Predictable thoughts about Improbable Outcomes

Image by StartupStockPhotos from Pixabay
If you want to start a conversation with a tabletop gamer, all you have to do is ask their opinion on GMs fudging die rolls. Everyone has an opinion, a theoretical best-practice policy, and everyone has a preferred approach in the real world – and the two don’t always match. Some people even have different preferences when they are a player compared with what they do when they are GM!
In my Quora feed the other day, I came across an answer by Dave Rickey to the question “How do game designers engineer luck into video games?”. Dave is a video game designer, so he is well-qualified to answer the question. I found his words so interesting that I reached out for permission to reprint them here, which he was generous enough to approve. I’ll be back afterwards to discuss the relevance to tabletop RPGs.
![]() |
How do game designers engineer luck into video games?Answer by Dave Rickey, Game Designer, Reformed Contrarian Seriously, the Random Number Generator (RNG) is only the beginning, we do all kinds of things so you can feel like you’re having ‘good luck’:
|
![]() |
Okay, so let’s talk about these in a Tabletop RPG context. I’ll keep my numbering the same as Dave’s so that you can match my commentary to his “tricks of the trade”.
- one-shot kills can be hugely entertaining for a player when he’s the one inflicting them, and promote screams of outrage when a player is on the receiving end. Hard-capping the amount of damage inflicted can be trickier to achieve in a Tabletop RPG unless you couple it with a one-shot weapon that must then be discarded – which usually gives one or more PCs a free shot while the enemy changes weapons, a factor that should be taken into account when deciding what the cap should be. See also “Altering The Combat Equations”, below.
- 1a. Warning Shot variation: I like to use this approach when I want to boost the tension at the table. Giving the enemy a really nasty weapon that does horrible things to the landscape beside/behind a PC or other obvious threat and that they will never get to use at full effect again is a great way to amp things up – if used sparingly. It gets fairly obvious fairly quickly.
- Artificial Stupidity – I prefer to use “Induced Desperation” or “Touch Of Panic” versions of this idea, because “artificial stupidity” tends to get noticed by players and can cheapen and undercut the desired sense of victory and achievement. “Induced Desperation” means that one of the PCs gets in a “lucky shot” on a miss that affects something the NPC is worried about, or that they are up against some sort of time limit that the PCs don’t know about; it causes NPCs to make mistakes, fire wildly, or choose to attack another PC without administering a coup-de-grace. You need to really sell this through narrative and roleplay or it becomes obvious.
“Touch Of Panic” works the same way, but implies that the NPC is worried about something that the PCs don’t know about (because the NPC appears to be winning) – something that transforms a potential victory into a potential loss. It’s all well and good inflicting 10 points a round (or whatever) but nowhere near as good when you need to do 20 a round in order to finish before reinforcements arrive, or your power pack runs out of juice, or whatever. The core of the concept is that however successful the NPC appears to be, they are also battling some sort of clock – and while they may be winning the first fight, they are losing the second, which induces Panic and wild attacks that often go astray. Once again, this needs to be sold properly through narrative and roleplay to be believable. Another variation that can work is the “Premature Coup-de-Grace” or “Premature Gloat”, both of which permit the almost-defeated to get a free kick in – which can sometimes be enough to turn the tide.
For example, I once hit the PCs with a Big Bad who they knew was almost unstoppable because of his affinity with the legendary weapon he wielded, and indeed he whittled through their ranks like wheat from chaff, not stopping to administer killing blows because he had to move on to the next enemy. Finally, the last PC was almost at his mercy when he made the mistake of counting his chickens before they hatched – he went for a Premature Coup-de-Grace, which gave the first PC he had “defeated” the chance to knock the weapon out of his grasp, which in turn gave the almost-defeated character a free shot at an obviously-vulnerable spot. Weakened massively, he fought gamely, but one by one the other PCs recovered and rejoined the fray, and eventually the bad guy went down. Knocking the stuffing out of the PCs with an obviously superior and intimidating enemy raised the stakes enormously, and the turning point inverted the odds, calculated on the assumption of the results from the initial battle.
- The Stormtrooper Marksmanship Academy gets a little obvious when you don’t have a lot of action to distract the players. But there are ways… See “Altering The Combat Equations” below.
