A Quick Tip for Fight Scenes With Multiple Enemies

I love a character who’s badass enough to take on two, three, four enemies at once. And writing that sort of scene is really fun. There are all sorts of ways to make it believable, and lots of places to turn to for how it works. But I want to make one suggestion.

You know what the most unrealistic thing in most war movies is? The wounded to kill ratio.  It’s like, every time the hero shoots a bad guy, the bad buy dies.   That’s never been accurate.  Even in the ancient world, where the lack of medical knowledge drastically increased mortality rates, you hardly ever hear of a battle where the wounded to killed ratio is lower than 3:1. In modern warfare the ratio is significantly higher. So the next time you’ve got your protagonist surrounded by bad guys, why not disable all or most of them instead of killing them? A sword-thrust through the upper thigh, a broken arm, a sharp rap to the head, a punch to the solar plexus.  Not only will it feel more believable, but by reducing the body count, you make it more powerful when you actually do kill someone. And more powerful is more gooderful, right?

Just a thought.

Published by

Avatar photo


I play the drum.

30 thoughts on “A Quick Tip for Fight Scenes With Multiple Enemies”

  1. Oooh. That is interesting. Is there a source in particular you get the ratios from? Now I’m curious about different time periods and types of combat.

  2. I’ll add another: When your Awesome Hero has taken a good number of attackers out, and especially when the AH has taken out the majority of the attackers, the rest should run away, or at least try to. Losers rarely flee because they’re cowards; they flee because they’re smart enough to do the math, and no one wants to die for nothing.

  3. Matt: Various history books. You often get, “x killed y wounded” in military history, and I can’t remember ever coming across worse than 3:1 even in discussions of ancient Greece. US Civil War may have been the worst because of the degree to which military tech had outstripped medical tech, and most battles were 4:1 or 5:1 wounded to killed.

    Will: Good point.

  4. I agree with Will if you are talking real life. The other thing is to get the attackers to be spaced out so you are only having to fight one at a time, again real life. Defeat the leader first and the rest may just leave.

    But that might not make a good story or movie. As you say, keep pokin’ at ’em.

  5. Mr. Doyle,

    If you’re interested in a few modern stats, the wounded to killed ratio for France in all wars ranged from 7:1 to 15:1 1775-2008, the US from 2:1 to 9:1, Russia 2:1 to 5:1 (stats from mid-1880’s), and Israel 3:1 to 8:1. This is from pg. 8 of “Dead Wrong? Battle Deaths, Military Medicine, and Exaggerated Reports of War’s Demise” by Tanisha M. Fazal.

    If you’re interested in more specifics, the
    Journal of Special Operations Medicine, Winter 2007, pp. 68-9 lists specific wound:death ratios for the Soviets in Afghanistan ’79-’88, which ranged from 1:3 to 1:5.

  6. The most useful casualty to make is the one that not only takes an enemy out of the battle, but takes a companion to get the wounded one to safety.

    Now that could be well done by an author – planning to be in battle, but needing to save his companion in the face of opposition.

  7. Lee Child’s Jack Reacher novels, while anything but realistic, does have Reacher disabling more opponents than he kills, since he is often without weapons.

  8. You could surely do something interesting with this.

    Like, the hero gets attacked by four guys and he kills one and disables the other three. Concussion for two and a dislocated shoulder. He does first aid and walks away.

    Then he is attacked by the three guys and five more. He disables one and runs away, he kills the fastest one chasing him and runs away. He meets up with one of the guys he did first aid on and three strangers, and the one he spared points him out and he has to fight them.

    And each time he sees that the guys he didn’t kill are causing him more problems, he gets less interested in not killing them.

    By the time it’s 3/4 done he routinely kills all the wounded. A hamstrung guy is trying to crawl away. “Please! No! I’m begging you!” He says nothing but just walks up and kills him. No point in talking to the crawling dead.

    By the end he’s noticed that whenever he leaves live witnesses they cause him trouble….

  9. Bad cliche–

    The bad guys, with superior numbers and surrounding the good guy, still only attack one at a time. They should all attack at once. But it would be cool if they did attack all at once, and the hero still figured out a way to thwart them.

