Two nights ago, I got to try Memory Quest for the first time.  It was definitely a unique social experience.  I've never had a roleplaying session like it.

It worked extremely well.  Much of the credit goes to the fact that we had a super-fun group of people who play together regularly.  I think trust issues would ruin this game.  But, whatever; trust issues will ruin a lot of RPGs.  I gotta say I'm pretty happy with this.

Matt opted to be the Agent Player, while Marsha, Mendez and I played the Whispers.  Matt sat on the end of one couch, Mendez and I sat on the opposite end of another couch, and Marsha pulled up a chair, her back to Matt, so she could huddle with me and Mendez.

Mission One

Matt started narrating, we all started whispering about him, and it was really fun for us to guess what he might be up to.  After a little while of talking without getting any feedback or even much eye contact, Matt worried that he wasn't being listened to.  Fortunately, this went away as soon as we started throwing Obstacles at him.

Matt's narration was slower and more detailed than I'd expected.  I enjoyed it, and loved that it gave me time to think before he'd gotten too far.  But at the same time, I was worried: "Oh fuck, he's gonna take 2 hours on this mission, and then when we kick him back to square one, he's gonna be pissed."

Anyway, the Whispers rolled like shit, turned up way too many "no reminder, 1 damage" results, and Matt only collected 2 Reminders before completing his mission on the penultimate Activity box.  His sequence went:
OO*OO*SOOWO*OO*
You can see that between his 2nd and 3rd steps (the *s), he did some probing (S,W), and we hit him with a bunch of Obstacles (3 Os to get through).  In the process, he took 7 damage.  He got pretty gunshy about interacting with the Reminder opportunities we threw at him -- just the kind of mixed signal I was hoping for.

(Best moment: Matt smuggled his replica painting into the museum inside a fake Vermeer, which the Whispers co-opted as a reminder: "You see your children in the painting!  Do you want to look closer?"  We wanted it to be touching and grabby.  Matt: "Hell no!"  Haha!  Apparently it was more creepy!)

So Matt stole the Mona Lisa, replaced it with a fake, and smuggled it across the border.  There was no "approaching Nemesis" narration at the end because (a) he'd forgotten the *s and had checked of 11 boxes, not 15, and (b) while the Whispers easily got the hang of interjecting Obstacles, throwing in Nemesis color was far messier, tending to bring out a traditional player-GM RPG back-and-forth.  It was fun, but would have derailed the game if I hadn't stopped it.

Matt announced completion of his mission.  The Whispers all huddled as I found the key line in the text, which we delivered in unison, "You wake up in the morning.  Where are you?"

"Fuck!" Matt yelled.  I got worried.  "No, it's cool!" he said.  "Now what?"  Whew!  This was what I'd been hoping for.  Yeah, it had been an hour or so, but he hadn't fought hard to win his mission success, so it wasn't like he was deprived of something he'd earned.  At least, that's my take.  To succeed in your mission in Memory Quest, you simply describe how you succeed.  Some part of you has to know there's gotta be more to it.

Mission Two

We took a break for Matt's new mission and the Whispers' new Nemesis.  (We switched from monkeys to genies, bypassing Mendez's suggestion of robots.)  I told everyone how I'd expected each mission to go a little faster.  I asked Matt if he felt like the first round was too long, and he said, "No," but I still gave a little demo of how the Agent Player could narrate faster.

Matt gave it a shot, and really started churning out completed steps before the Whispers had much of a chance to figure out what he was up to.  We started flinging Obstacles to slow him down, then made up Reminders afterward.  This wasn't what I'd intended, but was totally fine.  This semi-randomness illustrated to me that it isn't hugely important to give the Agent stuff relevant to his mission.  Fortunately, the countdown mechanics seem to be just right so that every Reminder is precious, and every extra bit of incentive helps!  So we never abandoned our attempt to understand what he was up to.  Also, it was just plain fun.

This time around, we got in a completed mission (blowing up magnetic North to get the attention of aliens) in about 40 minutes of play.  And 3 Reminders!  Just one short!  Again, "You wake up in the morning.  Where are you?"