- 3a. Conservation of Ninjitsu variant Good one, if you can avoid it becoming obvious. Video game designers have the advantage of being able to test their fights repeatedly to see whether or not they are too tough; Tabletop RPG GMs don’t have that ability, and have to go with their best guesses based on prior combat performance. See also “Altering The Combat Equations” below.
- A Hidden Health Buffer is a clever solution for a video game but not one that will work very well in a Tabletop setting where a player is responsible for tracking the Health of his character. If you can persuade your players to let you track all the damage without announcing how much is being inflicted, just telling you what their current total is at the start of combat, it can work – but good luck getting them to go along with that.
- Extra Health Buffs: These are more frequently handed out just before they are likely to be needed in a tabletop RPG, a practice that can cause problems if they aren’t needed after all and start to accumulate. Avoiding this problem requires the GM to keep careful track of how many the PCs already have, because asking how many healing potions characters have just before they find more makes your intent blindingly obvious. I overcome this difficulty with a prior combat encounter which requires a character to use one or more healing potions, giving the GM to ask how many the character has left, and get the same info from the other players. This enables the “cache” to be discovered to be adjusted in size.
That reminds me of a fun moment from long ago: Healing potions in this campaign were usually kept in a bottle of particular shape, so that you could find one in your pack by touch without taking it off (an idea I got from the unique shape of the traditional glass coke bottle). On this occasion, the PCs discovered a treasure cache which included one of these bottles – but when they opened it in the middle of the next battle, all that was inside was a note from a well-known brewer of such potions which read, “I.O.U. 1 Healing Potion – Dabny” (not the name of the Brewer)…
Extra Ammo Refills: TT RPGs are notoriously bad at tracking ammo, for all sorts of excellent reasons. It’s not that there aren’t rules for doing so, it’s just that they are so damn tedious. So this can only be used in selected situations – such as Cinematic Combat – when the entire “Running out of ammo”/”Discovery of ammo cache”/”Reload at the last possible second” sequence can be delivered in narrative form. Sometimes, you can use a “dud” to reduce the amount of ammo a player thinks he has in order to induce this whole sequence. This works by eliminating the paperwork overhead and preserving only the intensity of the situation.
- Fiddling The Odds: see “Altering The Combat Equations” below.
- Forced Success after X sequential failures: this can be a lot more difficult to pull off in a Tabletop RPG because the mechanics of arriving at a chance of success are not hidden beneath the surface, as they are in a video game. It’s a lot easier to do the inverse: Forcing failure on an NPC after X sequential successes. But that brings me to…
Altering The Combat Equations
Any combat equation has four parts: Roll, Modifier, Probability of Success, and Effect. You normally have the same set of equations for the enemy, as well, though the specifics will vary. Of these, for the player’s attacks, the roll is known to the player, the modifier may or may not be known, the probability of success is known if the modifier is known, and the first-order direct effect is known – any interpretation or secondary effect is usually the GM’s province – but needs to be credible, according to the primary effect. That makes that equation difficult to manipulate.
Which leaves the enemy’s attacks. Of these, the player may or may not know the roll, won’t know the modifier (but will probably notice severe inconsistencies in it), usually won’t know the probability of success, but can estimate both it and the modifier from repeated attacks, and needs to be told the effects, both direct damage and secondary effects, though they will know the rules as they pertain to their character – making secondary effects a minefield to be manipulated very cautiously.
So, if you want the capability to manipulate the flow of combat in your game – even if you don’t use it, or don’t intend to – you need to take some active steps. First, you have to keep something hidden from the players. In terms of player attacks, that can only be the modifier to their combat rolls and the probability of success; you will have to do all the mental arithmetic required and announce simply “hit” or “miss” before moving on to effects. You can’t make the die roll for the player, and you can’t hide the modifier if you announce what the character needs to hit, because it’s simple arithmetic to get one from the other.
Now, the only way that you can actually hide the combat modifier is by adding more ingredients to the mix than just the combat bonuses that the character gets automatically. There are two ways of going about that – keeping all the modifiers together, or dividing them into two separate categories. The first choice requires the player to announce his combat bonuses at the same time as he announces his roll; the second permits him to apply his combat bonuses to the die roll and simply announce the total. But it also means that the GM needs to explicitly think of the “additional modifier justification” as something outside the control of the player – and it has to be there with every roll. For example, if you justify it under the heading of “environmental effects”, you need to be able to articulate those effects and their environmental causes every time.
