Optimized .357 barrel length?

[GypsmJim]

I have been handgun hunting for deer for over 40 years. Ever since the state made it legal. I started with a Blackhhawk .45 Colt with a 7 1/2" barrel.

I chose that caliber because of extensive reading - most of the pundits said a .357 was not adequate unless in the hands of an expert.

Prior to that shotguns were the only option during the regular season. My personal practice was 50 yards or less. I never had a deer that didn't drop in his tracks.

My first handgun shot was at 50 yards. It was a heart shot, but the deer ran quite a ways, thus tracking was required. After that I limited my shots to 25 yards. No problem with that.

Now I use an Encore with a 15" barrel in .308 Winchester. I'm back out to 50 yards again, with no problem.

I don't hunt with a Ransom Rest, nor am I able to rest on a tree. Shots come quick with no time to prepare. Just my opinion, but .357 / 100 would not be in my future.

 

[Mark8]

Found a couple of 50 yard targets. Can't lay my hands on the 100 yard, but do remember the 158 grain being better than the 185 in these targets. If I can find it, will post it.

Thanks for looking for and posting the targets, very much appreciated!

Obviously didn't shoot as good on both targets, but mathematically, one is 1.882 moa and other is 3.7. Posted both so you wouldn't think I was making it up. Bet there are others on here that have shot better groups than this with a revolver.

Shooting a good 3-shot group at 50 yards every once in a while is not the same thing as consistently shooting 2" full groups at 100 yards. However, that is not important, your shooting is great, and you are right, neither do I possess a .357 Mag revolver capable of consistently doing 2" full groups at 100 yards nor could I shoot such groups if I had a revolver that could.

If you happen to find a couple of those 5-shot or 6-shot 2" groups at 100 yards, please post them. We all love revolvers, I'm sure I'm not the only one who would be tempted to put up a printout at the local range, you know, just to tease the semi guys and gals.

 

[rosewood]

Thanks for looking for and posting the targets, very much appreciated!



Shooting a good 3-shot group at 50 yards every once in a while is not the same thing as consistently shooting 2" full groups at 100 yards. However, that is not important, your shooting is great, and you are right, neither do I possess a .357 Mag revolver capable of consistently doing 2" full groups at 100 yards nor could I shoot such groups if I had a revolver that could.

If you happen to find a couple of those 5-shot or 6-shot 2" groups at 100 yards, please post them. We all love revolvers, I'm sure I'm not the only one who would be tempted to put up a printout at the local range, you know, just to tease the semi guys and gals.

If I an find them ,I will and I may be inspired to take it out and make some more targets. Will have to swipe a scope off one of the contender or encore barrels to put on the Taurus. As for the contender and Encore, my 14" 7-30 waters has turned in some 1/4" groups and the 12" 44mag Encore has done 1.25" groups at 100 yards. Yeah, I know they are short barrel rifles virtually, but they are still considered handguns. The shorter barrels in the Contender and Encore many times shoot better than the rifle length barrels do. The weak link is usually the trigger puller.

I can say, I have never shot a group as bad as 6.5" with this gun. It will consistently shoot way better than that.

Rosewood

 

[Swiftflyer]

I beg to differ, a much younger man then with much younger eyes but with a decent rest that 4" Colt would put all 6 rounds in a paper plate at 100 yards. A little luck perhaps and the cat was broadside and motionless.

 

[Abbynormal]

I've yet to shoot my 6" M27-2 since getting cataract surgery on both eyes. I've been doing 1 gallon milk jug lifting arm exercises and a piece of wood and rope to exercise the wrists.
I shot 6" and less back in the day at 100 yds and now have more accurate loads for the .357 to try.

 

[BB57]

With all due respect, the very best 6" competition revolvers in .357 Mag using the best precision cartridge selected for the individual revolver will do 6.5" groups at 100 yards from a Ransom rest under ideal conditions. No handgun can do 2" groups at 100 yards.
​

The AR-15 pistol in 9mm Luger at the bottom has a 8.3" ballistic advantage barrel and shoots 1 MOA out to 200 yards.

This MP5 pistol is 2 MOA accurate and will do 2" groups at 100 yards.

