Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics

Dice substitute for Jenga towers

A topic by Speak the Sky created May 16, 2020 Views: 12,254 Replies: 40
Viewing posts 1 to 19

I've made a tweet thread about this, but I thought I'd make a thread in the jam too.

Obviously a Jenga tower is a bit more specialised than dice or cards, and can't really been done easily and properly online, so here's a simple mechanic that uses d6s (and an online dice roller like

  1. 1st pull: roll 100(!)d6, take out the 1s,
  2. 2nd pull: roll the dice you have left, take out the 1s etc.
  3. 'lose' when you run out of dice.

100d6 is a lot, which is why you should do this digitally! Here's what one random run of this looks like:

At the start you're pretty safe, but you lose dice fast. In the middle the tension is building. In the back end you have a final stretch with few dice before you lose.

And here's how things look on average:

Most games take at least 20 rolls/pulls, and most are done by 40.


Here's the same thing if you roll a d20 every 'pull' and end the game if you roll a 1:

As I say in the tweet thread, you could wait until the second 1 to end the game, but it's not much better. Waiting until the third 1 just takes too long. You could also do nothing for the first 20 pulls and start rolling the d20 on the 21st pull, but then there'd be no mounting tension, no dwindling hope, no visible tracker for how bad things are going. The 100d6 mechanic has a story baked in. That said, I'm nobody's boss, people can go with whatever mechanic they like!


This is a very interesting mechanic, though as you say - 100d6 is a hell of a lot to manage!

I've been given another suggestion via Chris Sims on Twitter that I wanted to share, along with one of my own. Chris' is:

Start with 9 tokens. When instructed to pull from the tower, instead roll the die. If the die shows 6, you lose one token. If your number of tokens falls to between 5 and 2, you lose a token if the die shows the number of tokens you have left or a higher number. When you have one token left, you still need to roll a 2 or higher to lose a token. So, for instance, if you have one or two tokens left, a roll of 2 or higher causes you to lose one token. When you lose your last token, the “tower” falls.

That led me to what may be a simpler mechanic. Start with a counter that you set to 30 (the average number of pulls in a Jenga game - but you could set this to anything you like). When you're instructed to pull from the tower, reduce the counter by 1 and roll a d6. If you roll equal to or higher than the counter number, the tower falls.

Realistically you don't start rolling until the counter hits 6, obviously.

Keep the ideas coming!


Yeah, that's why I say to use a dice-rolling website or discord dicebot—the number of dice doesn't really matter then.

Here's how Chris Sims' idea works out in practice:

It gives a good curve of average results, with a difference in the story it tells in each game:

Where the 100d6 mechanic usually has a drawn-out tense ending, the token mechanic sharply curves to the end with a sudden disaster (because the probability of rolling over the number of tokens increases as tokens decrease). It can also randomly plateau earlier, like a moment of stillness, calm before the storm. It might be more swingy, but I'm not sure.

Your method is obviously gonna be a lot more consistent—you get 24-30 pulls. The action is always gonna happen over at most 6 turns. People can set exactly how many pulls they get before the risk kicks in (by changing token count) and set exactly how many pulls they get once that happens (by changing die size).

I think I still prefer mine, since it's both simple and dynamic, but the others have their advantages if people want a different story progression, more control, or for the mechanic to get out of the way until the final stretch.


I like how over the top 100d6 feels, along with how it diminishes. It's a crazy cool idea. My only issue is the minor one that to do the 100d6 pretty much requires an electronic tool. Part of the appeal of the game is its analog and somewhat meditative nature. Again, a minor issue, though.

See also...


Don't get me wrong, the tower has its own specific feel rooted in the physical existence and order of the object and the tactile skill involved, and the feeling that you fucked up when it falls—these are meant as substitutes when a tower is impossible or impractical. I feel like the 100-die mechanic probably works better in sci-fi, or any setting where numbers are more significant, since you can tie it into a % value (% oxygen remaining, % of humanity left after alien attack, % chance of survival derived by AI etc.).

