Sure, I was just saying I'd like to hear the specific reasons for it. As in: has it been determined how long the most "efficient" hype cycle is? How much extra revenue can one anticipate for every dollar of marketing spent before...say 3 months from launch? Stuff like that would be interesting to read about.
To my understanding, publishers found that the vast majority of pre-release marketing is only paid attention to with the initial debut trailer, and the content very close to launch, so they pretty much collapsed the marketing cycle to fit that. This also lets you display content that is very polished in all your marketing materials instead of having content from a long time ago coming back to haunt you.
A lot of the most effective marketing also comes from betas and streamers, so they need freely playable content that doesn't have a ton of bugs/issues, which only tends to happen near the end.
On top of that, direct marketing is both very effective and cost effective, and they don't need to do ad buys as far in advance for that to time up with major sports events and the like.
Finally, service games are a big part of this. You don't want to distract from your current games, because you want people to keep playing them as long as possible. Similarly, you actually want to be releasing a lot of information post launch about your existing titles and the updates they're receiving, so your marketing is often focusing on that instead of what is releasing next.