When used in ways beyond the standard outlined uses, wish carries a risk of stress:

Finally, there is a 33 percent chance that you are unable to cast wish ever again if you suffer this stress.

Assume here that a DM would resolve this using a d100 roll.

Are there any features/abilities in the game that can work to reduce the chance of this happening (either by modifying the roll or by some other method) when using wish in this way?

Make a Simulacrum, have them cast Wish instead

The material costs for a Simulacrum are only an arbitrary quantity of snow and hair/fingernail clippings, along with 1,500gp of Ruby dust. Wish itself is capable of generating an object worth at least 25,000gp, meaning it can generate 25,000gp worth of Ruby Dust (or a 25,000gp Ruby that can then be smashed into dust).

Meanwhile, Simulacra are required to obey their creators, both in actions and intent:

The simulacrum is friendly to you and creatures you designate. It obeys your spoken commands, moving and acting in accordance with your wishes and acting on your turn in combat. The simulacrum lacks the ability to learn or become more powerful, so it never increases its level or other abilities, nor can it regain expended spell slots.

So you sequence things like this:

  1. Acquire 1,500gp worth of Ruby Dust legitimately
  2. Create your Simulacrum (casting time: 12 hours)
  3. Have your Simulacrum wish for a 25,000gp ruby, smash the Ruby into dust (or wish for 25,000gp worth of Ruby Dust, if that doesn't violate the "one object" rule on Wish)
  4. The next day, (perhaps with your Simulacrum's help?) assemble the next Simulacrum, and create the new one (casting time: 12 hours)
  5. Have this Simulacrum wish for whatever it is that you actually wanted to Wish for
  6. Repeat 4-5 until you run low on Ruby Dust, then return to step 2

Using this process, only your Simulacra will risk losing the ability to cast Wish, while you, the original caster in full control of these Simulacra, will never need to risk your own use of the spell.

Disclaimer: DMs reserve the right to make Simulacra used in this manner revolt and turn against their creators. This may happen even though the explicit rules written by WotC expressly forbids them from doing this. I claim no responsibility for any "Fighting my Evil Clone!" Campaigns that get sprung forth by this obvious abuse of game mechanics.

Adventurer's League does not permit this.

The FAQ for Adventurer's League play is rather explicit about this exact combo:

You Are You; and So Is He. If a simulacrum you have created casts wish, both you and your simulacrum suffer the stress associated with casting the spell—including the risk of being forever unable to cast wish again. The inability to cast wish extends to any simulacrum you create in the future.

So this combo is not permitted in AL play.

  • Comments are not for extended discussion; this conversation has been moved to chat. – mxyzplk Nov 10 at 16:21

Wish itself can, if there are two casters in a party with access to wish

You undo a single recent event by forcing a reroll of any roll made within the last round (including your last turn). Reality reshapes itself to accommodate the new result. For example, a wish spell could undo an opponent’s successful save, a foe’s critical hit, or a friend’s failed save. You can force the reroll to be made with advantage or disadvantage, and you can choose whether to use the reroll or the original roll.

Obviously this means that any items which cast wish for you (utilised on your behalf by another creature), or other creatures, like Genies, or a simulacra, could also do it.

Using this strategy the odds of various scenarios are:

  1. The odds of the first spellcaster not being able to cast wish again is 33%*33% = 10.89% (they have to fail both times in the scenario I'm laying out).
  2. The odds of the second character not being able to cast it again has odds 33%*33% = 10.89% (the first caster has to fail in order for the second to even have cast wish and risk the stress).
  3. The odds that at least one of the two casters suffer the stress is 33% * (1-(67%)^2) = 18.19%.
  4. The odds that both casters suffer the stress is 33% * (33% * 33%) = 3.59%.

(thanks to frodo skywalker for pointing out an incorrect assumption with my maths for the previous version of the last two points)

So if you go into a situation assuming this strategy you can use math to reduce your chances.

Using more people you can reduce the odds of any individual spellcaster losing their ability to cast wish to an arbitrarily low level.

If you use the infinite staircase of simulacra the chance of the stress being suffered by the original spellcaster can be reduced to effectively zero given enough preparation (assuming that a stress suffered by a simulacra does not affect the original caster).

As noted within the comments, it is possible that a DM may choose to interpret the second wish as having caused the first never to have been cast. This situation is also desireable as it gives the 1st caster their 9th level slot back and the opportunity to cast wish again, either to:

  • roll back a stress suffered by the second caster (if the second caster suffers the stress)
  • attempt to cast the same wish again
  • Comments are not for extended discussion; this conversation has been moved to chat. – mxyzplk Nov 10 at 16:20

Step 1: Get an item that can cast wish.

