A Data-Driven Approach To Precision Rifles, Optics & Gear
Home / Ballistics / How Much Does It Matter? / How Much Does Group Size Matter?

How Much Does Group Size Matter?

As long-range shooters, we tend to obsess over every little detail. We think everything is important! After all, we’re trying to hit relatively small targets that are so far you may not even be able to see them with the naked eye. While you can get away with a lot of minor mistakes and still ring steel at short and medium ranges, as you extend the range small mistakes or tiny inconsistencies are magnified. So, most things are important … but to differing degrees.

So if we have a limited amount of time and money, where would we get the biggest return on investment? In other words, there are lots of things I could focus on (more precise rifle, better scope, more consistent handloads, more practice, etc.), but where should I spend my time and money to get the biggest improvement in the probability of getting a hit at long range? How do I recognize when I’ve hit that point of diminishing returns? Tough questions.

There are so many variables that it’s easy to get lost, and most of us end up doing our best to spread our finite energy and resources in every direction. Is there a data-driven approach to help guide us toward the most important factors to increased hits at long-range?

I’m glad you asked! Bryan Litz created the WEZ (Weapon Employment Zone) analysis tool to gain some insight into this dilemma. So I dropped $200 for the Applied Ballistics Analytics Software Package, which allows you to run your own WEZ analysis. This gives you the ability to systematically study how different field variables in real-world shooting affect the probability of hitting long-range targets. Here is his summary of this software package:

“The Applied Ballistics Analytics software tool is a full-featured ballistics solver that includes the capability to compute expected probability of hit using the same Weapon Employment Zone (WEZ) method described in Bryan Litz’s book Accuracy and Precision for Long Range Shooting. This tool allows a shooter to see how his rifle can be expected to perform under a wide range of conditions, and how errors contribute in causing a bullet to miss its target.

The WEZ tool appears to be doing what’s called a Monte Carlo simulation, which is a good way to model scenarios that have a certain level of uncertainty in the inputs. Monte Carlo simulations essentially play out hundreds or thousands of possible outcomes based on your inputs. The variables in each scenario are randomly populated within the ranges you set and according to a probability distribution. For example, if you indicated your rifle was capable of holding a 0.5 MOA extreme spread, then it might play out one scenario where it drilled the exact point of aim, another where it hit 0.2 MOA high, another where it hit 0.25 MOA low, another that hit 0.12 MOA to the right, etc. Those shots would all still be within a 0.5 MOA group. It does that same thing for each of the variables in every scenario (muzzle velocity, wind call, range estimation, etc.), then it plays out each scenario, and plots where that shot would land. After it’s ran 1,000 different scenarios, it looks at the results of all of those and calculates your probability of hitting the target based on the variables and uncertainties you defined. Here is a screenshot of this part of the program, and I highlighted some of the key variables you can tweak.

Applied Ballistics WEZ Analysis Screenshot

I’d already read Bryan’s book on this topic, but in his examples he really only used 308 Win and 300 Win Mag ballistics. I wanted to run similar analysis on some of the more popular precision rifle cartridges. After playing around with the WEZ tool a lot, I can say it was very enlightening! It challenged a lot of my long-held assumptions about how important different aspects were. As Bryan Litz said in his book, “Looking at each variable separately teaches us how to assess the uncertainties of any shot and determine how critical each variable is to hitting the target.”

Over the next couple posts, we’ll dive into a few specific pieces to the puzzle that we as handloaders tend to fixate on. We’ll start with a big one:

How much does group size matter?

Virtually every rifle shooter loves to print a tiny group on a target. There aren’t many things more satisfying than sending multiple shots into one ragged hole. But, is there a point of diminishing returns in terms of how tiny groups relate to your probability of hitting targets at long-range?

The chart below shows how your odds of hitting a target increase as you shrink the size of your group. All the other variables are fixed, and only the extreme spread of the rifle/ammo combination is changing. I’ve graphed two different scenarios, a 10” circle target at 700 yards, and a 20” circle target at 1,000 yards.

Effect of Tighter Groups on Hit Probability

Essentially, what the chart is saying is if you were firing at a 10” circle at 700 yards with a rifle capable of 1 MOA, you’d have an 69.7% chance of hitting the target. But if your rifle capable of 0.5 MOA, that would jump to a 78.3% chance of hitting that same target. So by tightening our groups to 0.5 MOA, we’ve increased our chances of hitting the target by almost 8.6%. If we continue to refine that load, and can get to 0.3 MOA then that boosts our chances to 79.9%. So there is only a 1.6% gain there, and if you’re able to go from a 0.3 MOA group all the way down to a tiny 0.1 MOA group, your odds only increase by 0.8%. Here is a look at what the shot simulations looks like for those scenarios:

How Much Does Rifle Group Size Matter

Did it surprise anyone to see that there was only a 2.4% increase in hit percentage from a 0.5 MOA group to a 0.1 MOA group? What about just a 0.8% increase from a 0.3 MOA group to a 0.1 MOA group? I’ll be honest, it surprised me.

The blue line on the chart above represents the 20” circle at 1,000 yards, and you can see the effect of tighter groups on hit probability is far more minor for it. The reason is at longer ranges most misses are due to wind, not vertical dispersion. Litz reminds us “Wind is usually the greatest uncertainty in long range shooting, and the cause of most misses. Improving ballistic performance can increase hit percentage at long range, but even high performance rounds are highly susceptible to wind uncertainty.” These simulations were ran with the ability to call the wind within +/- 2.5 mph, which is what Bryan Litz says is what a good shooter is able to do in scenarios he framed as medium difficulty. He says a novice shooter is typically closer to +/- 4mph, an average shooter is usually +/- 3mph, and elite shooter is +/- 2mph. These simluations were programed so that the shooter would be able to call the wind within 2.5 mph 95% of the time, and most of the time (68%) they’d be able to call within 1.25 mph.