That digital tower is cool, but unfortunately suffers from the same kind of slidy, bouncy instability I've seen in a lot of 3D stuff. Here's one attempt I made, screenshotted after 2 pulls:


Yeah, that digital tower is fun but totally wonky, especially for this usage. I wouldn't use it for this game.


I'm offering online alternatives for the die, the cards, and the tower in my submission, and this was the tower alternative I went for. It's not perfect at all - the physics make the tower likely to fall very quickly, compared to a physical one - but I figure any (free) digital option is better than none! 


Ideas I have had:

1.  Build Lego, duplo, coins, dice, pebbles, sweets etc. tower somewhere that's in the way of your everyday life. Or somewhere the kids walk past everyday. It's bound to get knocked over but when?

2. I considered this one for 'Adrift'. In Adrift you are thirsty and dehydrated. Start the game with a glass of water.  Take a single small sip. Everytime you are asked to take from the tower put a coin in the glass. When it overflows your raft has sunk.

3. Bake 20 cookies/cupcakes or whatever but one of them has chilli in it. When you eat the chilli cupcake it's game over.

4. Bake a cake and put it in the oven.  Stop playing when your nerve breaks  and you go check it.

5. Complete your daily log while balancing a book on your head. Each time you add a block add a book to your head instead.

6. Put a coin on your elbow then catch it in the same hand. When told to add a block add a coin. Drop the coins the game is over.

7. If you are in the pub (after lockdown) flip beer mats. Add a beer mat when told to add a block.

8. Do your daily log next to something sleeping. If they wake up it's game over.

9. Do your daily log next a randomly set timer/alarm. If it goes off during your daily log the game is over.

10. Replace the tower with buckeroo.

11. Replace the tower with kaplunk.

12. Each time you are asked to pull from the tower place a Domino standing up. Build an elaborate trail around the house. When it gets knocked over (and it will!) The game is over. 

13. Plant a seed.  Each daily log you water the plant. When you first see it's shoot emerge it is game over. Good for a Martian/terraforming game.


14. Replace the tower with operation. Would be great for a surgeon hack of the Wretched. Complete the daily log while operating. Get a friend with ketchup just off screen if all goes wrong.


I really like the 100d6 option. Honestly most people can get an app or load a site for this, so the number of dice shouldn't be an issue in most cases. Especially if you link to a good one in the document. Probably going to adopt a similar method, maybe d20's and pay around with those. Thanks for the tip Cat Evans. 

I'm planning on trying something with a deck of cards, 12 cards, then a joker, then 16 cards with the other joker shuffled in. When the second joker is pulled the tower falls. I'm not 100% about the distribution, but it's workable. 


I have been thinking of a way to approximate this S-shaped curve without needing tokens or software. I think rolling with advantage (roll two dice, keep highest) is a decent substitute. Closest I got is using a d100 with advantage:

  1. Each turn mark, how many turns have passed (this is your Counter).
  2. Roll a d100 with advantage (keep highest number).
  3. If max(2d100)<Counter -> lose;  else -> keep playing.

Did some quick calculations in Excel. X-Axis is Number of Turns. Y-Axis is probability that all rolls until that turn are successes.

This is pretty good I think. Thanks!


my favorite from this thread is chris's start at 30 and countdown.

"Start with a counter that you set to 30 (the average number of pulls in a Jenga game - but you could set this to anything you like). When you're instructed to pull from the tower, reduce the counter by 1 and roll a d6. If you roll equal to or higher than the counter number, the tower falls."

but my husband suggested rolling a d12 instead, for a chance of the tower falling after 18 pulls rather than 24.

i have the following paragraph in my game, which others can feel free to use also!

"If you don't have a jenga tower, you can play without one or use a dice mechanic to simulate the degrading nature of a jenga tower. Set aside a number of dice with their displayed values equaling 30, such as three 10-sided dice or two d20s with one die turned to 20 and one turned to 10. You will not need to roll these, but will use them to count down from 30 to 1. Each time you would pull from the jenga tower, reduce the number by 1. When you reach 12, each time this number is reduced, roll a 12-sided die (or two d6s). If the number rolled is equal to or higher than the number on your count-down dice, the tower falls."

Deleted 3 years ago

I don't have the maths to work this out, but what does the curve look like if you draw a random domino from a full set of 28 and continue until you get a double, removing dominos after they're drawn?


