I do, but I am not sure if it is answerable in those terms. Too much or too little cannot really be judged for something that isn't quantifiable. I like to approach this from the perspective of player experience goals. In a historical strategy game the player generally expects two things: first, to play in a world that plausibly reflects real world history, and second, to be able to have influence over the world with the gameplay decision they make.
I want to satisfy both expectations, but they inherently contradict each other: if the player can influence the world, they are theoretically capable of producing a world that does not reflect real world history. Another conflict is between historicity of processes and historicity of outcomes. You can easily force historical things to happen but then the artificiality of these events will be apparent to the player, which in itself is not historical. Conversely, you can accurately reflect some historical processes but they can fail to produce the same outcomes as in real world history, which also feels wrong.
In the best case scenario, game rules and goals are designed in such a way that the player wants to play historically. That is in my opinion the ideal experience for the player: you feel rewarded for sticking to history by positive outcomes, or you don't feel punished for picking optimal strategies by ahistorical outcomes. Around that, you can buffer the freedom of the player by scripting the AI to stick more closely to history, which players usually more readily accept, and by using elastic mechanics like expansion stability that allow you to stray from history to some degree with only small penalties, or to a larger degree only if you are good enough to overcome the penalties. The gradual nature of the penalty, and the sense of accomplishment from overcoming its more serious forms, makes the existence of a penalty that forces you to behave in certain ways less limiting and less frustrating.
Still, sometimes heavyhanded scripting is necessary. I'd like to keep the instances where this is used as infrequent as possible, and instead pick the most high impact situation to apply them. The most prominent examples of this are civ spawns that are guaranteed to happen, or the few big conqueror events that exist. I also usually try to exhaust other, less invasive methods of making certain outcomes happen, before resorting to scripting. And even when scripting is necessary, I try my best to make it context sensitive so it isn't completely jarring.
Two important design principles I have learned over time come to mind in this context. The first is what I call the Inertia Rule, which basically says that consequences of scripted events tend to stick around much longer in the game than they did in actual history, because of the number of years covered by a single turn. The second is that we only know one history (our history) on which to base the design of a historical game. That is important for several "what if" scenarios where railroading needs to be enforced because we know what follows the true course of history, but have no idea what follows all the alternative courses of history, so there is nothing to put into the game for those.
A common example for that is the recurring request to make the spawn of most medieval European civs conditional on the stability and existence of the Roman civilisation. On the surface, this makes sense: historically, these nations only existed because the Roman empire collapsed and various other states started filling its power vacuum. The fact that the game ignores that makes is less historical in process, but there is still not really any other option. European civs would go on to shape the remainder of history not only in Europe but most of the world. If they did not spawn, they would have to be replaced with... what? The only answer is nothing. So will northern Europe remain continually empty? That is also not historically plausible. Even if the Roman empire had not collapsed as it did historically, states and cultures in Europe would have continued to evolve, and e.g. native cultures in the Americas would have continued to evolve without European colonisation, but we don't know how. Any guess how they might have is also inherently ahistorical. Plus, if you imagine the perspective of, let's say, a player starting as Inca or America: they expect Spain and England etc. to be around, otherwise they're not having a historical experience.
One last thing that I didn't mention so far are ahistorical goals: I see those as a good tools to represent civilisations that did not reach their full potential because they fell to foreign domination. In that case, the ahistorical goal is a form of challenge, where the conditions of the civ accurately reflect history and the challenge exists specifically to overcome them as a sign of player skill.