So I recently had a discussion with an individual on what constitutes sufficient testing for homebrew before you publish it on this platform. Their view was that you had to have hundred of people or more test it in an actual game, to meet the testing qualifier, before you can publish it on here. My view was that my personal testing was sufficient to publish it on here, then take peoples feedback and do adjustments based on feedback as long as the feedback keeps it balanced with official subclasses for that class, and doesn't change my vision of what I wanted to create. While I would love to be in a position to have access to a few hundred DM's/players willing to playtest a subclass or other homebrew I create before posting it here, I am not, which is why I developed the following process for testing.
Lv.3 x20 easy/hard/deadly encounters(60 total).
Lv.7 x20 easy/hard/deadly encounters(60 total).
Lv.11 x20 easy/hard/deadly encounters(60 total).
Lv.13 x20 easy/hard/deadly encounters(60 total).
Lv.17 x20 easy/hard/deadly encounters(60 total).
Each level of difficulty in a tier has different monsters, and monsters don't repeat between tiers.
Use baseline official subclass of same class as my subclass by running each tiers encounters.
Use following die averages for calculating damage- 1d4(2.5), 1d6(3.5), 1d8(4.5), 1d10(5.5), 1d12(6.5)- which I was told by multiple people was the standard.
I use the same stat array for both the baseline and homebrew character, as well as the same race.
Both baseline test and homebrew test are conducted against encounters for a single character of that level.
No magic items are used, and I DO NOT min/max either the baseline or homebrew subclasses.
This process was found to be far more dependable for testing since it covers multiple levels of gameplay and consists of 300 encounters. Given the average player plays in a campaign that is run once per week, and most campaigns have 1 combat per session(this is being extremely generous), this testing is equivalent to 5 players testing each subclass in a year long campaign, or 1 player testing the subclass over a 5+ year campaign.
So my question to you is this....... Is my testing sufficient for publishing on this site or is the person I had this discussion with correct and no matter how much testing I do myself it doesn't count, and it can't be considered playtested until hundreds of players have played the subclass in their campaign?
Hundreds?!? Nah, no way. Nobody could ever publish anything if it took hundreds.
Technically you can publish anything, but that’s not what it’s intended for.
This is what I personally consider basic playtesting in addition to the number crunching.
Did you have anyone optimize the bejeepers out of it to see how far it can go before it breaks anything? One of the things I find most beneficial with playtesting something is to intentionally try to break it so hard it breaks the game. That’s one of the things you want to avoid, and only a munchkin is gonna find those like that, like a bloodhound. On a scale of 0 - 5, where 0 is guaranteed munchkinproof, and 5 is “oh 💩, well see what had happened was…” what’s it’s munchkin crash test rating?
Did you do any non-combat testing?* Did it have anything to do, or does it suffer from Da Champ syndrome?
Did you have at least one or two other people play it for a few sessions at each of the various tiers of play to see if they interpret anything differently than you intended when you wrote it?* As authors we sometimes develop blind spots in our work because we understand what the thing is sposta say, but sometimes we miss what it actually says. Recently I wrote an Action two different ways and just included both without mentioning it to see what some said during feedback. They said they liked the “Alt” version more because it allowed for a specific interaction which the other didn’t. (I thought they both said the exact same thing just worded differently.) What does it say that we don’t see and nobody will realize until it actually comes up in a game?
Did you test it without die averages, but actual rolling to find out how bad it might get if it runs away on a good night, or how bad it might get if the dice are frosty on a bad night?*
*Did it get tested in parties with other PCs run by other players to see if there are any funky combinations you didn’t realize? (Obviously cross-testing everything is unrealistic. Was it at least checked with the most interactive, or for support playtests the most beneficiary, of the PHB base classes. At the very least was it checked out with the stuff it was most designed to do?
How many times was it proofread for typos, misspellings, incorrect spacing or punctuation? Even in all the snippets too? Because, as an author, those drive me nuts when I miss them. (And I can’t spell, so I miss ‘em plenty. 🙄)
That’s what I shoot for anymore before I publish these days. Which is why I haven’t published much of anything in a while, it’s on a backlog of playtesting.
