Movement in Combat

Holmes’ second combat example is one of the few detailed explanations of early D&D combat; that it was edited by Gygax also lends it a certain gravitas. That said, there are two main challenges with this example:

  1. Movement rates,
  2. The rate of bow fire.

In this post, I’ll discuss movement rates. In a follow-up post, I’ll focus on the rate of bow fire.


The first challenge facing the reader of Holmes’ second combat example begins back on p9 with movement rates.

The Holmes manuscript has a normal man moving 240 ft/turn. In the editing process, Gygax made this the “exploring/mapping” pace and then doubled it for “moving normally” [1]

But what does “moving normally” really mean? In the real world, a steady walking pace is around 3 mph[2] and a strong march is around 4 mph[3]

For our purposes, let’s suppose “moving normally” with a bunch of adventuring gear is around 3 mph. That’s 15,840 ft/hour or 2,640 ft/ 10 minute turn. Right away it’s clear that Gygax’s 480 ft/turn is roughly one-sixth of a realistic walking pace (480/2640 = 0.182).

When it comes to melee combat, Holmes p20 specifies that “an unarmored man can move 20 feet per melee round”. Despite there being 10 melee rounds per turn, 20ft is 1/12th of the unarmored man’s exploring/mapping rate. There are two problems with this:

  1. Using 1/12th (rather than 1/10th) of a move arbitrarily reduces movement pace by one-sixth.
  2. But the big one is that combat doesn’t happen at “exploring/mapping” pace. In fact, Holmes is explicit that: “Movement (if any) is usually at a sprint”[4]Holmes Basic rulebook, p20.. Combat movement should therefore be calculated from a sprinting pace, not from a mapping/exploring pace.

Which brings us to Running.


The Holmes ms includes a running pace of treble the original movement pace (240 ft/turn x 3 = 720 ft/turn). Unfortunately, Gygax did not treble the newly introduced “walking normally” pace for running but left it at Holmes’ original 720 ft/turn. Had Gygax trebled the new normal walking pace (480 ft/turn) running pace would have been (480 ft/turn x 3 =) 1,440 ft/turn.

Realistic run speeds vary widely by distance and fitness but an average, maintainable jog/run speed is something like 5-6 mph[5] while a New York Marathon qualifying time is around 8 mph[6] However, D&D combat is more about short bursts of maximum pace rather than maintainable run speeds. The average sprinting speed for healthy adults is around 10-15 mph[7]

Assuming that even an unarmored D&D character will be carrying some adventuring gear, let’s call it 9 mph. This is conveniently triple our 3 mph walking pace, as originally specified by Holmes.

9 mph = 47,520 ft/hour = 7,920 ft/10 minute turn.

So a realistic run pace is around 792 ft/minute (which couldn’t actually be sustained for a whole minute!) which is 132 ft/10-second round. So the Holmes’ official movement rate for an unarmored man of 20 ft/round is once again about one-sixth (20/132 = 0.152) of a realistic sprinting pace.


Gygax and later authors clearly recognised this issue.

In the first print of B2 The Keep on the Borderlands Gygax reiterated the doubtful 20ft/round movement rate exactly as it appears in Holmes[8]The Keep on the Borderlands 1st print (1980) p3, but in later prints, this was doubled from 20 to 40 ft/round [9]The Keep on the Borderlands 4th print (1981), p3 .

In the following iteration of the D&D Basic Game, Moldvay continued with the (now doubled) unencumbered movement rate of 40 ft/round and (finally!) updated the triple-speed running rate for “running at full speed toward or away from the enemy” to 120 ft/round[10]Moldvay Basic Rulebook p20. It so happens that 120ft/10 seconds is close to our real-world sprint speed (132 ft/10 seconds), above. A 120 ft/round sprint with adventuring gear is quite believable.

Zach Howard of the Zenopus Archives blog[11] commented that:

the Holmes numbers are so low that it takes too long for a fighter to get across a room to help someone else … I make a distinction between movement when actually engaged in melee with an opponent (where I use the Holmes numbers, although it doesn’t come up often), versus movement during combat when not engaged, when I’ve used the running rate in Moldvay

(Personal correspondance).


The simplest fix might seem to be to reduce turns and rounds to approximately one-sixth of their current duration. I.e., 2-minute turns and 2-second rounds would go a long way toward addressing the pace discrepancies. However, the 10-minute exploration turn is a staple of original D&D and is even believable while sneaking along cautiously, searching for tripwires or secret doors, or pacing out distances for detailed mapping. Furthermore, if 2-minute exploration turns were adopted then torch/lantern burn times, spell durations, wandering monster frequencies, and monster movement rates would all have to be adjusted accordingly. In short, this would be a sweeping change not undertaken taken lightly!

So, what else could we do?

Visibility is an important consideration underground. It’s plausible that movement underground is limited as much by visibility as it is by mobility; you tend to slow down when you can’t see where you’re going. As Holmes notes “humans and halflings will need artificial light or be reduced to half speed or less”[12]Holmes Basic rulebook, p9 . Holmes does say or be slowed by half, but even the 30 ft illumination provided by a torch or lantern is poor visibility compared to daylight. Plausibly, one-half of the normal daylight movement rates could be considered standard for underground movement by torchlight.

A further observation is that encumbrance is less impactful on walking rate (and hence on exploration/mapping rate) than it is on running rate. The Roman legions as well as modern soldiers continue to march at 3 and 4 mph all day even burdened with their heavy gear[13] If movement rates underground are already limited by visibility, it seems unnecessary to alter them again for encumbrance. Therefore, underground exploration rates could be set by activity (mapping, walking normally) rather than by encumbrance. The impact of encumbrance would be reserved for combat/pursuit movement rates.

So what would all this look like?


This table could (if desired) replace the Holmes MOVEMENT TABLE (p9).

Note that these movement rates are obvious products of 12″ and scale inch movement rates, and that encumbrance is primarily relevant to running/combat movement rates.

If these alternative movement rates are adopted, the monster movement rates given in Holmes per turn become movement rates per round without any further change. This would be very convenient.

This table retains the official 120 and 240 ft movement rates for the predominant game activities (albeit for mapping and exploration, respectively) while acknowledging that these are substantially slower than “moving normally”.

The underground or torchlight movement rate (1,200 ft/turn) might be useful when passing through a previously explored torchlit area, for example, whereas the daylight movement rate (2,400 ft/turn, around 2.7 mph) would rarely be used in the dungeon game. It is included for completeness.

Moldvay’s very believable “full speed” movement rates are adopted for 10-second combat rounds.


  • The movement rates given in Holmes are approximately 1/6th of realistic walk and sprint speeds.
  • Holmes’ by-the-book movement rates of 120 and 240 ft/turn can be retained for mapping and exploring movement, respectively. These movement rates are already so slow they require no further adjustment for encumbrance.
  • A normal walk in full daylight is around 3 mph, or close to 2,400 ft/10 minute turn. This could be halved underground to 1,200 ft/turn to account for poor visibility, awkward floors, and increased caution.
  • Combat/pursuit movement rates of 10x the scale inch movement rates were introduced in Moldvay Basic D&D. These are intuitive to calculate, practical in play, and close enough to realistic. As a bonus, the monster movement rates listed in Holmes can be immediately used as combat/pursuit movement rates per round.


3, 13
4 Holmes Basic rulebook, p20.
8 The Keep on the Borderlands 1st print (1980) p3
9 The Keep on the Borderlands 4th print (1981), p3
10 Moldvay Basic Rulebook p20
12 Holmes Basic rulebook, p9