Premise
The current system has a BN auto-evaluated at 30 nominations. The changes to the BN evaluation in the past year involve only mentioning objective DQs/objective issues in the maps that a BN nominates. I believe auto-triggering the evaluation based on disqualifications would allow the NATs to hone in on the problem much earlier than counting on nominations.

Reasoning
I believe assessing DQs instead of nominations align much better with the current system where objective DQs/objective issues are much more prevalent. Realistically speaking, a BN who nominates more will be more likely to get a DQ down the line before reaching an arbitrary nomination quota.

A BN, in the current system, has to reach 30 nominations, and usually, if a streak of DQs happen, depending on their luck, more is bound to happen. Instead of getting whacked when they reach 30 nominations, the BN gets notified earlier that something's wrong with their workflow, and can fix their workflow much quicker before a heavier punishment dawns on them.

The NAT will be more hands-on and give this particular BN more attention and before anything gets worse. Most BNs are also able to maintain a clean sheet of activity, so objectively speaking, the 30 nominations quota seems a bit of waste considering that a BN will get evaluated every 3 months anyway. There is also a potential of less evaluations to run through as a NAT as my proposed system is more efficient.

System
An auto-evaluation is added after a BN crosses a certain threshold of nominations/DQ. Let us characterize this threshold as less than the average (i.e. if the average is 5 noms/DQ, then the average will be 4 noms/DQ, for example)

In this case, the DQs are characterized as severe DQs, which exclude minor DQs (0/0s) like audio, tags, OD/AR change, etc. The NATs will then evaluate the BNs activity and give recommendations on how to reduce sloppiness. The overall evaluation system will remain the same, except that the auto-evaluations trigger on DQs.

Q&A
Q: Doesn't this punish active BNs?
A: On the contrary, active BNs get a check much earlier to slow down, instead of going until 30 nominations and finding out the hard way. I believe active BNs and NATs can communicate much better this way and act as a sanity check for both sides.

Q: Doesn't this mean less quality checks?
A: As mentioned above, the quality checks complement the current system better where objective DQs/issues are pointed out much earlier, instead of an arbitrary amount of nominations. If anything, there are more rigorous quality checks.

(to be updated based on discussion)

I am not sure if such a system has been thought of before, however, considering the current state of NAT, I believe this is worth consideration. I would like to hear the NATs and the BNs opinions on this, and I encourage people of all modes to comment instead!

kinda replying off immediate first impression but I feel like this shouldn't be prioritized atm?

most things mentioned in any given eval made it into ranked already. imo once other ways to explore incentivizing qa have been done then maybe it would be time for this

I don't feel like this is a necessary change.
Current system works fine.

For transparency purposes, we've been in talks with implementing some sort of "auto-evaluation" system for almost half a year now.

The purpose of this system would be two-fold:

- Easing the evaluation workload by not requiring much (if any) manual input if a BN seems to be performing flawlessly according to their metrics.

- Allowing BNs who are known to work well free reign, without having to worry too much about their evaluation if they are aware of their excellent performance already.

I would NOT make this only work with Disqualifications, or generate a card automatically if a certain threshold is reached. This should always be the decision of a human, not a machine. There's more nuance to DQs than just the mistake itself. Some s fearmonger over DQs, even minor ones. This would only add more needless stress to the mix.

We already approach s informally if we believe their performance is somewhat shotty (or, at least Mania does). More often than not, if someone is grossly underperforming, they would know even before their eval comes up, to see if they can correct course before that. Think that's a more "humane" solution to the problem being described above?

if u are talking about efficiency then this new proposal is still not scalable, cuz more BN activity = more work for NAT = we need more NAT? but not every BN is willing to become NAT, and u can't force them, so do we cap the BN size due to lack of NAT?

u can't 10x or 100x the BN population with system like this.

i think the effort should be put to improve BN quality of life (reduce effort needed to check/automate some things server side) which ultimately reduce mistake by BN and allow BN to focus on the map and not the boring stuff no one wants to check like meta, files, etc.

other examples be like perform meta fix post ranked without affecting players (or remove it completely from BN responsibility to a dedicated meta team or AI bot, which also enforce better consistency)

we need reduce workload on BN and on NAT through automation/separation of concern/leverage AI, not add more work.

correct me if wrong, but majority of DQ/Pop is related to objective stuff like meta/files etc that can be automated