The Pitfalls of D&D Beyond Data

[Mistwell]

I’m disputing the value of misleading and inaccurate data.

Pretty sure saying data is inaccurate is disputing its value. Are you really quibbling with that statement too?

 

[TwoSix]

Pretty sure saying data is inaccurate is disputing its value. Are you really quibbling with that statement too?

I believe the general dispute is of the value of this specific set of data, in particular the subclass pie chart. It's not a quibble with the overall data set that D&D Beyond has access to, or with its overall intent to show us pieces of it.

 

[Mistwell]

I believe the general dispute is of the value of this specific set of data, in particular the subclass pie chart. It's not a quibble with the overall data set that D&D Beyond has access to, or with its overall intent to show us pieces of it.

There is no "the" dispute. I am talking to Frog specifically, and you are oddly answering for him, and in a way I don't think he agrees with by the way though I am sure he will appreciate your general support. I think he does have a quibble with the overall data set and intent. Maybe not, but I feel very confident he did not tag you in to respond for him.

But bottom line here, he questioned the value of the data, I stated that, he quibbled with that though he again repeated he thinks it's inaccurate which is a direct questioning of its value, and I think it's a good demonstration of the kind of thinking were dealing with here when someone dispute using the word "value" while calling it "inaccurate". It's the kind of thing someone who is looking to "win a fight on the internet" would say rather than have a conversation where there is back and forth allowed.

I mean seriously, if I cannot characterize "it's inaccurate data" as "you're questioning the value of the data" then this isn't a conversation, right?

 

[cmad1977]

Eliteism over post counts wow lol.

The number of posts in a forum is clearly the best metric for assessing another posters reading comprehension and understanding of the topic at hand.

... wait...

 

[TwoSix]

There is no "the" dispute. I am talking to Frog specifically, and you are oddly answering for him, and in a way I don't think he agrees with by the way though I am sure he will appreciate your general support. I think he does have a quibble with the overall data set and intent. Maybe not, but I feel very confident he did not tag you in to respond for him.

But bottom line here, he questioned the value of the data, I stated that, he quibbled with that though he again repeated he thinks it's inaccurate which is a direct questioning of its value, and I think it's a good demonstration of the kind of thinking were dealing with here when someone dispute using the word "value" while calling it "inaccurate". It's the kind of thing someone who is looking to "win a fight on the internet" would say rather than have a conversation where there is back and forth allowed.

I mean seriously, if I cannot characterize "it's inaccurate data" as "you're questioning the value of the data" then this isn't a conversation, right?

If you want to have a private conversation with someone, take it to PMs. Otherwise, it's open season, and as you're simultaneously on the wrong side of the argument (IMO) and being fairly abrasive, it's totally OK to jump in.

And yes, the subclass pie chart has no value. If you don't control to make sure it's only querying for characters that have taken subclasses, it's about as useful as a chart of the most popular classes for characters whose name starts with 'B'.

Hopefully that's clear enough to skip past semantic debates about "value" or "inaccurate" or "data".

 

[Stalker0]

Does it mention anywhere how many data points were gathered to generate these numbers? Like the number of characters on beyond?

 

[FrogReaver]

Does it mention anywhere how many data points were gathered to generate these numbers? Like the number of characters on beyond?

Without double checking I believe the total number of characters on D&D Beyond was cited at 8.8 million. Of course there is also a bit of debate about whether those charts include all those character or some subset of them called active characters (whose criteria thus far is undefined to us).

 

[FrogReaver]

Pretty sure saying data is inaccurate is disputing its value. Are you really quibbling with that statement too?

Yes. I think it’s important to be more precise in this instance.

The chart they provided for subclasses was objectively inaccurate and misleading (basing this judgment on the graphed data not aligning with the graph title)

The chart for classes handles multiclassing in an unusual way, and because of that also doesn’t at all show what the graph title claims it shows. I’d call that inaccurate and misleading too but if you have a better word I’m open to it.

the subrace breakdown doesn’t have these issues. It’s fine in that respect.

Of course there’s also the normal question of how representative any dataset is of an actual population but that isn’t what’s making me characterize those charts as inaccurate and misleading.

As to value, I think their actual dataset is very valuable even with questionable representativeness. However, The graphs we got to see are mostly invaluable but there’s still a sliver of value to them once we are able to label what they are actually showing vs the incorrect claim of what they are showing.

 

[Mistwell]

If you want to have a private conversation with someone, take it to PMs. Otherwise, it's open season, and as you're simultaneously on the wrong side of the argument (IMO) and being fairly abrasive, it's totally OK to jump in.

You can jump it of course. But, you can't answer for someone else. Which is what you did. As for you having an issue with my tone, take that to a moderator and not to me.

 

[doctorbadwolf]

I want to add. In most business you have the business side and the technical side. I'm sure the technical people get everything we are saying. However, the business side wanted a few clean easy to present charts to show off to the public. They likely hammered away at exactly what they wanted over the objections of the technical people and ended up with what we have.

The guy doing to videos and presented the graphs is also the guy who headed the creation of the service, and helped program it. If there is a significant division at DDB between the "business side" and "technical side", he isn't representative of that divide.

Eliteism over post counts wow lol.

There was a person who used to post here, who in an interaction via PM that got us both a deserved vacation, tried to tell me that the forum in general clearly liked him more than me, because he had more XP than me.

Of course, I didn't bother pointing out that he also had been here for twice as long, and had twice as many posts, but only a couple hundred more XP...I just don't care about internet popularity contests enough to win that easily won "contest", I guess.

People are weird, is the point.

If not for the imperative to correct people who are wrong on the internet, ENWorld would be a ghost town.

Is that why you come here? Certainly isn't why I'm here.

Note the numbers involved with this data are more than political polling.

It was mostly kind of assumed that games were mostly low level and the basics were the most popular.

Even in AD&D Fighters were way more popular than Druids.

The only mild surprise was Tieflings and Dragonborn being as popular as they were. Even then it wasn't a massive surprise as power and popularity are not related (3.5 fighters for example).

Also see, 5e rangers. Most people aren't satisfied with them, according to wotc polling, but they are in the high middle of classes played, according to wotc AND DDB data. Because people play concepts that they like, even when the mechanics aren't good.

Without double checking I believe the total number of characters on D&D Beyond was cited at 8.8 million. Of course there is also a bit of debate about whether those charts include all those character or some subset of them called active characters (whose criteria thus far is undefined to us).

Active characters were loosely defined in the most recent dev update. I'm not going to go watch it a second time just to answer more precisely, but it was something along the lines of "characters who get updated". So, ya know, active characters.