I don't think that letting players create bad characters ("failed builds") is an example of bad design per se, I think the reason that leads characters to fail is what might be.
A game is often nothing more than a big puzzle. You, as a player, have the role of finding the ideal solution based on the pieces you have at hand. In some cases, there is a single combination of pieces that allows you to complete the challenge. In others, there are several possible combinations and discovering these combinations is part of the fun of the game. But whatever the case, it is undoubtedly the designer's responsibility to make clear what each piece represents.
If your puzzle can be assembled in several different ways, but in all of them you always need 3 specific pieces, these 3 pieces have a completely different weight from the others. So, these pieces would obviously need some kind of highlight: the player needs to be able to realize (without needing to be psychic or restart the game several times) which are "the rules of the game". That in the case here, would be "assemble the puzzle using these 3 pieces as a base". This applies, for example, to skills in an RPG. If your game has 16 skills but some of them (or a specific combination of them) are fundamental for you to finish the game, then it's your role, as a designer, to make that clear. If in your design you imagined 3 primary ways to advance in the game (say for example, combat, stealth and diplomacy), the player should be able to realize this clearly and quickly by the design of your game.
Several games offer skills related to persuasion. But in the overwhelming majority of them, this is a completely secondary skill. Why would it be different in your game? If this is a skill that has great weight in its design, it certainly needs to be highlighted somehow. Not to prevent the player from making "stupid" decisions, but because the player first needs to be informed enough to be able to realize that the decision can be stupid. Think of an RST game. You have several types of units and different constructions. When you start the game, you notice that there are air units, and that not all troops have long-range attacks. You, as a player, can make the foolish decision not to create any units capable of dealing with air units and be annihilated, and that's on you. What cannot happen is playing for 20 hours and suddenly, air units appear in the game out of nowhere and you're wiped out.
Good design, among other things, requires two concepts: consistency and telegraph.
Consistency is a little obvious and self-evident. The rules of the game need to be consistent overall. Not to the point where you become a slave to your own design, but to the point where the player can trust that the game will follow certain principles so that they can create a successful strategy that doesn't depend on trial and error (at least, not often). Interesting and clever games are those in which you manage to put together a strategy previously to deal with what the game will throw at you. So that when you fail, you can recognize that the failure was in your strategy and not in the design of the game that put you in an unpredictable situation.
And here comes the concept of telegraphing. This idea is very present in fighting games - most character attacks have certain frames (key frames) that show which action will be performed by the character. The more powerful the attack, usually more evident (and time-consuming) is this animation, so that it can be identified by the opponent. In other words, the game tells the player what is coming, so that the player can prepare to deal with it. Obviously the player can make wrong decisions that put him in a situation where even if he is able to realize what will happen, he can't do anything about it (like for example if he is in the middle of a jump and realizes that a powerful anti-air attack is coming), but this is just part of what makes the game be interesting.
A good game, with a good design, is the one that always informs the player about the options and the tools he has at his disposal, so that he can make a conscious and informed decision. If certain skills are more important than others, this needs to be clear. If it is impossible to win the game without a combat skill, it needs to be clear. If the design of the game's quests follows a specific theme, this needs to be clear. You should not prevent the player from making mistakes and dumb decisions, but you need to give all the information to your player so that he, even without knowing anything about the game before that moment, has the potential to win any challenge that appears on the primary path of the game.
It's okay to have extra challenges or alternative ways to complete certain quests or missions that could only be discovered by someone who has finished the game before (or someone who was very lucky), as long as this is not present in the "core" of the game. If upon arriving at a challenge the player was unable to continue playing because he wasn't able to guess a specific and arbitrary decision of the game's creator, then there was a design flaw here. If the player needs some knowledge that only the game creator (or someone who has already played the game) has when arriving at a certain challenge, the game has failed. It is a line that often seems blurred, but if you confront it with practical examples it usually becomes obvious.