# D&D 5ERolling Hit Points Tweak

##### (they/them)
PC hit points are very important. If you have too few you are likely to be taken out by a single attack from a brute of your CR.

With bounded accuracy and multiattacks from monsters, if you roll low for your hit points, you are in trouble.

Luckily you can just choose to take a set number of hit points to avoid that. Not only that, but that set number is highly than the average you would get from rolling.

Rolling for hit points can be fun though. If you want to keep rolling for hit points but don't want to lose out here is a slight tweak:

d12 = 2d6
d10 = d6 + d4
d8 = 2d4
d6 = 1d4 + 1d2

This weights your rolls toward the middle, has a minimum of 2HP, has the same maximum, and gives you an average equal to what you could choose without rolling.

Any problems you can see with it?

#### Shiroiken

##### Legend
Your averages would be slightly higher, but overall I could see it work. I would consider using 1d8+1d4 for d12 and 1d6+1d2 for 1d8, as they give you a flatter spread.

Personally, I'm fine with using average HP (even true average), rather than rolling. Those who like rolling however, would probably hate this. They tend to like the risk/reward aspect, and while your minimum is higher, you have much lower chances of getting a really high roll.

##### (they/them)
Your averages would be slightly higher, but overall I could see it work. I would consider using 1d8+1d4 for d12 and 1d6+1d2 for 1d8, as they give you a flatter spread.

Would it be higher? The average of 2d6 is 7. If you have 1d12 hit die you can choose to take 7. Rolling normally you would get an average of 6.5 which is lower.

I suppose which dice should be used depends on the goal and whether it is to just make the average work out right. I suppose using these numbers 1d6 should be 2d3.

Personally, I'm fine with using average HP (even true average), rather than rolling. Those who like rolling however, would probably hate this. They tend to like the risk/reward aspect, and while your minimum is higher, you have much lower chances of getting a really high roll.

I tried it with my group today. They seemed to like it. They don't like to just take hp. I am finding in 5e that HP are much more important than they were before.

I suppose I could just let fate decide things. Let them die if they roll low on their HP.

Personally this is a system I would use when playing instead of choosing the average. I like to let aspects of my character be up to chance. But rolling straight 1dX in 5e feels like rolling 3d6. It achieves the goal of putting some randomness in, but at the cost of potentially making an unplayable character.

#### CapnZapp

##### Legend
The way the average is rounded UP when you take the set value, means that taking a risk is not only not encouraged, but actively worked against.

For you to take a risk, there rationally needs to be a reward. Rounding DOWN would have meant such a reward would have been present, however small it is.

(Half a point is not much, but at least it's double the reward when you're rolling a d6 than when rolling a d12. Meaning wizards would gain more by rolling than barbarians in this scenario.)

But the rules round up.

Only a fool or compulsive gambler would choose to roll when this statistically hurts you, even before you factor in the risk.

Besides, rolling low hurts you more than rolling high helps you, especially at the lowest levels.

Last edited:

#### Shiroiken

##### Legend
Would it be higher? The average of 2d6 is 7. If you have 1d12 hit die you can choose to take 7. Rolling normally you would get an average of 6.5 which is lower.
I was comparing rolling to rolling, not to the "average" option available. [MENTION=12731]CapnZapp[/MENTION] is correct that you are always better of taking the "average" than rolling in 5E.

I suppose which dice should be used depends on the goal and whether it is to just make the average work out right. I suppose using these numbers 1d6 should be 2d3.
True.

I've decided that I don't like rolling 2 dice of the same type anymore, because the average is by far the most likely thing to roll. With 2 different dice, the bell curve widens out to give a larger group a higher percent, without making the average the most probable. For example, with 1d8+1d4, you are just as likely to roll a 5, 6, 7, 8, or 9, whereas with 2d6 you are most likely to roll a 7, then likely to roll a 6 or 8, and then 5 or 9. It's personal preference I guess.

I tried it with my group today. They seemed to like it. They don't like to just take hp. I am finding in 5e that HP are much more important than they were before.

