What method do you use to test new material/homebrew

[stoutstien]

I realized I've used the thief rogue as my "bar" for damage output and utility. When I'm tinkering with game mechanics I always use it as my go to proof.
What methods have y'all found to work well?

[Galithar]

Small steps and play testing.

Creating a new class with new types of abilities? Give the ability to one of my players as a free boon (usually a quest reward) and let them know it's play test material and may be altered/replaced/removed at anytime. Nothing beats a field test in my opinion.

[R.Shackleford]

I realized I've used the thief rogue as my "bar" for damage output and utility. When I'm tinkering with game mechanics I always use it as my go to proof.
What methods have y'all found to work well?

The Rogue is the class that I think is as close to perfect.The sheer versatility when it comes to build options (str or dex and int, wis, or cha). You can be a support character, a melee damage dealer, an item master, a skill monkey, a tank, or a ranged damage dealer... There are just so many options that one class gives you.

So when it comes to martial classes I tend to judge them based on the Rogue. This is one of the reasons the fighter is so "meh" to me, there is so much more it could have been.

Anyways...

For the most part my groups judge homebrew based on fun.

[Kane0]

1) As much system mastery as you can scrounge up. Just having more experience and breadth of knowledge does a lot of backburning for yourself
2) Community eyeball. Run the numbers yourself, then have others look at it too so they can run their own numbers and provide input. The earlier you can start getting feedback the better.
3) Isolated playtest. Sit down and run a minigame, just on your own if you have to. The greatest math on paper might still feel terrible to play at the table
4) Staggered implementation. Introduce in-game changes in the smallest packages possible, which makes it easier to, implement, measure and adjust as the table gets a handle of it