Personally I'd say best options are:

June (when kids are getting out of school)

September (allowing them to have a launch period and then a separate holiday period)

November (the typical holiday launch time)

Of those I think typical holiday launch time of November is actually the worst. Next year I'll be looking for either a reveal around February and release around June, or a reveal around May and launch around September.

And considering it's been many years since the last 3D Mario and Mario Kart, and those are the two obvious launch titles, ya gotta figure they will have those two titles ready no matter when they launch next year so I don't think concerns about software being ready would change their launch date within 2024.

If they launch June-ish have 3D Mario on launch day, Mario Kart maybe a month later (or maybe launch day also), have another big game for later summer or early fall, and one more big game for the holidays.

If they launch September-ish have 3D Mario on launch day, Mario Kart maybe a month later, and another big game another month later as the holiday game, or maybe Mario Kart is the holiday game.

If they launch during the holidays, then just have 3D Mario and Mario Kart at launch.

Having a strong launch title that ideally shows off the platform well (sorry but NSMBU did not do this for Wii U, a 2D Mario is a poor showcase for a new console) is more important that anything else frankly. 

I don’t think there is anything wrong with launching with 2D Mario as long as it’s a unique entry and isn’t followed by a large drought.

The Wii U had troubling sales metrics even from the immediate launch ... I don't think the Switch is anywhere near as successful if BOTW was "NSMB" instead. 

Even Nintendo fans are a pain in the ass for Nintendo on this, they won't let Nintendo off the hook if they don't have some kind of "epic" Nintendo title at launch. 

Luigi's Mansion for the GameCube was kind of the same thing "where's Mario 64-2? Why isn't this Mario 64-2? When are we getting Mario 64-2? Well fuck it I'm not buying until I see a Mario 64-2, I'm not buying a system for a little Luigi spin-off". 

But I guess it is what it is, Sony can launch a Playstation home console with a middling lineup and its no big deal, but for Nintendo they have to prove themselves all over again to their fans every new system even though it's just factual reality that if you like Nintendo IP you're gonna have to buy the Nintendo system regardless of the timing of those games. 

If you're Nintendo you need to make sure you have a new big ticket franchise installment of a series that really shows off the new hardware well. 

2D Mario is huge though, the ones on DS and Wii both sold over 30m. The "New" series of games has gotten tiring but if an exciting original one gets made it'll be among their biggest system sellers I think.

I think what happened was there hadn't been a new 2D Mario in such a long time and it played in perfectly with the more casual gamer approach the DS and Wii were going for, that it was a big deal then. 

NSMBU and NSMB2 on the Wii U and 3DS though ... and even really Mario Run on smartphones ... they sold quite a bit, but I think it was more like "oh OK, so you're doing this again, huh, I just played this a few years ago on my Wii/DS?" type of feeling with a lot of consumers.

And now with Switch it's kind of a different generation altogether, the 16-30 year olds driving Switch adoption ... they grew up with the GameCube and N64 and so on. 3D analog stick Mario is just normal to them, so now we see the gap between 2D Mario and 3D Mario is no longer that big anymore. 

You could kinda see this trend on Wii U and 3DS also ... NSMBU and Mario 3D World sold about the same, so did NSMB2 and Mario 3D Land. Neither NSMBU or NSMB2 I think were the system seller that Nintendo was hoping for (Nintendo banked hard on NSMB2 + cheap 2DS being a huge seller, but the 3DS didn't really see much sales acceleration, I believe it's first year holiday season with 3D Land + Mario Kart 7 remained the 3DS' best holiday season). 

Since this has previously been a topic of discussion here in various threads, I thought I would bring this point up from the last Nintendo investor Q&A

Question: Nintendo Switch was announced under the development codename “NX” in March 2015,
and then the official name was announced in October 2016 prior to its launch in March
2017. Looking back, do you consider this a success in terms of hardware marketing, given
there was a space of about two years between the initial announcement and launch? Do
you think a similar interval between the announcement and launch of a next-generation
platform will be necessary?

Furukawa: Looking back at the release of information leading up to the Nintendo Switch launch, we
announced the “NX” development codename in March 2015 during a joint announcement
with DeNA Co., Ltd. regarding our business and capital alliance (as it related to joint
development of smart-device game applications and its operation, and also the core
system development centered around Nintendo Account). When we announced our entry
into the mobile business at that time, we needed to let people know that Nintendo would
be continuing to focus on the dedicated video game platform business as our core
business. So, I believe that the timing of the Nintendo Switch announcement was a special
case.

We will provide information about hardware and software at the appropriate time for
each product and strive to reach a wide range of consumers.

So basically what I said ... they only announced NX so early (2015) because they were making an announcement for smart phone games and wanted to reassure investors that they were not getting out of the console business. 

As such it was a special case to announce NX that early and (by logical extension) does not mean Nintendo has to announce the Switch 2 on the same time line. Likely if Switch 2 is a 2024 product, they can just announce it in 2024 and launch 4-5 months later if they want (that was the actual length of the real Switch marketing cycle last time).  

The Wii U had troubling sales metrics even from the immediate launch ... I don't think the Switch is anywhere near as successful if BOTW was "NSMB" instead.