Rifle Group Size at 1000 Yards

Did you notice that? There is only a 5% difference in hit probability in a 1 MOA rifle and a 0.1 MOA rifle when you’re trying to hit a 20” circle at 1,000 yards! You can see that there aren’t many misses above or below the target. The dispersion is virtually all on the horizontal axis from wind uncertainty. So tightening groups on that size of target at 1000 yards, simply doesn’t have a significant impact. You may get more hits that were centered vertically on the target, but if you’re just looking at hit or no hit … there isn’t much of a difference in this scenario. Honestly, that surprised me, and I bet it did some of you guys too. As Litz explains in his Accuracy and Precision book:

At long range, the environmental uncertainties play a much greater role in dispersion. But at short range, the environmental uncertainties are less important and so hit percentage is more driven by raw precision capability.

This model helps illustrate the point of diminishing returns, and reminds us that when you reach a certain level of precision it takes an exorbitant amount of effort and money for relatively small improvements in performance. There are no right or wrong answers here! Of course, tighter groups are always better … but it’s up to each shooter to decide how far they’ll chase small improvements in performance. The benchrest guys take it to the extreme, but it’s up to each of us to strike the right balance for our specific circumstances. Hopefully this gives you a more objective perspective on how all that stuff contributes to the probability of getting a hit at long-range, and where the point of diminishing returns lies for one of the items we tend to fixate on the most.

One last point to keep in mind, is that all of this analysis assumes you have centered groups. That means they represent the best case scenario for hit percentage, since your odds only decrease if groups come off center. If you’re scope isn’t zeroed, or your rifle is canted slightly to one side, or your scope’s clicks aren’t calibrated correctly, or you pull the shot … then your hit probability can decrease dramatically. But these simulations assume we have all that stuff squared away.

Other Posts In This Series

This post was one of a series of posts that takes a data-driven look at what impact different elements have on getting hits at long-range. Here are some others posts in this series:

If you want to dig more into this subject or explore some of these elements for your specific rifle, ammo, and ballistics, I’d encourage you to buy the Applied Ballistics Analytics Package to run these kinds of analysis yourself. You could also pick up Bryan’s Accuracy and Precision for Long-Range Shooting book, which has a ton of great info on these topics and other aspects of shooting.

Enjoy this type of data-driven information? That’s what this website is all about. Sign-up to receive new posts via email.

About Cal

Cal Zant is the shooter/author behind PrecisionRifleBlog.com. Cal is a life-long learner, and loves to help others get into this sport he's so passionate about. His engineering background, unique data-driven approach, and ability to present technical and complex information in a unbiased and straight-forward fashion has quickly caught the attention of the industry. For more info on Cal, check out PrecisionRifleBlog.com/About.

Check Also

Muzzle Velocity Statistics for Shooters

Muzzle Velocity Stats – Statistics for Shooters Part 2

Part 2 in my Statistics for Shooters 3-part series focuses on how to analyze muzzle velocity consistency, which is critical for us as long-range shooters. This article shows how to apply the concepts from Part 1 to get more insight and make better decisions related to muzzle velocity. It provides practical answers to some age-old questions: Should we look at ES or SD? How many shots do we need to fire in a string? How do we get the most value from the shots we fire at the range? I spent an absurd amount of time arduously crafting this article and creating visuals so it was approachable by shooters who aren’t math nerds because I firmly believe these concepts can help a TON of people in the long-range community.


  1. Hi Cal

    Most of my graduate work was with stochastic models of far more complex systems than this, but from my experience, what ultimately matters is the distribution used to decide each of the state variables (each part of a system you are modeling). In the case of ground water plumes for instance, data sets with tens of thousands of data points could be used to determine with probability distribution function the variable follows so you could then plug it into the simulation. You’d run it a million times (literally) a thousand is low, and see what you got. You’d then do what we would call a history match, go back in time, look at an initial set of data, run the model to the present, and see how close they were. The biggest factor affecting how accurate the outcome was was how you described each one of the individual variables. Im unfamiliar with his software suite, but have read his book and saw the examples of a model that had one variable manipulated to show how it effects hitting a target. I point this out because while you can learn something from a model, unless you know what distribution a set of data follows, it is hard to accurately model a system.

    The ultimate limitation to any model is it’s a model and not empirical data. A guess- and a good guess- but it’s still a guess.

    For what it’s worth- those number seem correct.

    Thanks again for a great post- keep up the great work

    Sorry for the typos. iPhone. Bill

    • Wow, Bill. I appreciate the thoughtful response and insight. I totally understand that the number of times the simulation is ran can vary and should based on the specific application. I was actually guessing that it was ran 1000 times based on the shot simulation, but that was just a guess. I tried to email Bryan to get more details last week, but he has yet to reply.

      Ultimately, this is just a model. It is an approximation, but a useful one. It definitely can help bring some perspective and clarity to what has historically been a pile of wire coat hangers. It’s hard to tell where one stops and the other starts, and it was hard to untangle one variable from the rest of the pile.

      Thanks again for taking the time to give us an expert opinion on the subject! There probably aren’t a lot of shooters that could speak to something this complex, so I appreciate it.