Gut reaction is that you shouldn't publish on D&DBeyond until you have it as a google doc (or alternatives like gmbinder.com and homebrewery.naturalcrit.com ). D&DBeyond also has coding limitations that might not even allow some mechanics you want to use. This helps with spelling errors as mentioned previously.
As far as "game balance" then the short list I use and consider fundamental regardless of game system. It also assumes you have an intent or theme decided upon.
How does it impact the duration of the player's turn or round?
Does it have a use in play?
What is its power relative to other things present?
The first considers the action economy of not just the individual but the group. If it is causing one player to take an extraordinary amount of time compared to the others then that is probably an issue. This isn't restricted to mechanical things like adding additional actions. This could be rolling more dice thereby increasing the time to count then deal with math (yes, the physical ones), needing to sift through a table on some page constantly, or trying to figure out a longwinded spell effect. All sorts of things can do this including features currently in the game.
The second is where you can analyze with the pillars D&D 5e uses if it helps. Ideally you would cater your homebrew to what you player group is used to, but if you want other people outside to use it then it might need to be a more balanced spread (I personally weigh exploration and social together vs. combat but you don't have to). If you made weapon 1 deal 1d6 piercing damage, and weapon 2 deal 1d6 piercing damage then one of these weapons has little use in play considering they aren't mechanically distinct. Flavor is free of course, but shouldn't be necessary to make up for boring duplicates. Additionally, if there is a class feature that rarely comes up due to requiring far too many conditions to be met, so it also fails to have a use in play.
The third is probably the easiest, however it doesn't mean just damage/healing. If you made a cantrip that manipulated terrain on the same level as Move Earth then it is still too strong considering the implications of world building. It should also be noted that Mold Earth cantrip exists with minor terrain effects (from Xanathar's). What you have mentioned in your post is mostly concerning this. I also see some d&d optimized youtubers compare these things to a Warlock spamming Eldritch blast as a baseline. You could start there instead of trying to grind out so many encounters.
What I don't recommend is that you place a pedestal so high...even official content can be questionable at best. I usually feel about 1/4 or 1/3 of the official stuff fails at what I mentioned above for "game balance." Some of it I think is unintentional, like Truestrike, some of it is intentional, like Fireball, that they wanted to be "iconic."
Though I 100% agree with the others, my opinion is a little different. Posting here is a good way to get some of those varying viewpoints and opinions that may not otherwise show up while testing depending on your group.
Honestly, posting my stuff for feedback was one of the best things I did, even if "incomplete". I learned so much about the game and platform that I would not have known otherwise and it got me to the point where I have formulated my own process to streamline designing, much like you have.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
So I recently had a discussion with an individual on what constitutes sufficient testing for homebrew before you publish it on this platform. Their view was that you had to have hundred of people or more test it in an actual game, to meet the testing qualifier, before you can publish it on here. My view was that my personal testing was sufficient to publish it on here, then take peoples feedback and do adjustments based on feedback as long as the feedback keeps it balanced with official subclasses for that class, and doesn't change my vision of what I wanted to create. While I would love to be in a position to have access to a few hundred DM's/players willing to playtest a subclass or other homebrew I create before posting it here, I am not, which is why I developed the following process for testing.
This process was found to be far more dependable for testing since it covers multiple levels of gameplay and consists of 300 encounters. Given the average player plays in a campaign that is run once per week, and most campaigns have 1 combat per session(this is being extremely generous), this testing is equivalent to 5 players testing each subclass in a year long campaign, or 1 player testing the subclass over a 5+ year campaign.
So my question to you is this....... Is my testing sufficient for publishing on this site or is the person I had this discussion with correct and no matter how much testing I do myself it doesn't count, and it can't be considered playtested until hundreds of players have played the subclass in their campaign?
Hundreds?!? Nah, no way. Nobody could ever publish anything if it took hundreds.
Technically you can publish anything, but that’s not what it’s intended for.
This is what I personally consider basic playtesting in addition to the number crunching.
Did you have anyone optimize the bejeepers out of it to see how far it can go before it breaks anything? One of the things I find most beneficial with playtesting something is to intentionally try to break it so hard it breaks the game. That’s one of the things you want to avoid, and only a munchkin is gonna find those like that, like a bloodhound. On a scale of 0 - 5, where 0 is guaranteed munchkinproof, and 5 is “oh 💩, well see what had happened was…” what’s it’s munchkin crash test rating?
