Forum changes: Editing of posts has been turned off until further notice.

Main Menu

no islands in the pool

Started by Paul Czege, September 25, 2001, 09:32:00 PM

Previous topic - Next topic

Paul Czege

Hey everyone,

I ran the second session of my scenario for The Pool last night, and I have to say that if you'd been there, the one thing you'd have noticed is how difficult it was for the players to keep dice in their pools. Three of the four players, Matt, Tom, and Scott, bottomed out their pools during the first session, and spent the entire second session vacillating between zero and one or two dice. At zero dice, they'd take an action that gets them a die in their pool. On their next action they'd gamble it, hoping for a success and a MoV, and since they were often rolling just three to five dice in this situation, they were failing these rolls pretty regularly.

This was very challenging to me as a GM. Not only was I narrating the outcomes of the rolls they'd used to get a die in their pool, but I was repeatedly called on to narrate hosings when they'd bottom their pools out, again and again. By far the vast majority of outcomes to dice rolls were narrated by me. I narrated a hosing of Scott's character Grazel in which his ear was cut cleanly off by a Kriedetempek warrior, and another one where Grazel's charge through a combat scene unbalanced one of his mercenary allies in combat and resulted in that NPC taking two simultaneous sword thrusts from the Kriedetempek warriors he'd been defending against. I continually needed to push myself to think outside the traditional RPG box when narrating hosings, to introduce consequences when the player's intent was "to kill" the opponent that were not necessarily a killing or grievous injury to the character. It was insanely demanding, and I think recognizing how hard it was, the players started to offer suggestions each time the situation arose.

It's hard for me to know why the players were thrashing around so much at the bottom of the pool. I was very very aggressive with scene framing, in both this session and the first one. It's something I've been wanting to push the limits of, and learn to be good at. So it's possible that the situations they found themselves in were all so compelling and protagonizing that they didn't have much opportunity to take careful actions to get dice into their pools, because they so desperately wanted the MoV in every situation they found themselves in. It's also possible they're so insanely curious about the MoV that they wanted opportunities to learn what it could do, and push its limits.

It was fun to have to push my creativity so hard narrating outcomes, but going into the session I was expecting the players would be far more aggressive with their authorial power than they had been during the first session. And in actuality, I think they were definitely interested in being aggressive. It just didn't materialize for them the way they'd all hoped. And the result was a certain amount of evident frustration for them.

I'm thinking about replenishing all their pools up to starting levels for the next session, as an experiment at how quickly they bottom them out now that they're more experienced players. But that's not really a solution. And although I have ideas, I'm not sure exactly how to solve the problem.

I do think Scott's handling of pool management is particularly instructive. One technique he used was to ask for a scene and gamble for an MoV that he used to create the innuendo of an NPC being complicit with his character's killing of Sgt. Aminar Korg in the first session. It positioned him for greater effectiveness in handling subsequent events to have invented this relationship, because he'd be able to twist and influence the NPC in ways that would make enough sense to the narrative that I wouldn't call for a die roll. He got a lot of permanent potential bang from that one MoV.

Another technique he used, when he didn't have any dice in his pool, was to ask for a trait roll to have an NPC do something. His character Grazel had been hosed and disarmed by a Kriedetempek. Scott asked for a die roll, explaining that Grazel's perceived "reliability" was from his shared closeness with other mercenaries, and that one of the NPC mercenaries nearby would turn to kill the Kriedetempek that was on Grazel out of the shared relationship of mutually relying on each other. It was a stretch of that trait, but cool enough that I allowed it, and it worked so well on his behalf and contributed so well to the narrative that during character creation for The Pool in the future I'll probably use it as an example of why players should strongly consider traits that create ties between their characters and others. I'm thinking of it now as kind of The Pool equivalent of Kicker integration with a relationship map.

Still, they did struggle at replenishing their pools, even Scott. We had a discussion about it after the game session, and beyond Scott's techniques for getting the most out of his character's pool by creating linkages with NPC's, I suggested that they needed to get comfortable asking for scenes, and asking for flashbacks, to ask for trait rolls associated with those scenes and flashbacks that build and reveal the character in the context of the developing narrative and add dice to the character's pool. I'm thinking of this now as the real augmentation mechanic for The Pool. It's what creates the character.

And when I think about it, each of these three techniques for effective use and management of a player's pool hinges on player-NPC connections.

Whaddya think? Is it possible to be an island in the pool, and still be effective? How have your players been effective or ineffective at managing their pools? Can you discern anything that may have been a problem with the way I was running the game?