I suppose I could just let fate decide things. Let them die if they roll low on their HP.

Personally this is a system I would use when playing instead of choosing the average. I like to let aspects of my character be up to chance. But rolling straight 1dX in 5e feels like rolling 3d6. It achieves the goal of putting some randomness in, but at the cost of potentially making an unplayable character.
I'm glad your group liked it. Sounds like you found a great house-rule

#### ExploderWizard

##### Hero
I take a simpler approach to hit point rolling in my 5E campaigns. At each level the player simply rolls the appropriate hit die. If the roll of the die isn't at least the default average for the die type then they may take the average.

No risk. Just a shot at getting better than average. Its win-win for the players.

Monsters get the same break. I roll HD and the monsters either have average HP or better.

#### bgbarcus

##### Explorer
I give my players a choice of how to roll hit points each time they gain a level.

Option #1. Roll three dice and take the best one.
Option #2. Roll one die but with a minimum of the average for that die.

Option #2 is safe but option #1 offers better chances for a high roll. Most players take option #1 and there have been a couple of times when three 1's were rolled.

#### CapnZapp

##### Legend
I lowered the static number to the rounded down average.

This to encourage rolling hp, partially since I outruled rolling for stats.

And yes, already with such a small sample this shows how rolling bad is more negative than rolling good is positive.

I would definitely take the given number, at least for the first few levels to ensure my survivability before starting to roll around when I already have 30 hut points or so.

I feel bad for players that love to gamble, or perhaps simply rolls out of habit. The 5E rules screws you so bad...

TLDR The chance at rolling a 10 on the d10 doesn't outweigh the risk of rolling a 1 on low levels. You really shouldnt risk the roll unless the potential reward is made even greater than I made it. (see upthread for ideas that does make the risk of rolling worth taking)

#### HorusZA

##### Explorer
If you want to have a bell-curve without changing averages, there's a simple technique:
Roll 3 dice of the required type and pick the middle result.

This doesn't change the possibility that a player could still land up gaining just 1 HP but the chances are much reduced (using a d8, for example, the chance is 4.3% as opposed to 12.5%)

Last edited:

##### (they/them)
If you want to have a bell-curve without changing averages, there's a simple technique:
Roll 3 dice of the required type and pick the middle result.

This doesn't change the possibility that a player could still land up gaining just 1 HP but the chances are much reduced (using a d8, for example, the chance is 4.3% as opposed to 12.5%)

The idea is to change the average though.

The average result of 6.5 on 1d12 is too low when you can just choose to take 7.

For those who are increasing the amount of HP the players are getting, how do you find that? Why not just give max HP at each level at that point?

#### fewilcox

##### First Post
PC hit points are very important. If you have too few you are likely to be taken out by a single attack from a brute of your CR.

With bounded accuracy and multiattacks from monsters, if you roll low for your hit points, you are in trouble.

Luckily you can just choose to take a set number of hit points to avoid that. Not only that, but that set number is highly than the average you would get from rolling.

Rolling for hit points can be fun though. If you want to keep rolling for hit points but don't want to lose out here is a slight tweak:

d12 = 2d6
d10 = d6 + d4
d8 = 2d4
d6 = 1d4 + 1d2

This weights your rolls toward the middle, has a minimum of 2HP, has the same maximum, and gives you an average equal to what you could choose without rolling.

Any problems you can see with it?
I can always get behind attempts to trade linear dice rolls for curves (even if those curves are simple Vs). GURPS use of 3d6 in its core mechanic is a huge part of why we love it so much. The dice gawds generally hate me, which is why my major complaint about HackMaster is that it requires players to roll their characters' stats.

#### PeelSeel2

##### Explorer
We always played you rerolled your hit points at every level. If you got higher, they went up. If they where lower, they stayed the same. Helps average them out.

#### the Jester

##### Legend
My tweak is "roll but treat any number less than half the die's maximum as half the die's maximum". So, if you're gaining a d8 Hit Die upon leveling up, you can take 5 hp... or roll 1d8 with a minimum result of 4.