  2. Wow. Definitely interesting. Great article.

    • Thanks, Olin. It was as eye-opening for me as it was you. Very surprising. That’s why I felt compelled to share it.


  3. First and foremost, I am very disappointed that a penis size joke wasnt included. even a simple “The age old question, does size matter?” would have sufficed!

    But seriously this was a great article diving into several interesting topics, I am so happy to see the Monte Carlo simulation show up! i loved writing super basic ones in MatLab, but I should point out that I have also made some pretty good ones in Excel. If youre interested its super easy.

    Also to add, it should be pointed out that the seemingly exponential decay of returns is actually in part a function of the target. For the sake of simplicity, imagine your shooting distribution is a square that just barely fit the circle. (that is the target circle is inscribed in the square). Right now all of the edges of your square are misses, huge wasted area. if you squished the square so that it was half as tall (same width, still centered) you reduce the missed areas by a lot! but if you fold it in half again, because the circle is nearly vertical here, you dont gain as much.

    This would be directly juxtaposed by a rectangular target. imagine a small square (the target) in a large square (the shooting distribution). This time, as you reduce the distribution vertically the removed loss area is actually linear until its shorter than the smaller square.

    I hope that made sense, I can make a pictured version and actually run the math behind it if you want.

    one last thing, this post is exactly why i believe groups are dumb. a gun that shoots 1MOA every other group and a gun that shoots 1 MOA every 100 groups are both technically “MOA capable guns”. I think we as long range shooters should be using standard deviation and pitch guns as “The guns sigma is MOA”.

    • Thanks for the comments, Adith. I definitely believe in using standard deviation from center of the group as a better way to represent shot distribution. It’s just that most people don’t talk about it that way, so I’m trying to use the same language as everyone else.


  4. Cal, I have logged on to you site so many times to read you articles about everything from scopes to powders. I have read Bryan’s book and am reading it for the second time. You have explained in detail on your writings that I have leaned so much about long range shooting. You are providing a great service to us and I thank you. Here’s the funny part, my rifle is at the gunsmiths and I’ve never fired a shot long range with it. I have went to matches at Rayners Range and asked questions from the shooters and watched them and looked at their rifles and they tell me what they think and why they do it. I can’t wait to start my journey into the art of long range shooting, but from your wisdom and how you explain it, I’ll do fine.

    • Thanks, EC. And man, that is an exciting time. I remember when I was waiting on my first long-range rifle to get back from the gunsmith. I’m a little jealous, because it really was such an exciting time to start learning all this stuff. This is one of those games that nobody ever knows it all. You can always keep digging and learn your whole life. If someone says they know everything about long-range shooting, you better watch them … because they’ll lie about other stuff too! We’re all still just trying to figure it out. Good luck, and thanks for the kind words.


  5. If only we shot at 2 moa targets in the prone position, targets that size typically are used in barricade or stress factor stages. Lets run the same data on a 1moa target. We shot a 6″ plate at 1000 yards at a match last year in a 15-18 mph wind from 4:30, amazingly enough some guys scored 2-3 hits out of 5 shots. I guess if all the MDs went to 2 moa plates for everything I would care less about precision ammo and quit when it was solid 1/2 moa.

    • Jim, I appreciate your point … but surely you aren’t implying that 6″ plates at 1000 yards are more common that 1.3-2.0 MOA targets. A 10″ target at 700 yards is a 1.36 MOA target, which is probably typical target size or maybe even a little smaller. If you are shooting 0.57 MOA targets (what a 6″ plate is at 1000 yards), yeah you need more precision … but realize most of us aren’t doing that. Yes, there was that one station you did that on that one time … and it sounds memorable (and kind of awesome honestly), but it would be shortsighted to use that as my examples here. I’m not saying it isn’t valid … I just don’t think it’s as applicable to my readers as the ones I’ve published. But it is a point well taken.


      • Your right Cal we don’t normally shoot sub moa targets, but seems 50% of the matches have at least one or two of them in the COF. Every match has at least one or two stages that requires us to engage one minute targets. The old saying that you can come to a rifle match with a 1moa gun and win is no longer true. My point is using the same gun derived data and a smaller target the hit percentages would not be near as close. If your readers are recreational steel shooters I suppose it’s not of any concern as to the extent of the systems accuracy. For match shooters and long range hunters the standard needs to be stepped up.

      • Absolutely! Sorry if it sounded like I didn’t believe that. I feel like a 0.5 MOA rifle is kind of the barrier to entry to be competitive. In fact, I want better than that, if only to give me a large margin of error if I pulled the shot. But what I’ve learned over the past few weeks related to this post, is there isn’t much of a benefit in trying to tighten that beyond a certain point. If you’re going to Oklahoma to compete … absolutely! After the humbling I received in that last match, I’ll take the best rifle and best ammo I can find … but I bet I’d improve my performance more if I just focused on getting better in the wind. The top shooters I watched, at least appeared to be able to read the wind as easily as reciting their home address. They were also experts in strategy and building stable shooting positions. All of that stuff comes with practice and experience. It’s just easy for us amateurs to be distracted by getting better gear or perfecting handloads, and we might be missing the elephant in the room. Spending time in my shop perfecting handloads isn’t moving me forward in the most important aspects. It’s not that it isn’t making me incrementally better … it’s just not addressing the primary reason I miss shots at long-range.

        Thanks again for the feedback, Jim. I appreciate your view. You certainly are an expert among experts in this field.


