventure so we could test how 20th lvl 5e worked. We had no other purpose than to test 20th level play. If that is not what you consider a test character, what is?

The implication made in the original post was that they were being made merely to see how you could build a 20th-level characters. That's what I'd call a "test character". Any character that actually gets played, is, to me, just "a character". Obviously this is totally subjective.