This 8" plate is out in the beans at 150 yards. Not a bad group for a "handgun"

"Never" is seldom as absolute as it sounds.

But I know braced pistols were not what you meant.

——

However, in terms of practical revolver accuracy, I routinely shoot 6" plates at 50 yards and go six for six with my 6" Model 19 and 6" Model 66 shooting off hand and generally stay away from the edges of the plate.

At 109 yards there are a pair of 8" x 18" targets at each end of the range. They are the bottom 18" of welding tanks mounted upside down on steel posts. I consistently score 5 out of 6 on them offhand and about three quarters of the time go 6 for 6.

A well made 6" .357 magnum revolver is certainly capable of hitting a 10" circle at 100 yards, which is the size of the vital area on a small deer, provided the shooter is proficient. If the shooter can consistently (a PK of 90%) hit a 9" paper plate at 100 yards offhand he can hit the vital zone on a deer.

——

My objection to hunting with a .357 Magnum handgun at 100 yards is the lack of velocity and penetration.

A 6" .357 magnum will launch a 158 gr bullet at 1300 fps or a bit more and generate just over 600 ft pounds of energy at the muzzle. At 50 yards the velocity and energy will be down around 1200 fps and 500 ft pounds and 1200 fps is already marginal at best for most JSP .357" bullets.

At 100 yards it's down to about 1100 fps and 400 ft pounds. But more importantly in practice it's ability to consistency anchor deer sized game is extremely inconsistent.


On the other hand a 20" short rifle or 24" rifle in .357 Magnum will launch the same 158 gr JSP bullet at around 1820 fps. That generates almost 1200 ft pounds of energy at the muzzle. In fact in a rifle it produces the same 1300 fps and 600 or so pounds of energy as a 6" revolver at the muzzle, but does it at 160 yards.

At 100 yards that rifle launched 158 gr bullet is traveling at just under 1500 fps with about 780 ft pounds of energy.

Both of my tang sight equipped .357 Magnum Rossi 92 .357 Magnum lever actions will shoot 2 MOA 5 shot groups at 100 yards with Federal 158 gr JSP ammo.

If I were limited to a straight wall pistol cartridge, the .357 Magnum in a rifle isn't a bad choice and it's consistently effective at anchoring a deer out to 125 yards.

 

[BC38]

According to the Ballistics By the Inch website, most 357 magnum rounds they tested hit their max velocity out of a barrel right around 18" long +/- an inch. So for a handgun, the longest barrel you can get your hands on is going to provide the best velocity and flattest ballistic trajecgory.

 

[BB57]

According to the Ballistics By the Inch website, most 357 magnum rounds they tested hit their max velocity out of a barrel right around 18" long +/- an inch. So for a handgun, the longest barrel you can get your hands on is going to provide the best velocity and flattest ballistic trajecgory.

In general you are correct. Longer means more velocity in a .357 magnum handgun.

But "hitting their max velocity right around 18" lomg +/- an inch" isn't quite how it works.

The folks over at ballistics by the inch take a barrel and cut it back an inch at a time on a chop saw.

They also test several brands and weights of bullets but they only fire three rounds of each type at each barrel length.

It's hard to come up with something worse than a three round sample size for determining velocity but by golly they found a way to do it. Those three rounds over two chronographs set back to back to generate 6 data points from those three bullets. The variation in each round is the same, they just measure it twice and add in two different sets of random instrumentation error.

It has a negative effect on the data.


Let's look at the Federal 158 gr JSP data from 18" down to 13". Again it's just three shots over two chronographs (first and second columns.

I have added the standard deviation (SD) as BBTI doesn't include it. SD is important as 67% of rounds fired will fall within +/- 1 SD of the average velocity 95% will fall within +/- 2 SD and 99.5% will fall within +/- 3 SD. For practical purposes +/- 3 SD is equal to the expected extreme spread in velocity.

So, with the SD of 30 for the 18" data 67% of the bullets in a box of 50 will have velocities between 1739.5 fps and 1698.5 fps. The extreme spread in the box will be a max of 1780.5 and 1657.5.

