You are assuming my players don't know I do this. I'm quite explicit in saying that I do this.
My game's not much about combat though, so...
If I designed a combat magus, and the SG then deliberately prevented me from being in combat, I'd be pretty annoyed. I think my character design is a pretty clear signal of my idea of fun.
Sure. But if you designed a combat magus for a game about politics, and we're told explicitly what the saga was about, then what?
Oh, sure. That's just not the vibe I took from the earlier post.
That's my point; it makes it much less fun, if the players know this is happening. It is OK in small doses, when used to compensate for catastrophic creature/plot design mistakes by the storyguide. But it should never be the default option.
The problem is, if it is the default, that the players will come to feel that (or are explicitly told that) if they have a clever idea (or a good dice roll) which unexpectedly foils the NPCs, the storyguide will just shuffle things around to counter the PCs good idea/luck. It doesn't matter whether we are talking about a combat, political, or other sort of story.
Clearly your mileage varies, but again you seem to be assuming that the point of this is in some way adversarial. It's not, and, no, it simplify doesn't have the deadening effect you suggest. Your assertions that it must do not accord with my experiences.
I think you're discounting the frequency that many of us have encountered the "killer GM." Changing things at the last minute smacks of the moves of a "killer GM" and the relationship between players and a "killer GM" is entirely adversarial. You really need to examine closely as to whether or not changing things at the last minute is going to make the overall scenario better, and what better means. You mention the game contract a lot in your posts, so I tend to think that you make this adjustment to encounters to keep them challenging explicit in the games you run. Do you always do this, so that the PCs never overcome an obstacle easily, or just sometimes? What is your rubric for deciding to change things last minute?
I generally prefer to stretch things out in the campaign rather than adjust the encounter. It's so much simpler to handle things in a broad story arc, and it doesn't invalidate the PCs actions entirely, and is much harder to detect what some players would call shenanigans. Players like to know the rules under which they play, the more the SG controls those rules and the world, the less the players will enjoy the overall game.
My rubric for when to change things is based on the feel of the table. There's a cure for the Killer DM. Just never play his games. It's that simple. The Ars Magica combat rules are not balanced, or fair, and if you look back to 2nd edition, you can see they were deliberately not designed to be. Faeries in ArM5 can change role in midcombat, deliberately, and it's not to make life easier for killer GMs, it's because simulating the feel of faerie stories, rather than the feel of realistic combat, is the goal of the book.
Also, I'd point out that your idea that I'm controlling the world less if I don't adjust monsters during combat is false, because I create the monster in the first place and if I were a killer DM who did not want to manage the combats in this way, then my monsters would just have greater combat power. I'm not clear how that's fair, and I'd suggest that you saying what my players will or won't enjoy is you projecting you needs onto my game.
Your last statement works two ways. You see that, right? 
Yes, and? 8)
The OP asked : "How do I make it easier to design adventures?" and my answer is "Don't sweat the detail on the monsters." Yours seems to be "Stat everything, because if you make stuff up as you go along then you are a cheater and the players hate that." How does your style assist the original poster? So far as I can see, you're just telling him to put in more work.
Ars is ridiculously overburdened with setup: why demand more?
Perhaps it is my experience with the Boy Scouts.
Adventure design is easier when I know who is on the adventure, before I've fully designed it. Whether I have to contrive story events to make it so that certain characters are unavailable, or engage in some meta-game discussion with the players is irrelevant. Where I have the most problems with an adventure is when everyone wants to bring their magus along on the adventure, oh and their companion character, too. If players want an adventure like that, I'll oblige, but it will be different than one designed for one or two magi who bring assorted grogs and one or two companions along. I'm flexible enough to allow for some last minute changes, too.
I never said I made stats for everything, but I do stat the really important stuff. I also design the adventures for the characters going on them; a mix of challenging and easy encounters placed to make things interesting[1]. Not every encounter has full stats, but the ultimate, and possibly penultimate ones do. My method involves controlling the characters involved and designing accordingly.
I was a Scout leader for eight years (we dropped "Boy" from the name when we let girls in in the 1970s). 8)
[/quote]
I agree wholeheartedly with the idea that you should contrive moments for characters with particular strengths to shine, if that's what you mean by designing to the characters attending.
Scouting, here in the US, I think is still segregated. I can't be sure. I left before getting my Life badge, as I wasn't going to do the work necessary to be an Eagle Scout.
Sure, I want characters strengths to shine. Of course, sometimes that means highlighting another character's weakness.
At the risk of stirring the hornet's nest further, I think that AM5 is designed and sustained editorially as something of a sandbox game, and one of the principles of a sandbox is that the GM has everything figured out ahead of time and the players can do whatever they want, and go wherever they want, and whatever they encounter will be what is "actually there," by which they mean, "what the GM knew was there before we decided to come."