  6. Another insightful post -thanks Cal.

    However, an argument could be made that a rifle’s precision is not absolute but should be measured in standard deviation of MOA. So a rifle which traditionally claims a precision of .5 MOA actually has a SD of something like .12 MOA. Allowing for precision to be measured this way will show a little more effect of the tighter groups on long range hit probability.

    I wonder if Bryan’s program can be modified to play with this alternate theory?

    • Martin, I actually couldn’t agree more! But “when in Rome” … Most people wouldn’t know how to apply a shot distribution measured in SD, because we just aren’t used to using that measure. I try hard to make my posts approachable by just about anyone, so I stuck with extreme spread measured in MOA. But that’s not because I don’t believe what you suggested is a more accurate and valuable way to look at it. Thanks for the comments.


  7. As interesting as that is, I don’t think the simulation of wind calls is accurate. This appears to assume that EVERY wind call has up to 2MPH of error. This probably makes sense if you’re trying to model cold-bore shots. But there are other contexts…

    E.g., during a highpower or F-class match, shooters quickly adapt to current wind conditions and good ones make quite accurate calls shot-to-shot. At that point, group size makes a big difference and is highly valuable. Typically there are 2-3 stable wind conditions, and I’ve watched the high masters as they wait for conditions to return to a known state when it gets weird. They might be 2MPH off on their sighters, but by the time they’re into the string the calls are far more accurate.

    • Great point, Josh. In a Monte Carlo simulation, the inputs are generated “randomly from a probability distribution over the domain.” That means they aren’t all 2 mph error. Some are dead on wind calls (possibly even most), some are +1 mph, some are -2 mph, etc. I tried to contact Bryan Litz to get more details on how it works, but he has yet to respond. That might be more details than he is willing to release, which is reasonable … but at the very least it isn’t that there was 2 mph of error on each shot.

      Remember, EVERY one of the variables are generated for each scenario. So if you defined a 0.5 MOA rifle precision, and a +/- 2 mph wind confidence. There might be one scenario where the wind call was dead on, but the shot went 0.2 MOA to the right. Then might be another were the wind was -2 mph causing the shot to go slightly left, but shot also went 0.25 MOA left … compounding the problem even more. Then there was likely another scenario where the wind call was off, but the shot position was off in the opposite direction causing a dead on hit … and we never even knew our wind call wasn’t right. That kind of scenario can be the most confusing, because we feel like we know what the wind is doing at that point … but we don’t.

      And F-class is a different kind of shooting than what I typically do and what most of my readers do. I typically shot in tactical/practical matches and also hunt. Lots of times that means you only get 1 shot at a target … no sighters. You aren’t able to walk the bullet onto your target. Also, you aren’t always shooting in the same direction or environment, which means you aren’t shooting in the same wind. On one stage you might fire west directly into the wind, north with a full crosswind, and northeast with some kind of quartering wind. Then the next stage might be right up next to a cliff that has a severe updraft, and totally changes the wind profile. It’s just very different from laying down in one place, firing most of your shots in one direction, and in relatively stable wind conditions. I’m not saying it’s harder! They’re both challenging and fun … just different. In F-Class shooting, precision might be more important than what the models here show, because you are shooting for score … not hit or miss. So shots centered on the target count for more. This analysis is more for a type of shooting like tactical matches shooting steel targets, or hunters who know they have a 10″ vital area and anywhere in that area results in an ethical and quick kill. The steel doesn’t count more if you hit dead center, just like the animal isn’t more dead if you hit the specific hair you were aiming at. F-Class is a different game.


  8. A WAG based on the intuition of a master sniper’s intuition would probably be at least as precise/realistic as those “models for man-made global warming” have proven.

    I think a really well edited and narrated interactive video on reading mirage, range wind cues and the laminar wind velocities would yield 10X more “gongs” than all the load development / COAL adjustment / barrel harmonic dampener mumbo jumbo.

    Show me some slow motion shots with the sun or a good spot light shining off the boat-tail of a VLD arcing way up and across the relative wind…give me video and doppler plotting from behind, profile and plan views…put them all split screen slo-mo. Repeat them three times telling me what your are gonna tell/show me, tell/show me and then tell me what you told/showed me !

    You sure tore a ragged whole in us equipment freak’s buying performance myths.

    • Well, I might have argued with you a month ago … but after learning all the stuff I’m presenting in this series, I think you’re spot on. Most of my training and time over the next year will be spent trying to get better at calling the wind. Instead of buying competition dies and a Prometheus Gen II scale (which I really was considering), I’m thinking about paying to go to Todd Hodnett’s long-range shooting class so he can help me get better at wind. I feel like spending the money on that would result on more of my bullets connecting with their intended target.

      I will say that Magpul produced a DVD training set featuring Todd Hodnett that is helpful. I’ve waited Bryan Litz’s DVD set, and it’s good too, and also own Rifle Only’s Training DVDs … but I think Magpul’s The Art of the Precision Rifle DVDs was the most helpful for me of all the ones I’ve seen.

      I wish I had the equipment to do those kinds of videos, but I don’t. I really have 1000 ideas for how I could use a high-speed camera if anyone wants to donate one!

      Honestly, this rocked my world as much as anyone. That’s why I felt like I had to share it, even though I fully expect a few people to totally dismiss me on this stuff or even pick up stones. Ultimately, I’m just trying to figure this out, and this was enlightening for me. I’m tickled if others find this stuff helpful too.


