Random violence stats I've compiled

The #1 community for Gun Owners in Indiana

Member Benefits:

  • Fewer Ads!
  • Discuss all aspects of firearm ownership
  • Discuss anti-gun legislation
  • Buy, sell, and trade in the classified section
  • Chat with Local gun shops, ranges, trainers & other businesses
  • Discover free outdoor shooting areas
  • View up to date on firearm-related events
  • Share photos & video with other members
  • ...and so much more!
  • BehindBlueI's

    Grandmaster
    Rating - 100%
    29   0   0
    Oct 3, 2012
    25,895
    113
    These are mostly cases I worked in some capacity (lead or assisting), but do include some from detectives I've asked to notify me of instances of citizens fighting back so that I can get information from Sex Crimes, etc. that I don't work.

    19 total cases. I'd hoped for a larger sample, but after I culled domestics and criminal v criminal encounters I had fewer than I thought I would. I've gone through over 300 cases at this point to get these, which is a combination of two factors. Few people are targeted by random crime, and fewer fight back when they are.

    Total shots fired:
    N/A (weapon used was not a firearm) x5
    0 shots x5
    1 shot x4
    3 shots x3
    4 shots x1
    11 shots x1


    Number of shots to resolution (suspect disabled/began to flee or citizen disabled/disarmed):

    0 shots x7
    1 shots x6
    3 shots x1

    Distance at time of resistance
    Contact x 7
    0-3' x 5
    3-7' x 3
    7-21' x 3
    21' + x1 (bad shoot)

    Was speed of the draw a consideration in the outcome:
    Yes, from body carry x5
    Yes, from off body carry x2
    No, had firearm in hand prior to need to engage x3
    No, had other weapon in hand (knife, blunt, etc) prior to need to engage x 2
    No, draw speed of weapon was not a consideration (disabled prior to beginning draw, bad shoot and suspects already fleeing before 1st shot, etc.) x5
    N/A, unarmed resistance x2

    Reload:

    0 incidents

    Additional equipment needed:

    0 incidents

    Win/loss/tie (tie defined as at least one party on both sides disabled/killed)

    W x 13
    T x 1
    L x 5

    Outcomes of win

    Suspect fled without sustaining injury x6
    Suspect killed x3
    Suspect disabled/psychologically stopped due to injury x3
    Suspect restrained (not seriously injured) x1


    Outcome of tie:

    Victim injured during commission of crime/suspect injured during flight

    Outcome of loss
    Victim failed to deactivate thumb safety, victimized
    Victim carrying with empty chamber, killed
    Victim killed reaching for weapon
    Victim engaged 2 suspects he can see, shot from behind, suspects flee
    Victim wrestled suspect, was shot, suspect fled.


    Citizen weapon used:
    Victim weapon
    Handgun-pistol x7
    Handgun-revolver x5
    Handgun - other x1
    Unarmed x2
    Golf club x1
    shotgun x1
    knife x1
    taser x1

    Suspect weapon
    Unarmed x13
    handgun x7
    unknown weapon x1

    Number of victims

    1 victim x14
    2 victim x3
    3 victim x2

    Number of suspects

    1 suspect x 10
    2 suspect x 5
    3 suspect x 4

    Citizen averages vs single suspect:

    0.67 shots

    80% won encounter

    Citizen averages vs multiple suspects:

    0.57 shots
    50% won encounter

    Citizens with firearms won 64% of the time. However citizens who could employ their firearm (removing those reaching for off body carry, failing to deactivate safeties, etc) won 81% of the time. Other methods of resistance had an 80% success rate. The element of surprise was more important than the specific weapon employed, ie ambushing someone with a golf club and knocking their teeth down the stairs was better than attempting to reach for a handgun while under direct observation of an armed suspect. Citizens using other weapons were more likely to use ambush tactics and were also closer to their opponent.

    Due to the small sample size, this should be viewed as preliminary information at best. I will attempt to build on this database and will attempt to update every 20 incidents, (ie next update at 40 incidents)
     
    Last edited:

    T.Lex

    Grandmaster
    Rating - 100%
    15   0   0
    Mar 30, 2011
    25,859
    113
    LOVE what you're trying to do here. Statistics are part of what I've done professionally for a long time.

    Quick question as I just skimmed your post before going deeper. What do you mean by:
    Total shots fired:
    N/A (weapon used was not a firearm) x5
    0 shots x5
    1 shot x4
    3 shots x3
    4 shots x1
    11 shots x1

    Number of shots to resolution (suspect disabled/began to flee or citizen disabled/disarmed):

    0 shots x7
    1 shots x5
    3 shots x1

    How can the total number of times 0 shots were fired (5) be less than the times 0 shots were fired before resolution (7). Even if you take into account the non-firearm resistance (5), what is the discrepancy between the 7 and 10?