This is the curve if you stop once you've pulled every double—if you only need to draw one then it's like rolling a 1 on a d4 (since there are 7 doubles in the total 28), which ends very quickly. Pulling all works pretty well, with some specific features:

  • it always ends after turn 28 at most
  • 28 = four 7-day weeks (plus a possible 28 Days/Weeks Later reference in particular if that's what you're going for?)

[no comment]

For real though thanks for running the maths on that - I wouldn't know where to start!

Submitted (2 edits) (+3)

I made a very basic html version of the 100d6 mechanic by Speak the Sky that handles everything for you. When I say basic I mean reaaaaly basic, I'll work on making in prettier when I can be bothered to do CSS. I hope you are okay with this Speak the Sky. Otherwise let me know and I'll take it down.

You can also download this file and include it in your project if you like:


Completely fine with me!


Great work!


This is wonderful!


So I posted on Twitter without thinking to come check here and see if someone had tried working up an alternative. Lots of neat ideas! 

My thought was "What if you rolled 5d6 (maybe 6d6?), and have to roll higher than the total number of 'blocks' you've 'pulled.'" If you don't, it tumbles.


there are 54 blocks in Jenga so that would be 9 x d6. 


Something to keep in mind with when you’re rolling multiple dice is that the probability isn’t uniform. On 5d6, you are much more likely to roll somewhere around the 17-19 range than in the 27-29 range, for example.

Submitted (1 edit) (+3)

My current thinking about the Jenga tower is that what it represents for the game, rather than the strict mathematics of it, is the more important part. So, what should it represent?

  • A growing sense of dread
  • Mounting pressure
  • A chance for the player to interact more directly with the game.

I like the 100d6 method that Sky posted, but I think it misses that last point a bit - the player doesn’t actually do much. I propose a revised version to address this problem. 

You start with 100 points of supply, named as appropriate for your game. Whenever your supply is tested, scratch off any number of supply, then roll that same amount of d6s. If you can make a total of 16 or higher  on your dice, you’re in the clear.  If you can’t make 16 or higher, you’re dead.

16 was chosen because 600 / 36 rounds down to 16. 600 is the maximum of 100d6, and 36 is the practical maximum of pulls you can make from a standard 18-floor block tower before you’ve cannibalized the tower entirely. This not only lets the player decide between a more risky approach that’s likely to be more effective in the long term (choosing few dice) or a safer option that’s going to be costly later (choosing many dice). It also gives the player something mechanical to do, in that they have to add up the numbers.


I completely agree that the game needs something tangible that represents the situation you are in. For me the wobbles, the tension of probing and prodding the blocks and having to stack them back on top made for a very enjoyable role-playing stimulus. Also having something delicate and precarious to look at was also an incredible source of inspiration and motivation to play. 


It's an interesting option, but I think it runs into some problems by aiming for more complexity. From some quick anydice-ing:

  • 7d6 works best: ~75% of the time you get to roll 14, then run out of dice
  • a mix of 7d6 and 6d6 gives you slightly more range, but your chances get worse enough that it's probably not worth it
  • anything else means you die in a few turns or waste dice (7d6 gives you a 98% chance of success per roll, 6d6=~90%), 5d6=~70%)

You could make the target less harsh (e.g. >=10) to extend the game a bit.

The real problems for me are 1) it has more and clearer responsibility for failure than jenga (where pulls are more drawn-out, skill-based, and obscure). Basically, it feels more arbitrary, but somehow also more my fault. 2) It adds a gamble, but everything is always at stake, so there isn't really any choice. That, and choosing which block to pull affects every future pull, but choosing how many dice to roll only affects how long the game can go on for.

I mean, these both have something in common (any failure = instant death), so what if there were multiple things at stake that you distribute dice between? That'd give more weight to your choices before and after you lose the things, since you could have pointed oracle questions about them even when they're gone. Alternatively, multiple layers of failure (e.g. someone posted an idea in the idea thread about a journalist on the case of a serial killer, who kills again every time the tower falls).

anything else means you die in a few turns or waste dice (7d6 gives you a 98% chance of success per roll, 6d6=~90%), 5d6=~70%)

It has admittedly been a little while since I played Jenga, but I would estimate that a good number of the pulls I've made (especially in the later stages of the game) were well below 70% chance of success. 5 is definitely the lowest I'd go in terms of "reasonable chance", though; it drops off pretty harshly below that. A range of 14-20 reasonably successful pulls sounds about right to me.

