Most players play variants of the same bidding system because learning a significantly new bidding system is a significant amount of effort. Most players aren't very good even with the only bidding system they know. Ask them to learn a new one and they would be perpetually confused. Most players aren't that serious.
I'm now going to ignore the mass of non-so-good players and focus on the advanced and expert players.
In most countries, the main bridge-tournament organizing association regulates what bidding systems are allowed, precisely to limit this strategy. You're not allowed any bidding system you want, and bidding systems and or conventions that are likely to be unfamiliar to opponents are banned, particular in games where each pair faces each other pair only for a small number of boards. (In North America, it's quite common for good-to-expert players to complain about the regulations put forward by the ACBL on the grounds that it bans legitimately useful conventions that are not that hard to prepare for.)
In tournaments without such restrictions, which are mainly very high level tournaments such as world championships, players are required to disclose their bidding systems to their potential opponents several weeks ahead of time so that their opponents can prepare.
Even so, there are some practical reasons why bridge players don't play highly unusual systems all that often. The first is that it is actually hard to design a good (human-playable) system that's very different from any of the commonly played systems. (Minor variations don't do much to confuse opponents.) If you make up a completely new system, even if you are good at it, the system is likely to have some significant weaknesses (which may simply be that it's very complicated and you need to spend a lot of mental effort remembering what your bids mean).
A second is in how professional bridge players make a living. There is no prize money in bridge tournaments. Rather, bridge players are paid by wealthy sponsors who want to win (or at least do well) in tournaments by having good players play as their partner. It's unlikely that their clients want to put in the effort to learn an unusual system or that their clients will be able to play the system well. Even the professional players who usually play with the same partner (because they are hired to be part of a team where the client plays with their professional partner at one table and two professionals play with each other at the other one) change partners sometimes when their usual partner is unavailable for a tournament (or does not want to be vaccinated and hence is taking a possibly temporary forced retirement).
(Most players at all levels play with several different partners. I once played four different systems in a week; I can handle that but most players would be quite confused. And there are subtle things I tend to get wrong when switching systems.)
A third is that a significant amount of bridge is played with matchpoint pairs scoring, where your score comes from how you do on each hand compared with a large number of other players. If you are a very good player, you play the cards better than everyone else and can score 60% (which is usually close to enough to win) every hand just by being in the same contract as everyone else. Playing a different contract than everyone else introduces additional randomness, which you don't want. Certainly, there will be games where the hands suit your system, you get to better contracts, and score 65%. But 60% is first, and 65% is also first, and there are also games where the hands don't suit your system, you get to worse contracts, and score 57%. (Let's suppose your system is actually better, so you lose less when you are unlucky than you win when you're lucky.) But now 57% is third.