  9. Very good read. These numbers are good for know distance shooting. Throw in a elevation error variable and i would think you could argue the fact that a guns accuracy becomes pretty valuable. Just the thoughts of a shooter way less talented and knowledgeable then mr litz.

    • Glad you found this helpful, Nick. There is an aspect of elevation error represented by error in pressure. The air pressure changes based on elevation. A lot of veteran shooters speak in terms of Density Altitude (DA), not just altitude. DA is a combination of a few things. Here is an excerpt from Wikipedia on DA: “Both an increase in temperature, decrease in atmospheric pressure, and, to a much lesser degree, increase in humidity will cause an increase in density altitude. In hot and humid conditions, the density altitude at a particular location may be significantly higher than the true altitude.”

      So each of those variables are represented in this tool: Temperature, Atmospheric Pressure, and Humidity. I left the amount of error for those variables to the default settings, which appear to match the specs on a Kestrel Weather Station. By that I mean the software is using a +/- 1 degree temperature error, because that is what a handheld Kestrel weather meter is capable of doing. The same thing appears to be true for those other variables as well.

      Thanks for the feedback!

  10. Nicely written article and well broken down analytics. This confirms my thoughts precisely on the factors that really make the biggest difference on your target hit percentages. Correctly reading the, (1) wind influence from your muzzle to the target and all distances inbetween; (2) rangeing your target to within a couple of metres, and, (3) having consistant projectile velocities, truely are the main variables you need to get a handle on for more hits on that long range target. Certainly barometric pressure, reletive humidity and temperature play their part, along with spin drift, azimuth & coriolis effect in altering the trajectory/path of your projectile somewhat, however nowhere near major as the factors of the first thre (3). Thank you for your time and dedication to the sport & science of precision shooting.

  11. My old comment is waiting moderation, probably from the childish jokes about size haha, here is the adult portion of it copy pasted.

    But seriously this was a great article diving into several interesting topics, I am so happy to see the Monte Carlo simulation show up! i loved writing super basic ones in MatLab, but I should point out that I have also made some pretty good ones in Excel. If youre interested its super easy.

    Also to add, it should be pointed out that the seemingly exponential decay of returns is actually in part a function of the target. For the sake of simplicity, imagine your shooting distribution is a square that just barely fit the circle. (that is the target circle is inscribed in the square). Right now all of the edges of your square are misses, huge wasted area. if you squished the square so that it was half as tall (same width, still centered) you reduce the missed areas by a lot! but if you fold it in half again, because the circle is nearly vertical here, you dont gain as much.

    This would be directly juxtaposed by a rectangular target. imagine a small square (the target) in a large square (the shooting distribution). This time, as you reduce the distribution vertically the removed loss area is actually linear until its shorter than the smaller square.

    I hope that made sense, I can make a pictured version and actually run the math behind it if you want.

    one last thing, this post is exactly why i believe groups are dumb. a gun that shoots 1MOA every other group and a gun that shoots 1 MOA every 100 groups are both technically “MOA capable guns”. I think we as long range shooters should be using standard deviation and pitch guns as “The guns sigma is MOA”.

    • Sorry, Adith … I think I was asleep when you made the comment last night, so I’m working through them and replying this morning. This is still just a hobby I do in my free time, so I can’t always be as responsive as I’d like to be. I do appreciate your feedback.


      • oh not it wasn’t a complaint! I assumed my word choice got it caught in the spam filter. I mean you put so much money into and produce some of the best articles out here. It would be pretty awful of me to bitch that you don’t monitor the comments 24/7.

  12. Cal – agree with the others above. Another great post. I have the books and software and have been playing around with the variables – and it does indeed show that wind and environmental factors seem to play the biggest role in accuracy. It’s nearly impossible to rule out human factors, but the modeling also reflects what I’m seeing in the field – with a low ES and SD on the muzzle velocity, nearly all misses are due to bad reads on the wind. Keep up the great work and looking forward to the next few posts.

    • Thanks, Jeff. Good to know this stuff aligns with your experience in the field. Thanks for taking the time to share that. – Cal

  13. Cal, my advice to you and other shooters who want to better your performance and wind skills is to shoot some f-class matches, your 6 or 6.5 shot on a bipod in f-open is fine, you may have to remove your brake. The point is every shot you get instant feed back in the form a score and a spotter disc in your previous bullet hole. Now some might argue the value of this but I will tell you I came into PRS with 4 years of f-open under my belt and what I learned about reading wind off of mirage, flags and foliage all came from f-class. It will train you to look at all indicators and that will translate to you using the indicators on a tac type event. Dust blowing from impacts or misses, grass moving on the burm, mirage from you to target. All things that you will program your mind to notice during an actual event. Other things you will learn is how mirage and specifically a boil will affect your POI, how cloud cover vs intermittent clouds to sun affect your POI, and how a rt or a left wind can affect your elevation. Not to mention at every match you will spend equal time shooting, spotting, and pit service. The time spotting for someone else is as valuable or more so than your actual shoot time.

    • Great tip. I’ve only shot on the 1000 yard range at Whittington one time, but getting that instant feedback from the guy in the pit was really helpful … and a lot of fun. I just might try to find a place to do that. I go to Midland all the time, and I think they have a big F-class crowd. I bet I could learn something from those guys.


  14. precisionaccuracyresearch

    This is awesome!