Now there are two problems with using SD with a sample of 3 (or 6). Random chance affects the shots fired. In some cases you get lucky and all three shots in the sample are within 1 SD and you get a low SD (like the 13" data) However with three shots the odds are you will also get at least one 2 SD shot, and a larger SD. And if you are as lucky, you will get a couple 2 SD shots and a larger SD. Which is to say, the SD with a small sample tends to be larger than it probably really is.

Ideally you'll shoot three 10 shot groups and include all 30 rounds for the SD calculation. That will give you move very accurate average velocity numbers and an accurate SD.

In the BBTI data the three shot sample also means the velocity isn't very accurate as any one round being in the second or third SD will skew the average velocity significantly higher or lower.

I've added the differences in velocity from bottom to top between each inch of barrel difference. You can see adjacent barrel lengths with differences of -24 to +58, and yet a total difference of only +45 from 13" to 18".

18"
1698 1692
1742 1726
1738 1720
Average = 1719
SD = 20.5
————————— (+30)
17"
1692 1664
1696 1675
1694 1710
Average = 1689
SD =16.4
————————— (-50)
16"
1709 1741
1732 1763
1730 1756
Average = 1739
SD = 19.5
————————— (+31)
15"
1679 1691
1715 1718
1717 1728
Average = 1708
SD = 18.8
————————— (- 24)
14"
1706 1714
1721 1729
1755 1765
Average = 1732
SD = 23.4
————————— (+58)
13"
1672 1677
1668 1668
1680 1678
Average = 1674
SD = 5.2


What it tells me is that it's junk data. Now, if you look across all the .357 data, you might see that in most barrel lengths the highest velocities occur at 16". But the differences between the 16" and 18" velocities are generally less than the SD of those velocities, meaning the differences between velocities are actually less than the shot to shot variation at a given barrel length.

——

In practice I have a Rossi 92 20" short rifle and a Rossi 92 24" rifle that are identical except for barrel length. Based on a 30 round sample, I get an average velocity of 1805 fps with the 20" barrel and 1820 fps with the 24" barrel.

Thats not a big change with 4" more barrel, but it's still an increase. There is still more pressure behind the bullet than in front of it and that's the case until length is well over a yard.

Bullets reach a point where the gain in velocity per inch gets progressively lower, but the extra barrel doesn't cause them to slow down.

 

[Nevada Ed]

A longer barrel length will give you a longer sight plane which will increase accuracy potential. Back in the day when I had good eyes, we used to shoot bowling pins at 50 yards off hand. At 100 yards, if I hit 3 or 4 out of 6, I considered it a good day. Great fun with targets. Game however, is not always standing still or broadside.

My M19-5 with a 158 Lswc factory magnum load would do very well
at 100 yards on steel pigs, at our range from a bench table and no bags, or two hands on a no wind day.
This however was, 34 years ago, with a 6" tube.

 

[BC38]

In general you are correct. Longer means more velocity in a .357 magnum handgun.

But "hitting their max velocity right around 18" lomg +/- an inch" isn't quite how it works.

The folks over at ballistics by the inch take a barrel and cut it back an inch at a time on a chop saw.

They also test several brands and weights of bullets but they only fire three rounds of each type at each barrel length.

It's hard to come up with something worse than a three round sample size for determining velocity but by golly they found a way to do it. Those three rounds over two chronographs set back to back to generate 6 data points from those three bullets. The variation in each round is the same, they just measure it twice and add in two different sets of random instrumentation error.

It has a negative effect on the data.


Let's look at the Federal 158 gr JSP data from 18" down to 13". Again it's just three shots over two chronographs (first and second columns.

I have added the standard deviation (SD) as BBTI doesn't include it. SD is important as 67% of rounds fired will fall within +/- 1 SD of the average velocity 95% will fall within +/- 2 SD and 99.5% will fall within +/- 3 SD. For practical purposes +/- 3 SD is equal to the expected extreme spread in velocity.

So, with the SD of 30 for the 18" data 67% of the bullets in a box of 50 will have velocities between 1739.5 fps and 1698.5 fps. The extreme spread in the box will be a max of 1780.5 and 1657.5.