So far, nobody has taken the 5 hps.

#### AriochQ

I run a campaign where most combats are pretty challenging. I allow my players to roll, but if they roll less than the average, they are allowed to take the average (rounded down). It is pretty lenient, but it is no fun gaining a level and rolling really low for HP's. And the extra few HP's allow me more leeway in making the encounters tougher.

The way I see it, the average for a d10 is 5.5 The average for my method would be 7.5. It inflates HP's a little bit, but avoids those low HP rolls.

#### bgbarcus

##### Explorer
I run a campaign where most combats are pretty challenging. I allow my players to roll, but if they roll less than the average, they are allowed to take the average (rounded down). It is pretty lenient, but it is no fun gaining a level and rolling really low for HP's. And the extra few HP's allow me more leeway in making the encounters tougher.

The way I see it, the average for a d10 is 5.5 The average for my method would be 7.5. It inflates HP's a little bit, but avoids those low HP rolls.
That matches my thinking. Using the best if three rolls means most characters get high hit points for each level. But gambling on the dice is also fun. One character had a two really bad rolls (three 1's at second level and highest roll was 2 at third for a rogue). His low hit points became a fun part of role playing the character as he became very good at hiding.

That was in the final play test when rogues had d6. When the PHB came out and they were bumped to d8 I let him reroll all levels.

##### (they/them)
We always played you rerolled your hit points at every level. If you got higher, they went up. If they where lower, they stayed the same. Helps average them out.

What does that typically look like?

What extremes have you seen?

Good to know that other people are taking measures to ensure HP are high.

#### Mercule

I don't see anything wrong with the OP's method.

Another option is to roll two dice each level, and literally take the average of them. It sounds odd, but the math is pretty close. Take the cleric rolling d8. You can simulate a d4 by halving a d8 and rounding up. Do that twice and call it a day. That actually gives 3/64 (8, 8; 7, 8; 8, 7) chance of an 8 and only 1/64 of a 2 (1, 1). I'm not going to check the math on the rest of the curve.

If you want to up the average while retaining an even distribution, you could roll the next smaller die and add 2 to the result. The above Cleric has a range of 3-8 hp, with a flat distribution. If the desire is to eliminate the extremes, add just 1 instead of 2. The Cleric now has a range of 2-7.

If you want a bell curve that centers high, do both: roll two dice of the next size down, halve that, and add 2. The Cleric gets 2d6/2 + 2 for a range of 4-8 and an average of just above 5.5 (functionally 6). Really, though, all classes average around 75% of their hit points (d4=>5, d8=>6, d10=>8, d12=>9) with a minimum of 3 and the same maximum they have, now. A small tweak can scale the minimum, too.

#### pming

##### Legend
Hiya.

In my games, if a player rolls for HP's upon gaining a level and doesn't like his result, he can opt to have me roll it for him...but he keeps it, even if it was lower. This usually works out better for the player, as a fighter rolling a "3" has a much better chance of me rolling a 4 or higher than a 1 or 2. That said, just yesterday the Sorcerer rolled HP and got a 1. Obviously he had me reroll... and I got a 1. But, as I said, that almost never happens.

This keeps the players honest in their rolling, knowing that if they get a sucky dice roll they can just have me do it in stead of trying to 'cheat' and then feel bad for doing so (or risk getting caught).

^_^

Paul L. Ming

#### Staffan

##### Legend
In the past, my group has used a rule that if you're unhappy with your hp roll, you can reroll it with a die one step smaller, and you can invoke the rule multiple times. So if a fighter rolls 2 on his d10, he can reroll with a d8. If he then rolls a 1, he can reroll with a d6. This gives some cushion against super-low rolls without raising the average too much.

#### Larac

##### First Post
I have always had them Take AVG. if the roll was low.

If a player rolls badly two times at low levels the Char. becomes a hindrance for the GM and the player.
Just not worth it.

Replies
32
Views
2K
Replies
21
Views
2K
Replies
10
Views
1K
Replies
5
Views
1K
Replies
3
Views
2K