Step 2: Trap yourself in a zone of silence. Arrange it so you can get out without spending an action, but only with a good roll; say a 20. Failing should not do damage, just return you back into the trap.

Step 3: Create a Simulacrum or other "sacrificial" being (like another party member) to use the wish item.

Step 4: Repeatedly try to get out of the silence trap. When you do, immediately cast wish using a spell slot.

Step 5: If you suffer blowback, the Simulacrum wishes a reroll of your escape of the trap, with disadvantage.

If you fail to get out of the trap, you could not have cast wish, so you could not have suffered blowback.

If you require a natural 20 to get out of the trap, then the wish'd reroll has a 1/400 chance of getting out of the trap.

While in the trap, you cannot cast wish, as you are silenced.

So, you are basically immune to blowback.

Using multiple wish-casting items reduces the chance by a factor of 400 for each additional one.

This makes it take an average of 20 rounds to cast Wish, but a 1 in 1200 chance of blowback sure makes it worth it.

Note that you can generate a more complex system. For example, arrange it so that person B has to first escape (rolling a 20), which gives person A a chance to escape (also rolling a 20). If and only if both escape then person A casts wish.

If A gets blowback, then person B wishes a reroll.

If person B gets blowback, then sacrificial wisher C (simulacrum, other party member, etc) with a wish granting magic item wishes B rerolls, or if B's reroll of A doesn't work, C also wishes B to reroll their escape.

After an average of 400 rounds, A and B escape.

Then A casts wish. 67% chance everything is fine. 33% chance of blowback.

If blowback, B casts wish. 67%*399/400 chance we have recovered and can try again tomorrow. 67%/400 chance C has to force B to reroll, with a 33% C blowback, 67% success, and 399/400 that everything is undone.

33% chance B has blowback (and either A failed or not). Then C wishes and has a 33% blowback, plus a 399/400 chance to undo both A and B's wishes.

So 33%*(33%+67%/400)*33% C (item holder) can never wish again, aka 3.63%.

33%*(33%+67%/400) chance to burn a charge on the item (aka 10.9%).

The chance that either A can never wish again is 33%/400^2 or about 1 in half a million.

For B to never wish again, A must first fail (33%), then B must fail (33%) then reroll must fail (1/400) for about 1 in 3600 chance.

If you have N wish casters and 1 sacrificial item holder, the first N-1 wish casters have a 1 in half a million or lower chance of never casting wish again.

The last one has about a 1 in 3^N*400 chance of never casting wish again.

The number of rounds it takes everyone to get out is 20^N. This will get ridiculous pretty quickly.

The trap requires all previous people exit before the next one can even try; so the trap complexity gets pretty insane pretty fast.

  • "If you require a natural 20 to get out of the trap, then the wish'd reroll has a 1/400 chance of getting out of the trap." How do you figure? Since you already attempted and made the roll the first time, doesn't the reroll also have a 1/20 chance? – Mark Wells Nov 11 at 21:23
  • Note also that in Step 4 you have to be able to escape the trap and then cast wish in the same turn. Otherwise, by the time you get to Step 5, it will have been more than a full round since you escaped the trap, and wish can't alter history that far back. – Mark Wells Nov 11 at 21:40
  • 1
    @MarkWells Wishing a reroll can force disadvantage. And yes, you need to be able to get out of the trap on a 20 on your turn without spending an action (so you can spend it on the wish). – Yakk Nov 11 at 22:37
  • You explain how to do it, but not why it works. Could you add an introduction briefly summarizing the reason why this might work or not? – Olivier Grégoire 2 days ago


Finally, there is a 33 percent chance that you are unable to cast wish ever again if you suffer this stress.

This is the final word - “there is a 33 percent chance ...”. This is not a roll that is subject to other game mechanics that affect rolls, it is the most specific thing that applies.

Using another wish to undo this one doesn’t work - it only allows you to reroll something, not to undo the casting. By the way, using wish to reroll also causes stress.

It doesn’t matter what circumstances you try to contrive; if you use wish to do anything than duplicate a spell, you suffer the stress and run the gauntlet:

The stress of casting this spell to produce any effect other than duplicating another spell weakens you.

  • 5
    I wonder if there's anything in the text you can point to that supports the notion that the word "final" is being used in its logical sense, rather than its sequential sense (terminating the list)? I find the notion you present intriguing, but wish it had a bit of support. – nitsua60 Nov 11 at 13:54
  • The emphasis on "finally" is clever but not really necessary. I think the point is that the "33 percent chance" is something unique to wish and outside any of the usual game mechanics. – Mark Wells Nov 11 at 16:07
  • 1
    Sorry but finally only indicates that this is the last point they're going to make after all the other points they made in previous sentences. It does not modify the rest of the sentence at all. – candied_orange Nov 11 at 19:23

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.