    In a real life prison killing attempt I am familiar with because of my work, one antagonist, the “bomber,” presented in front of the target with the prospect of a fistfight and drew the attention while a second confederate, the “hitter,” approached from behind with the shank. The target in that scenario was able to respond quickly enough to the hitter to keep from getting stabbed more than a couple of times and stayed on his feet until corrections officers intervened.

    Not sure how that could fit this concept but Steve’s post called it to mind. Maybe the writer could conduct a scenario where there is some kind of time limitation on the attackers, and even though he can’t win against such long odds, the hero can create a stalemate until the clock runs out.

  10. skzb: “I never knew it went as low as 2:1. Thanks for the info and link”

    You’re welcome, but oops, I don’t know how I mis-pasted the link. The actual source is at:

    Dead Wrong? Battle Deaths, Military Medicine, and Exaggerated Reports of War’s Demise


    The charts only have lines every 5 points, so the difference between wounded to kill of 2:1 and 3:1 is sort of a guess

  11. Can we include the “one hit to the head will instantly and perfectly render an enemy unconscious” problem if we’re talking about movies/television? I get so tired of that one. I even got tired of it in Buffy. It’s worst in heists: your story/plan depends completely on just being able to knock the guard out with a bonk on the head? That is a TERRIBLE PLAN.

  12. Actually, just wrote that scene–discussion of how reliable the knock-out to the chin is vs back of the head.

  13. I don’t know whether base of skull is less effective than just behind the ear. Both are pretty good, and better than chin but often harder to reach.

    But the biggest thing is that it’s much much easier to knock someone unconscious if they aren’t expecting it. One hit usually works when they aren’t expecting to be hit at all, and it’s quite unreliable when they do expect it.

  14. Strangle holds are the thing I’ve seen most misrepresented in fight scenes. Oddly, they are usually depicted taking way too long to work, but then having unrealistically drastic effect. An intermediately trained grappler can choke someone unconscious in under 10 seconds, not from blocking the airway but from closing off blood flow to the brain. But if you let go as soon as they go limp, the opponent will generally come to in 10-30 seconds or so, a little disoriented. If you actually wanted to kill someone with a choke, you’d (generally) have to continue applying pressure for around a minute after they went completely limp.

  15. My two cents: Having been a wrestler and not a boxer, I don’t know that much about knocking someone out. But for coaching I had to do concussion training, and they emphasized that most concussions are caused by the back of the head striking something. This usually happens when an athlete falls down backwards and his head strikes the ground.

    Most coaches tend to dismiss such hits because they don’t appear all that bad. But they will test after a viscous strike to the front. So the training emphasized that frontal strike rarely cause concussion, but strikes to the back of the head frequently do.

    James R: When I was wrestling in high school, we used to play around with choke holds. (We’d try them on each other—yeah, I know, we weren’t exactly wise.) The best we discovered was putting the front of your victim’s throat in your elbow and squeezing their throat between your biceps and forearm. It was hard to get it right, but if you did, the victim would get dizzy immediately and pass out within 3 seconds.

  16. According to the forensic pathologist from my last murder trial, if a standing adult of normal height falls back and lands on a hard surface with the back of their head, they will most likely die. Keep in mind, however, that my expert said he was a quack. So there’s that.

  17. Will–

    After your post last night, my wife and I read about Octavia’s passing. Very sad. I did not realize she was awarded the “Genius Grant.” Amazing writer. You knew her?


    Cross examination of an expert is a hazardous endeavor. Experts almost always know vastly more about the subject matter than the lawyer questioning them. Plus, they are usually competitive jerks who want their side to win the case so they will be retained again. In the case I was describing, about the best I could do on cross was tie him to his autopsy and point out he had not looked at the scene photos when forming his conclusions (man died nearly a year after the attack, but never regained consciousness). Then I put my expert on the stand to explain why the State’s expert was wrong. The specific mechanisms of the brain injury were critical because the State’s witness had a different version of the attack.

  18. JB: yep, you discovered a proper jiu-jitsu/judo style choke. the positioning of the crook of the elbow at the front of the throat ensures that the actual pressure from the squeeze is on the arteries to either side of the trachea.

  19. Now for the real question: if a dragaerean or an easterner fell, impacted the back of their head, and died, would they be revivifiable?

  20. You want to be careful about whacking someone on the back of the head. It’s possible for such a strike to pop the eyeballs right out of the head. Or so I’ve read, anyway.

Leave a Reply