[ This Message was edited by: Paul Czege on 2001-09-25 16:33 ]
My Life with Master knows codependence.
And if you're doing anything with your Acts of Evil ashcan license, of course I'm curious and would love to hear about your plans


Well, from the point of view of someone who's read but not played The Pool, it seems to me that by far the most effective way to not spend all your time "thrashing about at the bottom of the pool"1 is to suck it up, not take the MoV, and expand your pool up to a level where you can gamble it all and have a good chance of success or gamble part of it and have a decent chance of success.  I'm going to go out on a limb and suggest that the intention of The Pool is to make MoV's moderately rare, memorable, and exciting, something that you do when you're seriously committed to a scene and want a memorable and lasting victory, not the result of most successful rolls.

If your players feel like they need MoV's on most rolls, it may be because they perceive a high threat level of the game, and feel like they need every last advantage they can squeeze out of the succesful roll.  In which case, if you toned down the intensity a little bit, they might spend less time down there.

Or maybe it's just that they wanted to try out a cool new system, and, with a bit of experience, they'll be more, uh, cautious managers of their pools.


1 Nice phrase.  :smile:

Mike Holmes

I assume that you've read the flipping the pool thread where I define Anti-Pool? The "thrashing" phenomenon is exactly why I came up with Anti-Pool. I haven't played either version, but I predicted the thrashing problem statistically. Essentially the problem is that the way to get the best chance to succeed is to spend all your dice every time. Over time you'll have more successes than if you ration them out. The problem is when the failure does occur then the character is left diceless and "Deprotagonized". Getting back up to a reasonable level will be difficlt unless the player doesn't want an MoV for a long while.

Anti-Pool solves that. Any chance you might try a scenario with Anti-Pool instead? Just to see how it differs?

Member of Indie Netgaming
-Get your indie game fix online.


It's actually an interesting thing to study The Pool.

Let's suppose that you have three rolls in a row, and a nine die pool, plus, for each roll, you'll have two dice base (ie, either gifts from the GM or from your traits).  What's the optimal strategy, the most likely to get you the most success?

That's non-trivial to figure out.  There are a lot of possible outcomes.  (It's even more complex if you add a utility to having dice in your pool after the final roll).

One possible approach is to bet all 9 dice with every roll.  So, that gives you an 86.5% chance of the first success.  Then, on your second roll, you have an 86.5% chance of being able to get another 86.5% chance of success, and a 13.5% chance of having only a 30.5% chance of success.  Then, on the third roll, you've got, uh, lesse, a 74.8% chance of being able to get an 86.5% chance of success, and a 26% chance of being able to get a 30.5% chance of success.

Would it make sense, then, to say that, following that strategy, your expected successes are:

.865 + .748 + .647 + .041 + .013 = 2.314 successes, on average?

If so (and check my math, because I'm one of those people who likes to do Monte Carlo simulations whenever possible), then that's the number to beat.  If you can, then that at least somewhat implies that the "thrashing at the bottom" problem is a problem of players not utilizing the system, while if you can't, it implies that it's a system problem.


Paul Czege

Hey Mike,

I assume that you've read the flipping the pool thread where I define Anti-Pool?....Any chance you might try a scenario with Anti-Pool instead?

There's definitely a chance. I thought Anti-Pool was a very interesting concept, and I'd like to see how it played out. Right now I'm not sure what I'm going to run next. I'd like to run The World, the Flesh, and the Devil, but a compelling situation and setting has yet to present itself to me. I thought Alex Knapik's $2,059 in Dead Presidents on Gaming Outpost was an interesting setting, but I'd never use GURPS for it, and The World, the Flesh, and the Devil isn't really suitable. Anti-Pool is definitely a contender for that. But after Scott's Sorcerer game, and all the Sturm und Drang in my current scenario and in world events, I'm really ready for something way lighter. Right now I'm thinking either my own Servitude for Wuthering Heights or Elfs.

Still, regarding The Pool, I think it's early to say it's broken. I've only run two sessions. It's possible as players gain expertise with pool management and the techniques for getting things done by way of NPC's that I wrote about in my first post, the thrash-phenomenon will go away. It's also possible that I'm an overly brutal GM and other GM's won't see the same pool management dynamic at all. And beyond that, James V. has emailed me ideas he's exploring for tweaking the rate that dice flow into a player's pool, and they're pretty inspired.