Did you do any non-combat testing?* Did it have anything to do, or does it suffer from Da Champ syndrome?
Did you have at least one or two other people play it for a few sessions at each of the various tiers of play to see if they interpret anything differently than you intended when you wrote it?*
As authors we sometimes develop blind spots in our work because we understand what the thing is sposta say, but sometimes we miss what it actually says. Recently I wrote an Action two different ways and just included both without mentioning it to see what some said during feedback. They said they liked the “Alt” version more because it allowed for a specific interaction which the other didn’t. (I thought they both said the exact same thing just worded differently.) What does it say that we don’t see and nobody will realize until it actually comes up in a game?
Did you test it without die averages, but actual rolling to find out how bad it might get if it runs away on a good night, or how bad it might get if the dice are frosty on a bad night?*
*Did it get tested in parties with other PCs run by other players to see if there are any funky combinations you didn’t realize? (Obviously cross-testing everything is unrealistic. Was it at least checked with the most interactive, or for support playtests the most beneficiary, of the PHB base classes. At the very least was it checked out with the stuff it was most designed to do?
How many times was it proofread for typos, misspellings, incorrect spacing or punctuation? Even in all the snippets too? Because, as an author, those drive me nuts when I miss them. (And I can’t spell, so I miss ‘em plenty. 🙄)
That’s what I shoot for anymore before I publish these days. Which is why I haven’t published much of anything in a while, it’s on a backlog of playtesting.
Creating Epic Boons on DDB
DDB Buyers' Guide
Hardcovers, DDB & You
Content Troubleshooting
Gut reaction is that you shouldn't publish on D&DBeyond until you have it as a google doc (or alternatives like gmbinder.com and homebrewery.naturalcrit.com ). D&DBeyond also has coding limitations that might not even allow some mechanics you want to use. This helps with spelling errors as mentioned previously.
As far as "game balance" then the short list I use and consider fundamental regardless of game system. It also assumes you have an intent or theme decided upon.
The first considers the action economy of not just the individual but the group. If it is causing one player to take an extraordinary amount of time compared to the others then that is probably an issue. This isn't restricted to mechanical things like adding additional actions. This could be rolling more dice thereby increasing the time to count then deal with math (yes, the physical ones), needing to sift through a table on some page constantly, or trying to figure out a longwinded spell effect. All sorts of things can do this including features currently in the game.
The second is where you can analyze with the pillars D&D 5e uses if it helps. Ideally you would cater your homebrew to what you player group is used to, but if you want other people outside to use it then it might need to be a more balanced spread (I personally weigh exploration and social together vs. combat but you don't have to). If you made weapon 1 deal 1d6 piercing damage, and weapon 2 deal 1d6 piercing damage then one of these weapons has little use in play considering they aren't mechanically distinct. Flavor is free of course, but shouldn't be necessary to make up for boring duplicates. Additionally, if there is a class feature that rarely comes up due to requiring far too many conditions to be met, so it also fails to have a use in play.
The third is probably the easiest, however it doesn't mean just damage/healing. If you made a cantrip that manipulated terrain on the same level as Move Earth then it is still too strong considering the implications of world building. It should also be noted that Mold Earth cantrip exists with minor terrain effects (from Xanathar's). What you have mentioned in your post is mostly concerning this. I also see some d&d optimized youtubers compare these things to a Warlock spamming Eldritch blast as a baseline. You could start there instead of trying to grind out so many encounters.
What I don't recommend is that you place a pedestal so high...even official content can be questionable at best. I usually feel about 1/4 or 1/3 of the official stuff fails at what I mentioned above for "game balance." Some of it I think is unintentional, like Truestrike, some of it is intentional, like Fireball, that they wanted to be "iconic."
Though I 100% agree with the others, my opinion is a little different. Posting here is a good way to get some of those varying viewpoints and opinions that may not otherwise show up while testing depending on your group.
Honestly, posting my stuff for feedback was one of the best things I did, even if "incomplete". I learned so much about the game and platform that I would not have known otherwise and it got me to the point where I have formulated my own process to streamline designing, much like you have.