    Oh, and I totally think you can further develop these observations into a training/seminar topic.

    ETA:
    If possible, a further breakdown of Total Shots and Shots to Resolution might be separate good guy v. bad guy shots. In other words, how many times does the bad guy open fire/return fire? I have an amateur's suspicion that even when a bad guy has a gun, there is at least an equal likelihood that they will flee before using the firearm.
     

    BehindBlueI's

    Grandmaster
    Rating - 100%
    29   0   0
    Oct 3, 2012
    25,895
    113
    I think one of the big take aways here is that those willing to fight and able to fight won 80% of the time in a "fair" fight of 1 on 1. Folks, that's huge. Untrained average Joes won 4 out of 5 times. That's higher than I thought it would be, I thought it was going to be closer to 50/50, but as I culled out criminal on criminal it kept going up.

    So if Average Joe wins 80% of the time, how much better will your odds be with some realistic training and regular practice? Probably knocking around 100%, although there are always factors you can't control simply increasing your draw speed a little bit, distracting the bad guy just a little bit, seeing trouble coming just a little earlier, that'd up these numbers. Multiple opponents, well, that's a factor you can't control but what about some training on managing multiple contacts? Or a realistic understanding of what you can and can't accomplish? Note the total and average round count is lower for multiple opponents, it's because they LOST before they got more rounds off. Is "faster" what you normally hear when people talk about multiple attackers? No, they focus on capacity. You have less time to use that capacity, so are you focusing on the right thing?

    Speed, surprise, ferocity of attack.
     

    BehindBlueI's

    Grandmaster
    Rating - 100%
    29   0   0
    Oct 3, 2012
    25,895
    113
    How can the total number of times 0 shots were fired (5) be less than the times 0 shots were fired before resolution (7).

    Because sometimes the total number of shots is greater than the number of shots required to resolve the situation. There are times when display only resolved the situation (0 shots) but the person shot anyway. Example:

    You are unarmed. You attempt to mug me. I pull a gun and before I fire any shots you begin to run. Shots fired to resolution: 0
    I complete my draw. You are still fleeing, you are 21'+ and are no threat to me. I shoot at you 3 times. Total shots fired: 3
     

    T.Lex

    Grandmaster
    Rating - 100%
    15   0   0
    Mar 30, 2011
    25,859
    113
    Because sometimes the total number of shots is greater than the number of shots required to resolve the situation. There are times when display only resolved the situation (0 shots) but the person shot anyway. Example:

    You are unarmed. You attempt to mug me. I pull a gun and before I fire any shots you begin to run. Shots fired to resolution: 0
    I complete my draw. You are still fleeing, you are 21'+ and are no threat to me. I shoot at you 3 times. Total shots fired: 3

    Ah - makes sense now. One wrinkle: in that example, and more generally, can there be a role reversal of victim/aggressor? That may be a bridge too far, because tracking the data would get more complicated.

    Are you keeping a spreadsheet? I certainly don't intend to try to make more work for you, but you have a great start on something that can be meaningful. Using the right tools early can help you down the road. I've been impressed with Google sheets, partly because it can be accessed remotely.

    Also, you may not have caught my edit - do you have data on victim/aggressor shots? Your data implies those numbers are available, but it isn't clear.
     

    BehindBlueI's

    Grandmaster
    Rating - 100%
    29   0   0
    Oct 3, 2012
    25,895
    113
    Ah - makes sense now. One wrinkle: in that example, and more generally, can there be a role reversal of victim/aggressor? That may be a bridge too far, because tracking the data would get more complicated.

    Are you keeping a spreadsheet? I certainly don't intend to try to make more work for you, but you have a great start on something that can be meaningful. Using the right tools early can help you down the road. I've been impressed with Google sheets, partly because it can be accessed remotely.

    Also, you may not have caught my edit - do you have data on victim/aggressor shots? Your data implies those numbers are available, but it isn't clear.

    Yes, I'm keeping a spreadsheet.

    No, there is no role reversal in these stats. In my example, you tried to mug me. Your criminal actions led to me shooting at you, even if it was a legally bad shoot. That means you are culled from the statistics as a victim, because you were not targeted randomly, you were targeted for your criminal activity. This is the same reason I culled out gang related shootings and dope related shootings. Including them would make the statistics irrelevant to random victims.