  15. Cal, you really are producing some of the best articles on the interwebs about long range shooting. There is a difference in shooting F Class where the smaller the group in the center of the target wins and Tactical/Hunting type scenarios, but BOTH significantly improve your skill levels in different ways. I’ve struggled convincing new shooters that the way to get better is to shoot more, and to shoot under adverse conditions. Even good/great shooters don’t like their scores when they shoot under really tough conditions. Ask any F Class shooter who has shot at Rattlesnake in Richland, Washington about shooting in the wind there! (From what I’ve heard, first F Class match there lots of big names showed up, getting them to come back has been next to impossible!)

    While shooting with the almost instantaneous feedback allowed using markers in an F Class match does help, shooting steel under tough weather conditions where one shot is all you get to score presents a bigger challenge at times.

    Regardless, shooting more often, and under tougher conditions makes you a better shooter. Spending countless hours in the reloading room looking for the perfect load is unlikely to do that!

    For those not willing to spend $200 for Bryan Litz’s program, here’s a much cheaper option that allows you to do some of the same things:

    • Thanks, Bill. Couldn’t agree more. Out here in West Texas the wind is always blowing! If it’s blowing 5-10 mph, we call that no wind. It’s tough to make yourself go out to the range to practice on those 15-30 mph days, but I’ve started doing it. You don’t get to pick which day the match is on, and it might be one of those days. Plus it makes you better. The worse is when that is a 25 mph wind that is shifting from 11 to 1 o’clock or 5 to 6 o’clock. Man, you feel like a noob on those days! If you don’t watch careful, you’ll easily be off target 6 feet! But, you learn a lot on those days … Even though it feels like a beating at the time.

      Thanks again,

  16. Hi Cal,

    You have some sharp readers. I’m really impressed by the insight in the comments.

    Just too piggy back on my original thoughts, the distribution used to describe each variable in the equation is key to this type of model.  When you look at the wind for instance, your 2 MPH simulation has a wind call within 2 MPH every time, for 1,000 simulations, and those calls, follow a standard distribution (looks like it from the plot, correct me if I am wrong).  In fact, if you plot the area that the shots could fall on a set of axis, you have 100% of the possible outcomes in a very refined space (and my guess is the ratio of hit to miss area changes in the same proportion as the results of the simulation). How accurately does this reflect real life (which is what you are supposed to be simulating)?  I’d argue , using admittedly anecdotal experience, it doesn’t.  Again, assuming a standard/normal distribution was used, it is the most abused pdf in statistics. Gathering an actual data set from wind calls (which would be incredibly difficult to do objectively) and determining which distribution function those calls follow, would be an improvement.  The same would need to be done for every variable addressed in the model. 

     I would bet that if I made a scale model of the 2MOA circle on a piece of construction paper and masked off the area outside of the possible impact areas on the target and surrounding, that I could hand my 4 year old a crayon, have him mark try to mark the target 1,000 times, and have a similar result.  I could change the size of the masked area for height, representing the 1, .5 MOA, etc… sub tension and produce a similar finding.

    Further, I would guess that if you run the same simulation with a two MOA box, that your results for the 1-.5-.1 MOA won’t be that different due to the shape (hit to miss area ratio) of the target (I’d also run it  more than 1,000 times).  So in this case, the shape of the target at a given size would matter more than the rifle. 

    A better title for the simulation would be, “How much does group size matter when shooting a 2MOA circle target while being able to hold a wind call within 2MPH 100% of the time”, or, “Dude, you are over thinking it”.

    Thanks again for your hard work and effort


    • Yeah, I’ve got a lot of sharp readers! Honestly, it’s more than a little intimidating to write posts at this point. When I know that I have PhD level statisticians, ballisticians, and every kind of engineer reading my posts, plus some of the best shooters in the world … yeah, I better have my stuff together. But, that’s also what makes the content better. The pressure to get my crap together and check and double-check what I present and how I say it is healthy. If everyone did that online, there would be far less misinformation out there. And I’ve also enabled comments to let my sharp readers correct me when I misspeak. While most people don’t like being corrected, I’m a natural learner so it’s a little easier for me. I just want to get to the truth, and if I’m not understanding something correctly, then I just want to know what the right way is.

      Now to your comments. Unfortunately, I don’t know if the software uses a standard distribution or something that reflects a real-world data set. It could work either way. The IT team I lead (in an unrelated industry) has designed Monte Carlo simulations that work both ways, based on a normal distribution and based on real, historical data points from the field. Honestly, it’s not a lot harder to do one or the other. I tried to reach out to Bryan to get more details on how this works behind the scenes, but have yet to hear back from him on that. It could work like you said, but it also might not. We’d be making assumptions either way. I’d honestly be surprised if it made a significant difference. I doubt it would invalidate the point I’m making in this article regardless of that one implementation detail.

      I will say the shape and size of the target matters. Bryan makes that point time and time again in his book. I tried to pick targets that roughly represented a typical tactical target, if there is a such a thing as “typical tactical target.” If you change the size and shape of the target, the results can vary significantly. For example, if you are shooting at an IPSC silhouette target, the SD matters even less because the target is really tall and very forgiving in terms of vertical dispersion.

      And yes … We’re probably overthinking it at this point! I’m just trying to give people a better perspective of how all this stuff plays into hitting steel targets a long way off. Hopefully I haven’t muddied the waters.

      Bill, thanks, as always, for your thoughtful feedback.


  17. I’m just glad the hundreds of hours I spent working on modeling finally have a purpose! I never got to really use it outside of my graduate work.

    Keep up the great work!

  18. I’ve been using AB for a while now, and it’s really highlighted what I should be focusing on for long range shooting.