[ This Message was edited by: Paul Czege on 2001-09-25 21:36 ]
My Life with Master knows codependence.
And if you're doing anything with your Acts of Evil ashcan license, of course I'm curious and would love to hear about your plans

Mike Holmes

Uh, Mike, if I can find the spreadsheet with all the possible combinations would you like to see it (I can probably work it up real quick even if I can't find the old one)? I ran it for all the base dice combinations. I garuntee you that given a starting pool of sufficient size that it makes more sense to use all the dice every time. Sure you will run out more quickly, but you'll have way more successes using the all the dice all the time method in total. The second best method is to use just one. That means you'll take longer to run out, but you'll fail at most of the attempts. So your expected results look something like either:

Win, Win, Win, Win, Win, Win, Win, Win, Lose


Win, Lose, Lose, Win,  Lose, Lose, Win, Lose, Lose, Win, Lose, Lose, Win, Lose

And the middle is even worse, usually. Which would you prefer?

And either way, you eventually end up at the bottom of the Pool thrashing about. Other than the ability for the GM to really hose the players at that point, and it being strenuously challenging, I haven't heard anything good about that part of the game.

This does not mean that The Pool is broken, however. Actual significance of events make some worth more than others. Therefore you can play around with these numbers some. I think it actually just makes for a cerain type of cycle.

Anti-Pool just makes for a different cycle, a constant flow back and forth. You win, then you lose. Any level of bet makes sense, because you have incentive to go each way. Go high and burn your protagonism to be sure of something, or go low risking failure but get dice back if you do. I like the control over protagonism that would give the player.

BTW, in the spirit of full disclosure, and for those that haven't worked it out like Mr. Sullivan, here's the short form,

Total--Chance of

Depending on how many dice you are actually risking you can calculate the number of times that you can roll before you can expect to run out. Multiply that total by the success rate above and you get the number of successes in that chain.


[ This Message was edited by: Mike Holmes on 2001-09-25 23:27 ]
Member of Indie Netgaming
-Get your indie game fix online.


Yeah, I think I would like to see it, 'cause the back-of-the-envelope math I just did indicates that it's better not to use your full pool in a three-run set.

Though I haven't been taking into account the possibility of adding dice to your pool through successes.


Hah-hah.  No, you don't need to send the spreadsheet.  I've convinced myself that you're right -- at least without thinking about getting new dice for the pool, the best tactic is always to use all of your dice.

Mike Holmes

I was thinking about it some more, and I have an even more compelling argument for those who might still be doubters.

My original analysis is based on the idea that you'd plan out all of your expenditures before hand. But since you can reconsider in between rolls, you have to consider that in what kind of decision will be made. Once you have made a successful roll (assuming that you're using the "roll em all" tactic) then the odds reset. You have the same incentive that you had originally 89% to succeed and lose nothing. The odds of failure do not increase. So unless you are deciding all of your rolls at the beginning of the game, you have a constant strong incentive to use all your dice. This is of course more true in cases where you really want to succeed and have that additional incentive to roll everything. Or in other words rolling just a few would only occur for very unimportant things and with nothing important on the horizon. Not very often the case in a RPG, especially one that wants you to only roll in important circumstances.

Member of Indie Netgaming
-Get your indie game fix online.

Mike Holmes

Here's a beter way of looking at it. However Paul's players played in the actual example, they ran out of dice in the first session and had none for the second. This is inevitable without giving up more MoV's than you take and still likely then. And that is the real problem. Once they are thrashing they have a very strong incentive to stay thrashing which is that they want their MoV's. And the number of successes and MoVs generated by this method are about the same as if you decided to just bet one die and then forgo the next MoV all the time as a method of maintaining your pool (though that system does allow you a big kick when you need it. But then you're back to thrashing after a while).

As I said before, this isn't an indiation that the system is broken. It just means that you are going to have this cycle where protagonism is determined by the available methods. If the cycle of starting with a bunch of successes and then having a long bad time in which the GM gets to make up a lot of the resolution appeals to you, then The Pool it is. I just prefer a different cycle that bounces back more quickly.

Member of Indie Netgaming
-Get your indie game fix online.

James V. West


Here are two ideas I have for tweaking the rules:

1) Everytime you roll a one, you get to add a die to your pool. If you can take an MoV, it costs you one die. If rolled 2 ones, then you can still get a die for your pool and do the MoV.

2) At the start of each new session, each player starts with 15 dice and subtracts 2d6 from that number. This gives you a range of 3-13 startind dice for each session, an average of around 8 or so.

