D&D General In defense of Shocking Grasp

[Kobold Stew]

Not quite -- that's the odds of three in a row.

If there have been ten rolls over the evening (two or three combats), the chance that three (or more) of them were ones go up to about ~1.15% (1 in 87). If the player has extra attack and has made 20 rolls, the odds go up to ~7.55% (~ 1 in 13). Plus there are four players at the table, so the odd s that someone on any given night get three ones jumps again, and it disproportionately affects the player with Extra Attack or Eldritch Blast.

It's gonna happen.

 

[MarkB]

For example, if it was his third roll of the evening and all were natural ones, Bob would have a right to be bummed. That’s a one in 8000 chance.

Been there. The first time I rolled an attack using Elven Accuracy (first roll for a brand new character) I rolled a triple 1.

 

[Clint_L]

Not quite -- that's the odds of three in a row.

That’s literally what I described in my example, yes.

If there have been ten rolls over the evening (two or three combats), the chance that three (or more) of them were ones go up to about ~1.15% (1 in 87). If the player has extra attack and has made 20 rolls, the odds go up to ~7.55% (~ 1 in 13). Plus there are four players at the table, so the odd s that someone on any given night get three ones jumps again, and it disproportionately affects the player with Extra Attack or Eldritch Blast.

It's gonna happen.

No argument. Still gonna suck if you roll ten times and three of them are ones. I feel ya, Bob.

 

[CleverNickName]

The point I'm trying to make is, if you pick up a d20 and roll it, you have a 1-in-20 chance of rolling a 1. It's the same chance as any other number on the die.

You roll it and get a 3. The odds were 1-in-20.

You pick it up and roll it again. What are the odds of getting a 3? The die doesn't "remember" your last roll was a 3, so the odds are still 1-in-20.

Folks want to assign a bell distribution to a singular die roll, but we can't because a singular die roll is always linear. The chances of rolling a 10 will always be 1-in-20 no matter how many times you roll. There are other dice systems (like 3d6 or 2d10) that have curve distributions...but not the d20 system.

 

[Valetudo]

The only problem I have with shocking grasp is that the 5.5 version took away advantage against metal armor. So I still use the 2014 version of that spell in my games. Otherwise if I where to ever play a wizard I would pick shocking grasp as one of my starting cantrips.

 

[Clint_L]

The point I'm trying to make is, if you pick up a d20 and roll it, you have a 1-in-20 chance of rolling a 1. It's the same chance as any other number on the die.

You roll it and get a 3. The odds were 1-in-20.

You pick it up and roll it again. What are the odds of getting a 3? The die doesn't "remember" your last roll was a 3, so the odds are still 1-in-20.

Folks want to assign a bell distribution to a singular die roll, but we can't because a singular die roll is always linear. The chances of rolling a 10 will always be 1-in-20 no matter how many times you roll. There are other dice systems (like 3d6 or 2d10) that have curve distributions...but not the d20 system.

Yes, we all understand this. We’re not stupid.

But that’s not what your example describes. Your example describes a player not asking “what are the odds” because he rolled a natural 1 once. I assume that Bob, like most everyone else, is aware of the simple fact that the odds of that or any other roll are 1 in however many sides the die has.

You describe Bob complaining because he has rolled a 1 an unlikely number of times in a set of rolls. The odds now are more complicated because we would need to know exactly how many rolls he had made, but a single player rolling three natural ones in a single session in one of my games would be noteworthy.

 

[Horwath]

The point I'm trying to make is, if you pick up a d20 and roll it, you have a 1-in-20 chance of rolling a 1. It's the same chance as any other number on the die.

You roll it and get a 3. The odds were 1-in-20.

You pick it up and roll it again. What are the odds of getting a 3? The die doesn't "remember" your last roll was a 3, so the odds are still 1-in-20.

Folks want to assign a bell distribution to a singular die roll, but we can't because a singular die roll is always linear. The chances of rolling a 10 will always be 1-in-20 no matter how many times you roll. There are other dice systems (like 3d6 or 2d10) that have curve distributions...but not the d20 system.

well, you can make custom d20 dice, been thinking about it:
d20 dice as normal, 20 sides, but sides are numbered:
1,3,5,7,7,9,9,10,10,10,11,11,11,12,12,14,14,16,18,20

average is still 10,5
there is still 5% chance to roll a "1" or "20".

 

[CleverNickName]

Yes, we all understand this. We’re not stupid.

Apologies if I implied anyone was stupid. That was absolutely not my intent.

But that’s not what your example describes. Your example describes a player not asking “what are the odds” because he rolled a natural 1 once. I assume that Bob, like most everyone else, is aware of the simple fact that the odds of that or any other roll are 1 in however many sides the die has.

I was trying to describe the Gambler's Fallacy, the perception that rolling a certain number makes rolling it again less likely. Bob (not his real name) really struggles with this--not because he's stupid, but because he has a rigid perception of what is considered "fair." Say he rolled a nat-1 three times already tonight? Well in his mind "it's only fair" that the laws of probability repay him with three nat-20s if he keeps rolling. Of course that's not how it works, and he gets frustrated and embarrassed about it.

Why did I drag Bob into this thread about Shocking Grasp? Oh right, someone mentioned whiteroom analysis, and it's one of my pet peeves, and I have ADHD. ~sigh

Sorry @Gorck , I'll keep to the topic at hand.

I'm a fan of cantrips that let me do more than just hurt foes. For attacks, my first choice for a cantrip is always Eldritch Blast, but my second choice is always going to be Chill Touch (1d8 damage and target can't heal, no save), Ray of Frost (1d8 damage and target is slowed by 10 feet, no save), or Shocking Grasp (1d8 damage and target can't take reactions, no save).

I understand they nerfed Shocking Grasp quite a bit in the 2024 revision--the target can't make opportunity attacks (instead of losing their whole reaction), and you don't get advantage against opponents in metal armor. Even with those (harsh, unnecessary) revisions, it's still a darn good cantrip.

 

[Gorck]

I understand they nerfed Shocking Grasp quite a bit in the 2024 revision--the target can't make opportunity attacks (instead of losing their whole reaction), and you don't get advantage against opponents in metal armor. Even with those (harsh, unnecessary) revisions, it's still a darn good cantrip.

The funny thing is, I don’t recall there being a major public outcry about the 2014 version being overpowered. So I’m not sure why they nerfed it in the first place. But, personally, I think I’ll homebrew the “advantage if the target is wearing metal armor” back in for my campaigns.

 

[MarkB]

The funny thing is, I don’t recall there being a major public outcry about the 2014 version being overpowered. So I’m not sure why they nerfed it in the first place.

Probably just to avoid the occasional debates over edge cases - anything from "do the studs in studded leather count as metal armour?" to "do I get advantage against metal constructs?"