    Be careful on your selection of variables, as it skews your data significantly.
    You’ve isolated wind and rifle performance in the above, but largely left out range and MV. If you redo your study, and set wind to 2 mph (elite shooter), MV to 15fps (way better than factory ammo) and range to 1% of distance (as per your rangefinder accuracy post) (10m @ 1000m), you’re going to get a very different picture of the inherent rifle precision. (i.e. there is going to be substantially less deviation as MOA increases/decreases at 1000m). It’s even more so with less ‘elite’ values. So for the average shooter; 3-5mph, 25-50fps and 10-25m could be more relevant, and then the picture completely changes.
    A more interesting study would be to take the same inputs (with 0.25, 0.5, 1, 1.5, 2 MOA) and run them against different distances from 100 to 1000, which would then better indicate the transition point from the rifle’s accuracy to the external variables of Wind, MV, and Range.
    This would then indicate to what distance rifle accuracy really matters.

    Keep having fun with AB, it’s a GREAT piece of software.

    • Great, comments. And yeah, I agree completely about the AB software. It is a lot of fun to be able to just play around with the variables and explore how tweaking different variables changes the dispersion. It has really been an education process that can’t fully be duplicated in posts, no matter how much time I spend thinking about it or designing graphics! I’d highly recommend it to any serious shooter. Yes, it’s $200 … but man I’ve learned a lot. I’d already read the book twice, and I’m still getting a ton out of the software. It’s incredibly empowering to be able to do ad hoc analysis on your specific rifle, ammo, ability, and scenarios.


  19. This is such a timely article, weird really, how often that happens. Just the other day, I was discussing with a shooting buddy what was more important, group size or velocity. What would you rather have, a 1/4 minute gun shooting 2500 or a 1/2 minute gun shooting 3000? With everyone so obsessed with pushing bullets as fast as they can, almost to the point of being dangerous, does that extra speed really matter. I am mainly talking about F-TR competition here, and the limitation of using the 308 case, but it applies equally everywhere. What I found was quite enlightening, and right along the lines of what you have written about here. For example, a 185 Juggernaut at 2650 moves 8.1 inches at 1000 yards for each one(1) mph of crosswind. The same bullet at 2850 gets blown 7.2 inches. So for all that extra pressure that you need to gain 200 fps, you will see less than 1 inch of difference on the target. What is much more important to realize here is that if you misjudge the wind by only 1 mph, both bullets will be blown out into the 9 ring, regardless of the speed they are started at. The fact that making the correct wind call trumps everything else at long range was made painfully clear by this experiment. Now here is something else fun to think about. You said in the second to the last paragraph “Of course, tighter groups are always better”. This is not necessarily true, because if you misread the wind by 1 mph with a 0.1 moa gun you will be 100% guaranteed to shoot a 9 (no part of the group can touch the 10 ring). But, if you misread the wind by 1 mph with a 1 moa gun, you still have about a 15% chance of shooting a 10 (the left edge of your group that got blown to the right can still catch the 10 ring). Not that I would ever recommend doing this, but it could be argued that a gun that shoots too good could be a disadvantage. Kinda like trying to shoot skeet with a turkey choke, if you know what I mean. I guess the point is, as your article explains, once your gun is shooting groups small enough to hit the target, it is up to you to get them there by making the correct wind calls. Great article, and follow up comments from your readers too. Thanks again for your effort in putting this blog together.

  20. Reblogged this on rifletalk and commented:
    This is the kind of detailed analysis that Cal’s Precision Rifle Blog simply excels at. Really challenges our view of matters we considered to be truisms in our sport. An excellent read.

  21. Great information here. I’m very appreciative of the data-driven approach you take both in testing and analysis. It’s very refreshing to find objective information in a world that is ruled by opinions stated as fact.

    Although I have been shooting since I was a child, I am fairly novice with long range. My question is; with respect to achieving the best groups possible, is one better off to persue hand loading or a precision rifle? Does one form of improvement have more profound impact on group size? I ask as I am limited by my budget. As I sit right now I can upgrade my scope or my rifle (rem 700 sps with vortex viper), but not both. I already have hand loading supplies and have been working with loads in trying to extend my range. So I guess my real question is how much benefit is there to be gained by improving my rifle? Thanks for all your hard work and keep it up.

    • Travis, those are great questions! I’m afraid there isn’t a clear answer, although everyone has their opinions. It sounds like you have a great scope/rifle setup to get started, but I’m thinking you’re probably to the point that you want to take it to the next level. If it were me, I’d upgrade the rifle first. You can go a couple different routes. You could sell that rifle and start fresh … or you could stick with your Remington 700 rifle as the foundation and build around it. If you went that route, I’d probably follow the same steps Todd Hodnett suggests in The Art of the Precision Rifle DVD Set (which is a great place to start and a great training DVD for long-range shooting). I’m pretty sure these are the steps he laid out, or at least very close.

