+Jeffro Johnson loves him some d4 Thieves. He’s pretty emphatic about it, as is his right.

In my Heretical DnD project, he recently quipped that when I asked a question about rolling a pile of d4s he thought, for a brief, glorious instant, that I was returning the Thief (now the, sigh, Rogue with a Thief subclass) to its glorious roots.

That got me thinking, though. How different is the d4 Thief from BECMI to the d8 Thief from Fifth Edition?

*The BECMI Thief and Fighter*

Let’s say that our thief has attempted – unsuccessfully – to pick the pocket of a lonely neighborhood fighter. Both are first level.

I’m going to first assume that we roll 3d6, but can assign stats. Based on my work with the Standard Array, those stats for both fighters, at the median roll (50th percentile in luck) are 14, 12, 11, 10, 9, 7.

Oof.

What does that mean? It means that there is precisely one score that will get a bonus: a +1 for the 14.

The Thief has a prime requisite of Dexterity, so that’s where his +1 will go. Leather armor, no shield. So basically, he’ll have an AC of 6, or a roll of 13+ will hit him.

Our fighter will have Strength as his choice, so he’ll roll 1d20+1 to hit, and thus will hit on a 12 or higher – 45% of the time. He’ll do 1d8+1 damage with his sword, or 5.5 points on the average. Each turn, he’ll deal an average of 2.48 points of damage.

In short, he’ll kill the thief on the average in about 1.01 attacks.

*The 5e Thief and Fighter*

Let’s look at the 5e fighter, using the 50% percentile instead of the standard array. They’re not that different, but the Standard Array actually represents the 45% percentile of die rolls.

The important thing for our human fighter is still his Strength, and his standard array gives 16, 14, 13, 12, 10, 9 – actually one better than the standard array in both the highest and lowest score. With the right selection of race – Dwarves, I’m looking at you, Mountain Dwarf – you can start with STR 18. This is impossible with the standard array.

Swinging a battle axe, then, he’ll roll 1d20+2 (proficiency)+4 (STR) for 1d20+6, doing 1d10+4 damage on a hit.

Our Thief also gets to play the min/max game, and we can cut right to the chase. If our fighter can start with STR 18, our Thief can as well – Wood-Elf Thief for the win.

With DEX 18 and CON 13, this gives him 9 HP. Studded Leather and no shield, but a +4 DEX bonus for armor class, and he’s AC 16.

So our fighter has to roll 10+ on 1d20, and will hit 55% of the time. He will do 5-14 damage, plus a bit more for a critical hit – an average on a hit of 9.78 damage. Or 5.38 per round.

This means our 5e Thief will, in general, withstand 1.67 blows from our fighter. A 1d8 battleaxe will increase this a touch. to 1.88 attacks to drop the thief to 0 HP.

**Parting Shot**

I had thought that with the higher damage values of the weapons due to STR bonuses and whatnot, that a d4 Thief would, in fact, be perhaps as robust as a d8 thief. But no. By and large he takes another swat to put him down – a bit less.

What would equality be? You’d need to have the typical HP of the thief equal the typical damage done in one turn by one swing of the sword. That’s 4.8 to 5.4 HP. Call it 5, and . . . you need to get back to the d4 Thief in order for the classes to be as fragile as they were in BECMI.

Why? Armor classes are higher due to higher bonuses, which offsets weapon damages. Higher bonuses from STR are offset by higher DEX bonuses, though average damage is higher. The real boost comes from giving 1st level characters maximum HP per Hit Die at 1st level. If 1st level 5e characters rolled dice instead of getting the max, a 1st level thief would need an average of 4.8 to 5.4 HP to be as robust as BECMI. That’s basically a d9 rather than a d8!

So yes! The d8 5e thief is actually *less *robust than the d4 Thief from BECMI if you roll the first level HP, but if you don’t, well, obviously he’s more robust.

It’s the lack of randomness for rolling hit points that makes the difference. If you gave all 1st level BECMI thieves 4 HP to start, then they’d wind up with almost exactly the longevity of a 5e Thief.

It’s like the designers thought about this or something.

Because of the way that damage works, the minimum time it will take for the fighter to kill the thief is 1/(chance to hit), which assumes that any hit is an autokill (true of a thief with 2 or fewer hit points). Crunching the numbers, I find that it will take an average of around 2.5 rounds for that fighter to kill any random first-level d4 thief of that AC.

*minimum

averagetime