    Yes, I have data on criminal's total shots and citizen's total shots. (and yes, I know the criminals are generally citizens as well, but it makes it more clear than 'victim' or whatever). I am still working on how to report these meaningfully, but in general when the citizen wins the bad guy fires no shots. When the citizen loses, the bad guy fires up to 3 shots. I don't want to just report the total and average, I think I'm going to break it down into "when X wins" and "when Y wins"
     

    injb

    Sharpshooter
    Rating - 100%
    6   0   0
    Jul 17, 2014
    390
    28
    Indiana
    This is cool! Sorry if this is in there and I missed it, but just out of curiosity, how did victims with handguns fare against suspects with handguns? It looks like the suspects in these cases were unarmed (13) far more often than the victims (2). Most of the victim losses seem to have involved both sides being armed.
     

    T.Lex

    Grandmaster
    Rating - 100%
    15   0   0
    Mar 30, 2011
    25,859
    113
    Yes, I'm keeping a spreadsheet.
    Awesomesauce.

    Yes, I have data on criminal's total shots and citizen's total shots. (and yes, I know the criminals are generally citizens as well, but it makes it more clear than 'victim' or whatever). I am still working on how to report these meaningfully, but in general when the citizen wins the bad guy fires no shots. When the citizen loses, the bad guy fires up to 3 shots. I don't want to just report the total and average, I think I'm going to break it down into "when X wins" and "when Y wins"
    Humbly, I think it should be a breakdown in each of the "shots fired" categories.

    Something like this, that could be relatively transferred to a pivot table:
    Total Shots During Encounter
    Shots Prior to ResolutionVictim outcome (w/l)
    VictimCriminalTotalVictimCriminalTotal
     

    BehindBlueI's

    Grandmaster
    Rating - 100%
    29   0   0
    Oct 3, 2012
    25,895
    113
    This is cool! Sorry if this is in there and I missed it, but just out of curiosity, how did victims with handguns fare against suspects with handguns? It looks like the suspects in these cases were unarmed (13) far more often than the victims (2). Most of the victim losses seem to have involved both sides being armed.

    Some of that is because of multiple suspects. If there are two suspects and one is armed and one is not armed, they count in both categories.

    I haven't broken down the stats that far yet, because at this point with only 19 incidents if we start drilling down too much and get to the point we only have a hand full of incidents that match what we're looking for, how statistically valid would it be? For example, only two people resisted unarmed (although I'm about to add a 3rd). One won and one lost. Does that make it 50%? Well, yes, but two incidents is tough to draw conclusions from. The 3rd is also a win, so now unarmed resistance equates to 67% wins. Again, is that valid? Only 3 instances? Maybe when I get closer to 50-100 I'll start breaking them out more.
     

    BehindBlueI's

    Grandmaster
    Rating - 100%
    29   0   0
    Oct 3, 2012
    25,895
    113
    Awesomesauce.


    Humbly, I think it should be a breakdown in each of the "shots fired" categories.

    Something like this, that could be relatively transferred to a pivot table:
    Total Shots During EncounterShots Prior to ResolutionVictim outcome (w/l)
    VictimCriminalTotalVictimCriminalTotal


    OK, HTF did you get the table in the post? I could have saved a butt ton of time formatting if I could have figured that out.
     

    T.Lex

    Grandmaster
    Rating - 100%
    15   0   0
    Mar 30, 2011
    25,859
    113
    I haven't broken down the stats that far yet, because at this point with only 19 incidents if we start drilling down too much and get to the point we only have a hand full of incidents that match what we're looking for, how statistically valid would it be?
    In the real world, as long as it is disclosed, the numbers are what they are. A small sample size only means that the ratio may not continue over time, not that the conclusions are invalid based on currently available data.

    [ETA: in statistics, there's a magic number where the sample size shows the trend, and more data only improves the precision, but doesn't really affect the trend. That magic number depends on what you're trying to measure, though.]

    OK, HTF did you get the table in the post? I could have saved a butt ton of time formatting if I could have figured that out.

    Copy/paste, but when you paste into the reply, make sure you are in "advanced." There is a button with "table properties." Click that with the table selected and have it show borders (grid). Caveat: some vBB versions are better at tables than others. INGO's current one is ok, but not great. Some formatting will get wonky, in my limited experience using tables here.
     

    jsharmon7

    Grandmaster
    Rating - 100%
    119   0   0
    Nov 24, 2008
    7,827
    113
    Freedonia
    Thanks, BBI!

    My take away:

    The odds of being the target of random violence, even in a large city like Indianapolis, are pretty low. The odds of winning that encounter if you can successfully deploy your firearm are high. Proper training and mindset make those odds much higher. Backup mags are handy but the total number of rounds needed to stop a threat are low. Train to scan for multiple targets rather than just practicing to turn your head side to side because you saw it on YouTube and it looks cool. Practice your draw stroke.
     
    Top Bottom