Now there are two problems with using SD with a sample of 3 (or 6). Random chance affects the shots fired. In some cases you get lucky and all three shots in the sample are within 1 SD and you get a low SD (like the 13" data) However with three shots the odds are you will also get at least one 2 SD shot, and a larger SD. And if you are as lucky, you will get a couple 2 SD shots and a larger SD. Which is to say, the SD with a small sample tends to be larger than it probably really is.

Ideally you'll shoot three 10 shot groups and include all 30 rounds for the SD calculation. That will give you move very accurate average velocity numbers and an accurate SD.

In the BBTI data the three shot sample also means the velocity isn't very accurate as any one round being in the second or third SD will skew the average velocity significantly higher or lower.

I've added the differences in velocity from bottom to top between each inch of barrel difference. You can see adjacent barrel lengths with differences of -24 to +58, and yet a total difference of only +45 from 13" to 18".

18"
1698 1692
1742 1726
1738 1720
Average = 1719
SD = 20.5
————————— (+30)
17"
1692 1664
1696 1675
1694 1710
Average = 1689
SD =16.4
————————— (-50)
16"
1709 1741
1732 1763
1730 1756
Average = 1739
SD = 19.5
————————— (+31)
15"
1679 1691
1715 1718
1717 1728
Average = 1708
SD = 18.8
————————— (- 24)
14"
1706 1714
1721 1729
1755 1765
Average = 1732
SD = 23.4
————————— (+58)
13"
1672 1677
1668 1668
1680 1678
Average = 1674
SD = 5.2


What it tells me is that it's junk data. Now, if you look across all the .357 data, you might see that in most barrel lengths the highest velocities occur at 16". But the differences between the 16" and 18" velocities are generally less than the SD of those velocities, meaning the differences between velocities are actually less than the shot to shot variation at a given barrel length.

——

In practice I have a Rossi 92 20" short rifle and a Rossi 92 24" rifle that are identical except for barrel length. Based on a 30 round sample, I get an average velocity of 1805 fps with the 20" barrel and 1820 fps with the 24" barrel.

Thats not a big change with 4" more barrel, but it's still an increase. There is still more pressure behind the bullet than in front of it and that's the case until length is well over a yard.

Bullets reach a point where the gain in velocity per inch gets progressively lower, but the extra barrel doesn't cause them to slow down.

Ummm, OK.

That's a very interesting analysis of their data. Just curious, but where did you come up with such minute details of the BBI testing methodology? I may have just missed it, but I didn't see an explanation of their testing process that described it to that level of detail. But that isn't to say I may not have just missed something.

FWIW, it is always easier to take someone else's test results and crunch the numbers to create an analysis that "throws rocks" at the other guy's conclusions. Especially if you don't have to produce any data of your own to refute the conclusions from their data. Just an observation...

From what I have read of other people's analysis of the data from BBtI, it seemed to be making the case that somewhere above an 18" barrel length, the friction of the bullet in the barrel starts to overcome the pressure behind the bullet that was causing it to accelerate.

This makes sense considering the fact that at some theoretical point, the amount of volume inside the barrel behind the bullet that is containing the pressure of the powder charge that is causing the bullet to accelerate becomes large enough that the pressure drops enough for the bullet to stop accelerating.

At that "break even" point the velocity would logically start to decrease rather than continuing to increase if there is still more barrel for it to travel through, because the force of friction between the bullet and the barrel would be greater than the force of the pressure behind the bullet.

At least that was the theory behind the analysis I read regarding why the velocity started to drop off after around 18" barrel length. I'm no mathematician, but logically speaking, that theory seems to make sense to me from a physics standpoint.

The exact length of barrel where friction and pressure balance out would logically vary, based on numerous variables. Things like the coefficient of friction of the bullet (soft lead, vs hard cast, vs copper plated, vs powder-coated, vs jacketed) and the quantity of the powder charge, and of course the burn rate of the powder being used, would all factor into that result.

You seem to be saying that it happens at somewhere around a 36" (a yard long) barrel. The folks at BBtI who did the actual testing seemed to think that it was somewhere around half of that - at around an 18" long barrel.

If you want to do the testing and publish the kind of comprehensive data that they have, as opposed to just doing a critical numerical analysis of THEIR data, then that would be very useful. Barring that, I think I'll go with their conclusions. No offense.