This is, I think, in contrast to a lot of more free form or story-based gaming where, for example, the GM changes things as the players are moving through the story because everyone at the table agrees its more fun that way. Tim's example of a dragon which suddenly develops acid blood is a classic example of this. It's not done to screw over the players, it's done because, hey, we have been building up to this dragon fight for three hours now, and we want it to be exciting. What is written down on the dragon's character sheet is, honestly, not especially sacred.
There are a lot of games today that do this, and there's a long tradition of it, from the recent Fate to classics like Amber Diceless and plenty in between. Every table is going to have a different tolerance and enjoyment level, but for my part, I very much enjoy both ends of the scale. And although AM is officially very much a sandbox, in which a GM can know a bewildering amount of detail about the setting before the players even make characters, it can absolutely be run -- and can be enormous fun -- as a more free form experience where no one even rolls dice, and where the GM and players are all making shit up as they go along.
It wasn't about you JL. Roberto is designed more for verisimilitude than anything else. A magus trained for combat. He came from another game and was ported into your saga by invitation of the then current players. And he wasn't the only combat magus. Just the better one at a younger age. And I like roleplaying him, with his assertive demeanor and being difficult to intimidate. The face off against Rotigers was pretty cool. Now I ported him into my saga, and granted another player liberal authority as a beta sg for a segment of the game so I can play him as a PC. With verisimilitude.
Which inspires a point I'd like to make about storytelling.
The GM should always be willing to adapt to the changing tastes and changing natures of his changing players. Especially in a PbP game where half of them come and go like the wind. I favor Plot Potentials rather than Metaplots. Verisimilitude. Let the stories revolve around the players. The GM will still do stuff he wants to do, and he can create potentials and encourage players down certain paths. But it is a give and take.
And another point: stats for an encounter should be consistent throughout the dice-rolling phase of that encounter. I have had it once whenre I was given the opposed sore I had to roll against, I beat it by one point, and the SG added two points to the target's Mentem score (a companion's Entrancement roll versus the MR of a magus NPC). If it is a combat encounter, stat the creature and be consistent. You can stat the creature/npc at the very last second if you want. I often do it mid-way through the adventure, so I can gauge the capabilities of the characters and their group dynamic and fit details to work with what players have accomplished. But once dice roll, stick with it. If they defeat the advisory in a quick anti-climatic fashion, deal with it. Sometimes this is a good thing. It rewards players for planning and designing and tactical prowess. Even if none of these things are theirs, they get that feeling.
And you can't always upscale the combat challenge like this was Dragon Ball Z. Instead, a saga should involve a variety of challenges so that players often have to rely upon each other or occasionally seek outside assistance. I don't run an "Adventure Saga". I run a saga that has adventure stories. There are political stories, romance stories, moments of family drama, and right now I am running a mystery story.
Truth is a three edged sword, Marko. Verisimilitude? Really? I'll agree to disagree. Roberto is a well designed character, but he didn't have the appearance of being real when viewed in toto. Verisimilitude, in a game, is a shared experience. There are places where I wasn't feeling it for Roberto. I said it there, and I say it here, I felt you were playing me and the meta-game, and not playing the character, certainly not the character that would exist in that particular saga. You were the player least interested in the whole Guernicus conspiracy thing, and you played Roberto that way.
I don't disagree with you here. Like I said, if an adversary is too easy, it's just an opportunity to demonstrate he was a lackey and not the boss.
Wow. That sounds like a lot of work!
And for the record, not even sandbox computer games do this. Because it's a lot of work.
The world revolves around the players. There's no point in going into anything beyond the most cursory of detail for any area that is outside the perception of the players. All the simulation rules are there for the players and the things the players regularly interact with. Beyond that, handy-wavy. It works for GTA. 
If you look under the hood of a recent AAA sandbox computer game, you'll find that a great deal of what you encounter in the game is actually assembled on-the-fly just outside of what you can see based on what you're doing, what kind of player you are, what behaviours you've been engaging in, etc. The trick is placing them there so they look like they've been there the whole time. The reason being trying to actually simulate an entire world is a huge waste of processor time - so instead you just simulate what the player can see and then guesstimate the rest.
You can trick people that the simulation exists even more with some anchoring techniques (give them some big-ticket items to latch on to and then only track the big-ticket items), which work equally in pen-and-paper and computer-game sandbox just fine. The trick is deliver with confidence. If someone contradicts you, just politely inform them that they must have remembered the event wrong. This is only a problem if one of your players is a habitual chronicler of everything... and if you have one of those just get them to confirm details instead.
Computer games excel at delivering with confidence. Anyone who has ever used computers will tell you that computers are always incredibly confident in what they put on screen... even when they're 100% wrong.
