The IP. In today's market, established IP will always be a safer bet than new IP. Shenmue is probably at its highest peak since 1999 right now and easily its most accessible. Even middling sales and returns are safer than a complete unknown, and YsNet have the rights to create all future instalments. There is no easier access to a legendary IP than this, if you want to take a chance on it.
There is, I suppose, an argument against this in all of 110's projects being new IP, but that's because none of that legacy IP are actually available to these developers. They're very much leaning into "Hey, do you like Ninja Gaiden? Well we've got a new game from some of the creators of those games" for Wanted: Dead, though. If they had access to the Ninja Gaiden IP, would they just be making one of those instead? I think so.
The technology. Shenmue IV is a direct sequel and would build on already developed code/assets. Much of that could go out the window with a new IP if it shares absolutely nothing in common with Shenmue. This is another reason why sequels and sister-IP are prevalent and a surer bet than brand new projects, because you're starting from the base of an already shipped game rather than a bunch of concepts and ideas that might not pan out.
Unless you are backed by a huge financial entity willing to take the hit if a totally new project fails, then more often than not you'll see developers looking to salvage and repackage stuff they've already built just to ease the technology burden. AM2 did this with Shenmue and Virtua Fighter. RGG Studio did this with Judgment and Yakuza. FromSoftware does this with Elden Ring/Sekiro/Bloodborne/Dark Souls/Demon Souls. What have YsNet got to hand? Shenmue assets, Shenmue code. There isn't really any point making a riskier not-Shenmue or a half-Shenmue when you could just make... Shenmue.
If you wanted to make some changes, doing them within an established IP space would still carry less inherent risk, which is why you see series gradually changing over time after several instalments.
Even with a blank cheque to create SEGA's next big game in the 1990s, Yu Suzuki decided to make an RPG spin-off of the already established and popular Virtua Fighter. Only relatively late in production did it become a new IP, and even then the VF roots run extremely deep in both its creative and technical composition.
The pitch. Shenmue IV remains the last game we know YsNet were pitching. There's nothing really more to it than that. Even if Deep Silver turned it down, they were free to continue developing it and shop around until someone else bites. We have no indication that they just gave up after the first try, and generally when you're trying to get a project made you hear plenty "No"s until you finally get a "Yes".
Knowing that 110's CEO has a high reverence for Yu Suzuki and Shenmue, and for all the other reasons above, if anybody was going to say yes to YsNet, it would be this guy.
A variable that could upset the whole theory is that YsNet's next game is something smaller and simpler than a Shenmue title. However the proven expansion of the studio indicates they're going bigger rather than smaller... or are working on two projects, which starts the speculation game all over again. It could also be that Yu doesn't actually want to do another Shenmue, but I find that hard to believe and it doesn't match his actions or words on the subject either.
Similarly, I can't predict the nature of human choice. There's no guarantee either Yu Suzuki or the 110 CEO would do what is "logical" as laid out in this post, but there's little actual incentive to take the curveball approach as I see it. Could be right, could be wrong, but if you walk through the practicalities of game creation, the logistics of financing, and what the publicly stated intentions of Yu Suzuki and the 110 CEO are... it's Shenmue IV.