What do you guys think of those changes? I don't want to get drastic. Having a large pool renders the dice less valuable. Having no dice renders the game pointless. I'm going to try it this way in my next session on Monday. As I told Paul, I haven't ran into this problem in my group yet.

Mike, didn't you email these probabilities back when I first posted the game? Interesting. I hate to quote Star Wars, but I'm goint to do it anyway: never tell me the odds.

The game is about fast, loose play. Gambling, losing, winning, these are all part of it. Getting to control things in moments of tension. All of it. I want to tweak the rules to the point that the "thrashing" problem is much, much less evident. Actually, it needs to be eliminated. Getting down to no dice is something that ought to happen from time to time, but not every damn game!

One thing I certainly don't want to see is a rigid strategy for die rolling. Screw that. That's not how I would want to have to play any game. And if the current rules support that kind of thing, then I need to change them.

I'm starting with the simple changes I detailed above. I believe they can be smoothly integrated with the way the game is played since they are not seperate forms of mechanics, but merely a simple use of what's already there. These two rules ought to result in more dice in players' pools, but not *too* many more.

James V. West

P.S. What's "Anti-Pool"??? Is there a thread on it here? I must be blind...



The Anti-Pool is, indeed, a threat hanging around somewhere.  In brief, it is Mike Holmes' suggestion to reverse how the Pool deals with things -- instead of losing dice on a failure and gaining dice on success, he proposes that you lose the gambled dice on a success and gain a die on a failure.

If you wish to not support a single style of die rolling, I suggest that you do not implement the "get as many dice as you have 1's" rule.  That strongly supports the strategy of rolling all of your (large) pool as you possibly can, every time, because, for the first time in The Pool, it's not just success, but degree of success which matters, and only with large pools will you ever get much in the way of multiple ones.

[ Editted because I made the world's most obnoxious typo -- substituting "loose" for "lose. ]

[ This Message was edited by: Epoch on 2001-09-26 19:09 ]

James V. West

Excellent point. Thank you for hitting me upside the head with it. One of the most important things about the idea for this game was that it was a simple "if you see a one, you win" kind of thing. No counting. Good call.

That leaves me with the die replenishing method, which I think still holds true to the feel of the game.

So, in the anti-pool idea you get dice for failing and loose them for succeeding? Let me think about that for a minute.

In that case, if you win a roll you lose how many? One die? Then you have to buy an MoV, so that's 2 dice gone for every sucess. For every failure you get one die? I can see it working, but I'd have to think on it more. The one thing about it that I don't like (if I'm understanding it, which might not be the case :wink:) is that the gambling is less interesting. If I gamble dice and win, do I loose them? If I don't gamble, or gamble low, and lose, I get more dice. Then, if I go for broke and gamble them all, where's the idea of risk? Am I getting anti-pool right, or am I misinterpreting?

I think the problem boils down to a simple matter of odds, as Mike pointed out. Odds are you'll fail when you roll few dice. If you fail, you loose more dice. So to fix that problem the odds would have to be tweaked up a bit. How?

Having larger pools? That means gambling will be more common. Ok, no big deal. BUt would the same problems then creep up because folks are throwing more dice in the roll, or would the problem lessen because they are winning those rolls more often? I'm getting a headache.

James V

Matt Gwinn

Has anyone thought about simply making traits cheeper during character creation?  As it stands, if I min-maxed my character to have only 3 traits with dice I would have at best 2d a piece and a Pool between 1 and 4.  The way I see it, my character wouldn't be much more skilled at those traits than he is at something he is trying for the first time with no dice.  That means that my chances of success and failure rely almost entirely on fate and my keen gambling sense.  I think being able to start the game with a 3 or 4 in at least a few traits would make a significant difference - especially if you bottomed out.

Kayfabe: The Inside Wrestling Game
On sale now at


Mike's original idea for the anti-pool was that you lose all the dice you used (from your pool, of course, not your traits) if you win the roll.

Example:  You have a trait of 2 dice and 6 dice in your pool.  You elect to use 3 dice from your pool.

If you win the roll, you lose those three dice, and now have only 3 dice left in your pool.  You make a MoV, or you may choose to take a die (so you end up with four dice in your pool).

If you lose the roll, you get an additional die in your pool, bringing you to 7 dice in the pool.

It's much less of a "gambling" concept and more of a "consolation prize" concept.  There's always a good and a bad side to any roll.  Either you win the roll and lose some dice or you lose the roll and win some dice.