Memory is incredibly mutable, and the only internal mechanism it uses for anchoring is narrative. As long as the narrative feels intact everything else will fit into place - even if it's totally different to what it was last time around. This is why eyewitness accounts are so unreliable - and its a phenomenon you can abuse the heck out of as a SG. It saves you a ton of work.
The best part is if you screw up details you have four realms to blame for it. "That forest wasn't here last time..." "A faerie needed it for her story." ... Now I want to make a mage that randomly plants forests.
I'm also not assuming that it is an adversarial thing. The worst offenders are actually nominally narrative/story-based story-guides. The problem is when the storyguide has A Very Clever And Interesting Story plotted out, but the players either through luck/idiocy/or cleverness run off-script. There is a great tendency for the storyguide to undermine (however well disguised) whatever the players have done, to put them back on-script. Otherwise the Very Clever And Interesting Story will not be told! However, it is much better (in my experience) for the story-guide to instead allow the story develop in the direction that the players take it via their actions.
See, this is actually the sort of thing that rapidly becomes problematic. It seems like a great idea for the storyguide, to say, create an opportunity where the Intellego specialist has a role. However, the question is what to do, if the players unexpectedly instead decide to resolve this opportunity with the Creo Ignem specialist.
One option is to try to change things around so that the Creo Ignem solution effectively fails.
Another option is to accept the player's solution, and try to create some other opportunity for the Intellego specialist.
A third option is to accept the player's solution, and take the story where it "naturally" develops, regardless of which player character will subsequently "shine", or whatever.
I think that the third option is far and away the best option.
What I advocate is "adventure design" that is essentially limited to the "beginnings" of adventures. The storyguide knows the start of the story, and he knows what the NPCs are trying to do. But the actual story depends on what the player characters do (or do not do). So, there is never a wrong thing for the player characters to have done that needs to be "corrected" in order to get the story back on track. The story is what the player characters do, not what the storyguide does.
The way I see it there are several situations where you can talk about the SG "cheating"
One is a perfectly well-designed and balanced group, who go on a story designed by the SG, with relevant stats. At some point there is a 'short-circuit' of the story. It could be a flaw in the design of the story. A SG able to think on his feet should decide whether to fudge things or go with it. It could seem like the adversary had made a mistake, noticed by a perceptive character and cleverly taken advantage of. If it's a total brain fart then the SG should agree with his Troupe that this mistake does not make sense and change things, allowing the players to adapt. Flexible players would readily adapt and ignore those wrong turns of events.
If it is a ridiculously lucky roll that ruins the climax? Well, some players love to win like this and would feel a SG somehow annulling this is cheating. As mentioned before, the SG could also go with it and see how the story develops from here. or simply change small things (it was not the big baddie, but a lackey!). If the players had not realized the power of the enemy yet - eg. by winning Init and striking first blow - it is easy to get away with. Still, some players call this cheating. So the best things may be to adapt the story, and let the events have consequences. IMHO good players also roll with this. A former player in a former saga most often rolled consecutive 1s every round (we call this "a Brian roll") so many combats ended quickly. His magus was Overconfident and this made it no less so. Because of success in the very first combat he kept on using mostly a completely mundane dagger, whit which he was barely skilled. And he also rolled very low occasionally and the magus often got hurt, and spent full seasons recuperating. My point is, the consequences of his luck was that the character developed his personality flaws further. Such matters could also have social consequences, reflect on Reputations, draw unwanted attention, or step on the toes of other magi's projects ("why did you kill that dragon, I was using that!").
Another problem is when a single character's design is skewed and it makes certain stories impossible to not have circumvented. It could be an Intellego specialist who makes mysteries and investigations trivial. Or a combat monster. Or a careful, cowardly even, magus with lots of escape possibilities - although this just means stories don't happen at all because the challenge is dodged. Or my absolutely most hated design: The overly defensive, resistant mastery type.
So if the SG designs a [Whatever] type story then the expert in [Whatever] goes along and the rest have nothing to do. But it should not be designed so as his specialty never comes into play. IMHO this is a problem needing to be dealt with in the design phase of magi and saga. What do we want to play? How can we do this. If one magus is a combat monster he wants and expects a high challenge, but nobody else wants to go on these adventures because it will kill them. These problems can't be solved with simply letting the mysteries happen while the Intellego magus is foisted off on an errand. And problems may even get bigger during the saga's progress since some magi become even bigger experts in some things the others don't touch.
I don't have any solutions except to again point out that certain actions have consequences and Reputations develop forming or spoiling social relations. Other magi can decide some expert is a threat. The combat monster killing the dragon draws unwanted attention. The resistance master looks mighty suspicious to Quaesitors. The escape specialist is labelled a coward and few magi want to associate with him.