      1. First, I’d replace the barrel with a match grade barrel. That is your biggest bang for your buck, and it will probably cost $650-750 for the barrel and gunsmithing work to install it. But, you’ve just replaced the biggest component that affects accuracy. If you can keep the stock from touching the barrel, I’d say you will likely see a substantial difference in accuracy at this point.
      2. Next, I’d replace the trigger or at least get it adjusted down to something less than what a Remington comes standard. Many veteran shooters run their triggers from 1.5-3.0 pounds. Most stock Remington triggers are closer to 6-8 pounds, and that can cause you to pull your shot off target. A Timney 510 trigger costs $140, or you can adjust your Remington 700. If you’re not comfortable doing that, I’d recommend paying a gunsmith to do that for you. The trigger is probably not somewhere you want to DIY … it is what controls the firing mechanism, so it can be dangerous if adjusted improperly.
      3. Next, I’d replace the stock with either a conventional stock by Manners or McMillan, or a chassis like those made by Accuracy International and others. Those options start around $700 for an adjustable tactical stock and go up to $1500 or more. If that’s too steep a $450 Bell & Carlson Medalist seems to be a budget choice and lots of guys say they’re happy with the performance they get out of it. HS Precision also has some stocks in that price range between Bell & Carlson and Manners and McMillan. I personally like an adjustable stock (length of pull and cheek), and would rather have that even if I had to back down from one of the bigger brand names. I have a Manners, McMillan, and AI … and all are outstanding products. You can see what stocks are popular in precision rifles in this post.
      4. Next, you can get a gunsmith to accurize your Remington 700 action, which means he does a bunch of operations to square everything up and bring a mass-produced action into tighter tolerances for increase repeatability. This is also refered to as “blueprinting” the action or receiver. It seems like the cost of this varies, but it looks like Score High Gunsmithing is advertising $695 for this.

      After all that stuff, you have a custom rifle. The cool part is you can do it over time. #1 and #2 are likely to yield the biggest difference, but the other steps will incrementally improve your precision as well.

      I’ve been thinking a lot about guys in your shoes. Next year I might do a couple different rifle builds to see which one yields the best rifle and scope setup for the lowest cost. I know getting into this can be a little cost prohibitive for some people, but it sounds like you have a good basis to start with. The Vortex Viper isn’t the best scope, but it’s also not the worse. The Remington 700 action can be a great foundation for a precision rifle build.

      There are other options of rifles you can buy for sub-MOA accuracy. Here is an article that talks about some of them: http://www.gunsamerica.com/blog/minute-of-angle-moa-accuracy-out-of-the-box/. That is similar to the test I’d like to do at some point. I’m sure there are others out there that are similar. Here is one more article I’ve read that might also help: http://www.outdoorlife.com/blogs/gun-shots/2013/05/new-rifles-best-rifle-rifles-review-rifles-test-gun-test-2013.

      You could also go with what I call a semi-custom rifle like those offered by HS Precision, FN, and Kimber … or you could obviously go with a full-blown custom rifle, and pick every component. Those usually start around $3500. There are a lot of good gunsmiths specializing in precision rifles that you could pick from to help you with that.

      I know that’s a lot of information. At some point I hope to be able to provide better answers to that question backed up by data and a few different configurations I’ve personally tried. Until then … I’m afraid this may be the best I’ve got. Sorry I couldn’t be more help!


      • Thank you very much for the quick and quality level response. I stumbled across your site months ago when I decided to improve my skills/equipment and I must say, reading all of your posts has been a fantastic help in my planning. After reading and understanding your approach to various aspects of shooting, I would also like to say that I respect and value your opinion as well. The viper I have seems to work pretty well, I haven’t done a box test but it hasn’t ever not returned to zero after dialing elevation. I think I might go the route of a build, the main problem I’ve been noticing is a large variance in the vertical dispersion of my shots at 500+ ranges. I’m sure that some (most) of it is my abilities, but I would like to take any accuracy variables out of the equation. Now I just have to decide between .243 and 7mm rem mag. Thanks again for your time and willingness to help others.

      • You bet! Both great cartridge choices. Can’t go wrong!


      • Hey Cal, Surely if you’re shooting past 400/500 meters, the precision of the rifle is less influential than producing highly consistent top quality ammo, being able to judge the wind as tightly as possible and getting the best range finder that you can afford.
        I would perhaps think along:
        1. Upgrade press, case prep and powder tools.
        2. Upgrade scope to decent FFP (at least a Burris XTRii 4-20 SCR reticle)
        3. Get excellent Rangefinder (see Cal’s excellent earlier rangefinder post)
        4. Get a Kestrell 4500 AB, Bluetooth and Shoot a LOT in wind.
        5. Upgrade rifle (6mm or 6.5mm), and with the Rem 700 action, you could do that bit by bit if budget is not available all up front.

      • Your definitely entitled to your opinion. That’s not the path I’d choose, but I also won’t say it’s wrong. I like 4 & 5, but 1 & 2 seem overstated … At least in my experience.


  22. But a big difference in reloading cost, with the 7 Rem using more powder and more expensive bullets. The .243 you can shoot all day because of the low recoil, while 50-100 rounds will beat up most shooters of 7 Rem Mags. You will get better faster with the .243! You can shoot a LOT more and shooting more equates to more chance to improve!

    • Just depends how you plan to use the rifle. I actually own both. I have a custom 7mm Rem Mag that I use for hunting and long-range target shooting out to 1600 yards (actually have hits out to 1 mile but that’s pushing it). I also have a custom 6XC (identical external ballistics of the 243 with 7 grains less powder) that I use for target shooting out to 1,000-1,200 yards. Both are outstanding cartridges, and great within their intended uses. I used the 7mm Rem Mag in precision rifle competitions for 2 years (before I built the 6XC), and didn’t feel like it handicapped significantly. You do need a good muzzle brake and need to shoot it a lot. I’m about to publish some data on muzzle brakes, and I have found some that reduce the recoil of a 7mm Rem Mag down to the level of my 6XC. Most can’t do that, but there is at least one that can and I have hard data to prove it!