FWIW, based on their published data, I already actually have gone with their data and chose 20" long barrels as the best compromise between performance vs easy handling & maneuverability when I was purchasing my 357 mag and 44 mag lever guns.

 

[jem102]

I don't remember the actual center area of the IHMSA turkey at 150 yards or the ram at 200 yards but I witnessed a lot of 40 X 40's with wheel guns during the years I shot the game. All done with factory Ruger's, S&W's and Dan Wesson's. .357, .375, .41 and .44. loaded max and maybe a pinch more... All those would have resulted in a down bird or whitetail. IMO

 

[Jim Watson]

Since a longer barrel gives higher velocity and a longer iron sight radius, then you need an 8 3/4" for hunting.
But a 6 1/2" is a lot more portable.
Elmer Keith wanted nothing over 7 1/2".

We shot IHMSA here for a few seasons. I rang a couple of hard set rams with the .357 that I had had an 8 3/8" barrel fitted to, so I went to a 6 1/2" .44. I now think I'd have been better told to work up a heavy bullet load for the .357.

 

[Bald1]

With all due respect, the very best 6" competition revolvers in .357 Mag using the best precision cartridge selected for the individual revolver will do 6.5" groups at 100 yards from a Ransom rest under ideal conditions. No handgun can do 2" groups at 100 yards.
​

I was open to your argument until you said "no handgun". Like Gypsum Jim I hunt with an Encore. Mine is a .243 with a T/C scope. I have killed woodchucks at 300 yds from the bed of a truck on West Virginia dairy farms. At 100 yds an unobstructed shot on a Whitetail is a lay up. I had a 6" 686 that my soon now owns that I would shoot at 100yds all the time. 6 for 6 on 9" steel was routine. There's many factors and I'm sure there are people reading this that can shoot circles around me. Therefore it's probably best to not speak in such absolutes.

 

[Bald1]

This video just came out this week regarding .357 barrel lengths.

The RIGHT Barrel Length - 357 Magnum Lever Action - YouTube

Fun video. Velocity results were interesting. I'm waiting for someone to get upset about no "obvious " backstop on the range.

 

[BB57]

Ummm, OK.

That's a very interesting analysis of their data. Just curious, but where did you come up with such minute details of the BBI testing methodology? I may have just missed it, but I didn't see an explanation of their testing process that described it to that level of detail. But that isn't to say I may not have just missed something.

FWIW, it is always easier to take someone else's test results and crunch the numbers to create an analysis that "throws rocks" at the other guy's conclusions. Especially if you don't have to produce any data of your own to refute the conclusions from their data. Just an observation...

From what I have read of other people's analysis of the data from BBtI, it seemed to be making the case that somewhere above an 18" barrel length, the friction of the bullet in the barrel starts to overcome the pressure behind the bullet that was causing it to accelerate.

This makes sense considering the fact that at some theoretical point, the amount of volume inside the barrel behind the bullet that is containing the pressure of the powder charge that is causing the bullet to accelerate becomes large enough that the pressure drops enough for the bullet to stop accelerating.

At that "break even" point the velocity would logically start to decrease rather than continuing to increase if there is still more barrel for it to travel through, because the force of friction between the bullet and the barrel would be greater than the force of the pressure behind the bullet.

At least that was the theory behind the analysis I read regarding why the velocity started to drop off after around 18" barrel length. I'm no mathematician, but logically speaking, that theory seems to make sense to me from a physics standpoint.

The exact length of barrel where friction and pressure balance out would logically vary, based on numerous variables. Things like the coefficient of friction of the bullet (soft lead, vs hard cast, vs copper plated, vs powder-coated, vs jacketed) and the quantity of the powder charge, and of course the burn rate of the powder being used, would all factor into that result.

You seem to be saying that it happens at somewhere around a 36" (a yard long) barrel. The folks at BBtI who did the actual testing seemed to think that it was somewhere around half of that - at around an 18" long barrel.

If you want to do the testing and publish the kind of comprehensive data that they have, as opposed to just doing a critical numerical analysis of THEIR data, then that would be very useful. Barring that, I think I'll go with their conclusions. No offense.