One method that I would probably use to mitigate the risk of death would be borrowed from the grand-daddy of Jenga-based horror games, Dread. You give the player the choice of whether or not to pull. If they pull without fail, their story is slightly more positive. If they don't pull, they don't succeed at what they try to do, but at least they know they're not dying.

Basically, it feels more arbitrary, but somehow also more my fault. 

Interesting! Personally, I would say that if I tip over a Jenga tower, it feels a lot more like my fault than if the dice I rolled ended up being low. That's not to dismiss your point of view: just saying that I don't share it.

 what if there were multiple things at stake that you distribute dice between?
Alternatively, multiple layers of failure (e.g. someone posted an idea in the idea thread about a journalist on the case of a serial killer, who kills again every time the tower falls).

I am all for adding more options for dice results! For your first case, I could see something like, rather than having a single goal at 16, you could do two separate goals at 9 and 9, or whichever distribution is more appropriate for the situation. Towerfalls that aren't gameovers are fun as well, and depending on the concept, they can definitely be a good fit.

A range of 14-20 reasonably successful pulls sounds about right to me.

Ah, 14 is the highest you can reasonably get, with pretty much the best strategy. 20 is what you can get if you succeed with every 5d6 roll, but if you roll 5d6 more than once then you're more likely dead than alive. Any mix of 7, 6, and 5d6 is worse than just 7d6—you get the possibility of some more rolls/turns in the future, but you're much less likely to get there.

It's very unforgiving—deviate from the strategy and you die quick or cut your game short by wasting dice. The problem then is either people go with the best strategy for part or all the game (so they have less/no choice) or they gamble and fail quick (so they don't play for long enough to have many choices). Most people aren't gonna figure out a strategy, so they'll go with option 2.

Interesting! Personally, I would say that if I tip over a Jenga tower, it feels a lot more like my fault than if the dice I rolled ended up being low. That's not to dismiss your point of view: just saying that I don't share it.

Yeah, 'more' was the wrong word—but it still has a single clear point of responsibility (making a single bad gamble) where jenga's drawn-out.


Well said on all points. (I would say that Jenga still has a single point of responsibility, but I digress.)  I've adapted this take on the system for my second submission, which is a lot more flexible when it comes to target numbers and failing a roll. Basically, we've got variable targets, multiple targets per prompt (on many, but not all) prompts, and failing to meet the prompt just means a less favorable outcome, rather than certain death. While I'm sure that each individual target number can still be optimized for, if one wants to play the spreadsheet game, the larger variance should mean that most of the problems we've discussed can be avoided.


can't sleep & just had this thought -- what about building a "dice tower" instead of pulling from a jenga tower? i mean who doesn't have 25755 d6s laying around


hmm yeah I like that. Or maybe build a house of cards from the cards you keep drawing. Maybe every time you draw a king you need to start working on a new level of the house of cards.


yessss! that's great!

(1 edit) (+5)

I'm late to the party, but here's a simple, non-Jenga mechanic that I've been playing with:

  1. Write down the numbers 1 through 6 in two lists with a bar in between them (1 2 3 4 5 6 | 1 2 3 4 5 6).
  2. Every time the game tells you to pull from the tower, roll a d6.
  3. If the number that comes up has not been crossed out from the left list, cross it out and continue with the game.
  4. When all of the numbers on the left list have been  crossed out, start crossing out numbers on the right list.
  5. When all of the numbers in both lists have been crossed out, the tower falls/alien breaks into the derelict/barbarians enter the library/etc.

The distribution (is that the right word?) for a 10,000 run trial looks like

with 90% getting to the 19th pull, and 50% getting to the 27th pull.