O*OO*O*BO* (which was all he had, given the 5 Activity boxes left over from Mission 1)

At this point, Matt said he'd totally be down to play a third round if he wasn't exhausted from recent sleep deprivation... but he was, so could I just tell him the ending?  I obliged, and he found it satisfying.  "Yeah, I did figure that you guys were trying to help somehow, at least some of the time."  Nice.

I'm pleased to say that the basic activity and premise of the game is functional!  Two main things to do next:

1) The easy: I'm streamlining the instructions, adding some examples, and planning to playtest the approach and arrival of the Nemesis, to see whether that's confusing to narrate.  I'll have a revised document up soon.

2) The hard: I need to refine what happens to the Activity and Damage tracks on each Reset and each new Mission.  These actions were inspired by the fictional premise, and make for good drama but potentially broken strategies.  I think it was the luck of the dice that these didn't come up in this session.

Design quandaries -- feedback appreciated

I think it might be good if the third full Mission produced Win or Die outcomes; I think most of the Agent Player's learning strategies will be done by then.  On the other hand, talking strategies could emerge late, on a 4th or 5th Mission -- SPOILER ALERT! -- e.g. the Agent Player going slower to allow more Reminder opportunities, or Whispers Players repeating some vital info over and over to elicit a "What did you say?"

I wonder if I should mention in the text that talking strategies are kosher?  I think I'll put "pace" as an Agent Player sidebar topic and suggest "try switching it up" and leave it at that.

It could be nice to add another formal layer of learning -- a way for the Whispers to signal to the Agent which Reminders have been co-opted.  Something regular but non-obvious, like all the bad stuff comes in odd numbers while the good stuff comes in evens, or all the bad objects start with the letter D or something.

I think 15 boxes is the right granularity, but the balance is really delicate.  If the game gets to a point where it's just a single Step, a barrage of Obstacles and a prayer that Reminders taken outweighs Damage taken, I only want that to happen once; the Agent then needs to die or wake up. 

I think a Reset needs to be, at least in most circumstances, clearly worse than a new Mission.  It's probably safer to keep it as a life-saver only, than risk having Whispers jump in and say, "You wake up in the morning!" in the middle of a mission narration for no apparent reason (to the Agent Player) because they're looking for a mechanical advantage.  That said, a Reset still needs to be playable; the odds on subsequent Missions are tight, so less than "tight" might effectively be "nil".  If a Reset just gets you another chance to get to the next Mission without dying, but won't wake you up, is that good enough?

Probability question: I am correct that the chance of getting 4 heads on 6 coin flips is 22/64, and the chance of getting 4 heads on 5 coin flips is 6/32?

Thanks,
-David

I don't have any immediate insight on your design questions, but you're correct on the probabilities (assuming you mean "at least 4 heads" and not "exactly 4 heads").

- Bax

Er, yeah, "at least", not "exactly".  Thanks!

Re: the design, I'm now thinking that the opportunity to act on your understanding (Activity boxes) ought to vary with the learning curve.  So, as players learn a lot in their first Mission, and less thereafter, the permanent Activity loss should follow suit.  I could just fix a track, like 15-12-11-10... 

10 seems iffy: if the Agent Player does nothing but On With the Mission, that's four boxes, leaving 6 Obstacles.  The odds of getting 4 Reminders through in 6 tries is only 1 in 3, and then the odds are worse if the Agent doesn't grab at every single Reminder.

I think the ideal would be if the Agent's chances are very good if he grabs the true Reminders and avoids the co-opted ones, and very bad if he can't distinguish them.  Random luck may work for this early on (the feedback is clear and immediate), but once it's the only thing to worry about (if the game goes that long), informed choices seem key here.

Any thoughts on patterns (odd/even, letter D, etc.), hints (repeatedly whispering "take it!" or "don't take it!"), or other options for giving the Agent Player "good Reminder"/"bad Reminder" info?

Updated document is up here.