Transparency Discussion: Shadow Nerfs

Hiya everyone (staff in particular),

Throughout a bunch of years on this server, I've tried to cling to the belief that staff is relatively (take that with a grain of salt) transparent.
As an extension of this, I'd like to believe that 'Shadow Nerfs' as such aren't a desirable thing.

However, the past years have shown numerous cases where 'shadow nerfs' have (supposedly) taken place. E.g. this post just recently.

The point of this thread is twofold. The first wish I have from this post is a clarification/confirmation from staff that these shadow nerfs do occur.
Secondly, a broader question towards the community, are shadow nerfs a problem? Is it more desirable for these changes/nerfs/fixes to get their own channel through which they are mentioned/confirmed/announced, or is the current state of things (guesswork, really, unless you're logging vast amounts of data regularly) fine?

Feel free to discuss.

Kind regards,

Ainz

 

I realise I forgot to include my own opinion, so here goes:

While I don't think the current state of shadow nerfs is a huge problem per se, I do feel like it would be preferable for changes to be public.

There's a few difficulties I can see with this, however. The first being that ML's droprates have never been public, which can be used as an argument against publishing changes made to them.
Secondly, one could argue that staff already has plenty of work to do and asking them to publish changes made might just add to that pile. Against this, I want to say that publishing a change should generally take significantly less time than the process associated with making such a change.

Finally, I could see people arguing that publishing changes would only further turn ML into a numbers simulator of 'x hours at y meso/hr results in z total profit'.
Here I'd like to raise the example of League of Legends as a counterpoint. Changes here are entirely transparent (e.g. skill x having damage/cd/range reduced by y) yet the meta still always requires a lot of exploration to see what the 'new best pick' is. As such, I don't think ML would run into similar problems.

Sorry for rambling~

 

I found this last bit particularly interesting. Why would empirical data (at a reasonable scale) not be a good indicator of the state of the game in terms of e.g. droprates? Obviously RNG is gonna RNG, but in a large sample size, it should even out.

 

In that case, I suppose my main question for this thread has been answered and the other question thus becomes void.

Many thanks to Kimmy and Pasta for their clarifications. I'll go back to clinging to my undying faith in staff's transparency!

Thread can be locked as far as I'm concerned.