However you approach it, the simple fact is that this means more work for the GM. Which is just one of many possible reasons that a GM might choose not to do it.
Which leaves only manipulating the enemy’s rolls. If you do this without also doing the same thing for the PCs, you must never let the players suspect that you are being anything other than scrupulously fair; because you are leaving yourself wide open for accusations of GM Bias against the PCs, and an extremely unhealthy us-vs-him mentality can arise.
Nevertheless it can easily be done – but it once again means hiding something from the players. Arguably, and ideally, it requires hiding everything from the players except the outcome – “Hit” or “Miss” – and the effects (“…doing 25 damage and stunning you for d6 rounds, I’ll let you roll…”). To explain why, let me tell you a little story…
The Discovery Of Manipulations
The more intelligent and analytical the player, the harder it is to conceal any manipulation of the combat equation from them. I once played with a lawyer (not a rules lawyer, I hasten to add) who could, after three or four attacks by an NPC, predict roughly what the modifiers of that enemy were, what the chances of success were, and could therefore predict what I had rolled – then alter his tactics accordingly. This was back in the days when I hid my die rolls behind a GM screen. It made him a very challenging player to GM for.
If the GM makes his die rolls in the open, it’s even easier for predictions – and becomes even more obvious when the GM is fiddling the odds. The key to doing so, I’ve found, is to create a dynamic environment in which some change in circumstances leads the player to expect that one or more of these known factors will change – and that the change will be unstable. That could be an environment filled with smoke or fumes, or a walkway that can collapse at an importunate time, or any number of such things.
You could even regard these as small-scale dues-ex-machina, but that’s generally overstating them – if that label can be applied, whatever you have in mind is probably too extreme. For example, a creek that just happens to flood at the right moment to knock an enemy off his feet momentarily is going too far.
There’s nothing wrong with occasionally going to far, either. But you need a curtain behind which to hide as you pull these levers – a secretive die roll after each round of combat, for example, supposedly to see when the flood arrives, plus some justification for the flood like obvious high-water marks. You can also get away with a lot more if the cure seems to be almost as threatening as the enemy supposedly was – so also sweeping PCs off their feet, demanding saves against drowning, etc, makes this seem less like the GM intervening and more like the GM adding a maraschino cherry of a challenge on top.
The Philosophical Debate
There’s also the deeper philosophical question of altering the odds at all. Some groups would rather the certainty of letting the chips fall where they may, others might prefer the GM to provide a safety net. I have a long-established track record of not killing PCs unless they do something stupid. I don’t generally include things like a savage playing with the controls of a nuclear reactor – the savage has no context in which to determine that this is a stupid thing to do. A character from the modern day – no matter how illiterate – has no such excuse. But he or she also has the capacity to make more rational and purposeful adjustments to those controls – which is not the same thing, and provides no justification for a “stupidity kill”.
Ultimately, there’s a lot of room under this heading for different answers, all of them right from some perspective and all-but-one of them wrong from other perspectives. Answers could range from doing everything in the open, to hiding as much as possible but never manipulating the results (and let the dice fall where they may), to only manipulating results under defined circumstances, to only manipulating the results for one faction, to a more egalitarian approach to manipulating the results.
One of the other comments to Dave’s answer speaks to exactly this point – Hahn Ackles wrote.
There’s a huge ideological divide here, as you no doubt well know.
I’m a very narrativist GM myself, much like you describe, but over on the Pathfinder forums I have debated with some people are way far on the other side.
I’ve spoken to people who say that fudging dice is morally wrong, people who roll everything out in the open and even freely tell their players target numbers, etc. It’s very fascinating what different people call fun and what hills people choose to die on.
The reasons that game designers play with the odds is to ensure as entertaining an experience as possible. Tabletop games have different tools available, but the same broad objective. It is useful to consider the techniques that they employ to see what can be transferred into our form of gaming.
And again, huge thanks to Dave Rickey for giving me permission to quote his answer to the Quora question, and to Lawson Shepherd for asking it in the first place. You can read the other responses to the question at this link.
Discover more from Campaign Mastery
Subscribe to get the latest posts sent to your email.
October 10th, 2019 at 12:42 pm
Now I know how to deal with table top players.
October 21st, 2019 at 3:21 pm
Rule Zero: Always have fun. Every other rule can be bent or outright broken, if the result is more fun.
October 21st, 2019 at 9:34 pm
That’s always been my philosophy, Dave :)