Submitted (1 edit)

Also late, but FYI here's the one I use in my game: take a tarot deck (78 cards), remove the Tower card and replace it about 1/3 of the way through the deck. Drawing the Tower card is equivalent to the tumbling block tower collapsing.

Alternatively, if you want to remove the concept of a tower altogether (useful if you're adding other mechanisms and want to reduce the number of items on the table), add a joker card about halfway through the deck of playing cards. This one isn't exactly equivalent in terms of probability, but it's close enough.


Ooo, considering I'm replacing the tower with a cat, the Joker is excellent for me. Thank you! 

Very late to the game, but getting three blocks of 12mm d6s from Chessex will give you 106 dice.

Could even do it with one block and rerolls. Should only be a couple times before you only have to reroll once.

(3 edits)

I'm late to the party but did anyone ever try this method? All you need one of each polyhedral dice (d20, d12, d10, d8, d6, d4)

When the game tells you to pull from the tower, roll the d20. If you roll a 1 or 2, replace the d20 with the next biggest die (d12). When you roll 1 or 2 on the d4, the tower fall.

According to my tests, the average number of pull is around 30 and the probabilities are very similar to other methods. Plus you only need 6 common dice.

EDIT: I just realized that this is not a great method since there's no immediate sense of danger as long as the die is not a d4. 

Deleted 88 days ago
(35 edits)

That's funny! We must be sharing a hive mind or something:

That is I am new to this thread, too, and while I was reading it I was thinking of the same idea as you: that is why not simply replace the Jenga blocks with a 'usage die' from the Black Hack, that we keep downgrading by one step (d20, d12, d10, d8, d6, d4...) whenever we roll a 1 or 2.

But the same as you did, I realized that the problem with that method is that it is too predictable and gradual. There is no danger (not even a remote one) that it might all 'fall' in a single pull or two. So scratch that.

Then I reached your post at the bottom of this thread and I saw that you had already beat me to all of this.



Sorry for the long spiral of edits here. It is just because I keep having additional thoughts on this:

1) What if the usage die that we are downgrading here step by step (d20, d12, d10, d8, d6, d4...) is combined with a second die roll - e.g. an exploding d6, or some similar mechanic - that sometimes makes that usage die downgrade by more than one step at a time?

2) To simplify that:  We could just use the usage die method on its own and downgrade the die on a 1 or 2 the same as we usually would. Except if it is a 1, then we downgrade the die by one step; and if it is a 2, then we downgrade the die by 2 steps.

3) Or if that doesn't feel dangerous enough: then in addition to doing the above, we also roll a small pool of d6's on the side whenever we roll a 1 or a 2 on the usage die: If the lowest result any of those d6's beats the 1 or 2 that we just rolled, then we downgrade the usage die by as many steps as that d6 result. 



Besides the problem that Avatar_Izzy found earlier, I have now run into another problem with this idea of substituting a usage die for a Jenga Tower:

On the good side: When combined with any of my additional rules above, it sometimes simulates the tension of pulling blocks out of a Jenga tower fairly well.

But on the bad side: There are also some runs where the usage die becomes stuck at one of its larger die-sizes & it takes an incredibly large number of rolls to get through that step.  There is no tension & it is no fun when it becomes stuck like that.

The problem there is that it all just depends on how the usage die itself rolls: So long as that die is involved, then there is always a possibility that on some runs it will just keep beating the odds; that it will just keep rolling & rolling for an absurdly long time before it finally rolls a 1 or a 2.

That could be fixed by combining the usage die with a progress track or etc to keep its number of rolls within bounds.

But I think that would be too cumbersome: Why combine a usage die with a progress track, when we could simply roll against a progress track's value & leave the usage die out of this?

So I'm abandoning this usage die idea for now.

But I'm not deleting my post here. I'm leaving it, just in case any of it is useful to Avatar_Izzy or anybody else.


One idea.
Write down the numbers 1-60 on a piece of paper. Starting after the first turn, each turn you cross the lowest number off the list.
Each turn you roll 1d100.  If you roll any of the crossed off numbers that's the same as a tower collapsing.

This gives you a 1% chance of disaster after the first turn going up 10% after 10 turns, 20% after 20 turns, 30% after 30 turns, 40% after 40 turns etc.