FWIW, based on their published data, I already actually have gone with their data and chose 20" long barrels as the best compromise between performance vs easy handling & maneuverability when I was purchasing my 357 mag and 44 mag lever guns.

They describe their methodology on their website:

BBTI - Ballistics by the Inch :: Actual Process

They talk about using two chronographs in their scope of project link.

They chose 18" barrel length for reasons they never state, although they are clear that they started with 13 handgun cartridges in mind, and expanded to 22.


To be fair, the BBTI folks are not the folks who make claims about barrel length and bullets going slower past a given length. Those claims are normally made people who look at the data and don't understand its limitations.

I most often hear that claim in reference to .22 LR, where it's commonly held that bullets "go slower" past 16". In most cases, even on the face of it, before you consider the SD within the data for a given round in a given barrel length being greater than the differences between adjacent barrel lengths, the BBTI data doesn't even support that. Yet this data is often used to make that claim:

BBTI - Ballistics by the Inch :: .22 Results

If you look at it you'll find a couple examples where the 16" velocity is higher than the 18" but only by a few ft per second and well within the range of expected variation. In a couple cases you won't find any significant difference between the 11" and 18" velocities. That's an artifact of small sample size and large SD, but people misunderstand what that means.

The small sample size gives an average that is greatly skewed by normal variation in velocity and it produces results that would in all probability be much different in subsequent tests. When you put the different brands and barrel lengths together all with that flaw, the longer the barrel length tested the smaller the velocity differences and the lower the likelihood that you'll get velocities that are statistically significantly different. But that gets mis construed as "proof" that .22 LR bullets don't go faster in longer barrels.

I've been using a chronograph in my load development since the late 1990s. I've also used a chronograph to measure .22 LR ammo for consistency and velocity for my match rifles and precision rifles and sporters. There is always some potential for barrel to barrel differences, and yet when testing statistically significant numbers of the same lots of ammo in the same rifles varying from 16" to 25", with several examples in the 20"-24" range, the longer barrels are consistently a little faster than the shorter barrels, and they generally have lower SDs in velocity.

Consequently, based on the BBTI data and testing methodology, and my own experience over the last 25 years I am not hesitant to make an informed rebuttal to the "bullets go slower in barrels longer than 16" in handgun calibers" myth.

I've even heard the same nonsense regarding .223 Rem/5.56x45.
There is a sweet spot and inflection in the graph right around 18" where the gain in velocity per inch starts to decrease past that point and that's one of the reasons why both Eugene Stoner and Ruger chose 18" for the AR-18 and Mini 14 barrel length.

But 20" barrels area still faster and 24" barrels as faster still, they just don't gain as much velocity per each additional inch. But that gets mis remembered or mis understood as "they go slower".

 

[BB57]

Fun video. Velocity results were interesting. I'm waiting for someone to get upset about no "obvious " backstop on the range.

The data he collected is a great example of small N and how velocity data can be skewed by a fast or slow round in the sample.

With the mag tech load data that he shows on screen, the 16" has a high velocity outlier of 1787 fps, while the 20" data has a couple low rounds at 1683 and 1693 fps. The average velocities are only 5 fps apart (1737 versus 1732) while the SDs for both groups are 36 fps. More to the point with a round 50 fps above average in the 16" group with just rounds fired that one round drags the average up 8 fps. At the same time the two rounds in the 20" group that 39 fps slower and 45 fps slower drag the average down 25 fps.

But the high SD of 36 in both groups is the kicker. It's the difference greater than 36 fps that make all three of those rounds above outliers that skew the data.

Those high SDs also mean there is no significant difference in velocity, even at the 67% confidence level, between the two groups, precisely because the outliers in a small sample make it impossible to measure differences between them. With the high SDs of 36, you won't be able to show a statistically significant velocity difference until you have a velocity difference of at least 37 fps, or until you have a lower SD. A 74 fps difference is required for a statistically significant difference (ie velocity difference not due to random chance) at the 95% confidence level.

If he'd had chronographed 30 rounds of each he'd have a lot more data points to determine the averages, with much, much less chance of high or low outliers impacting the average. In addition, assuming the ammo is really not that inconsistent, he'd have a much lower SD, probably down in the 10-20 fps range. That more robust data set would then allow an actual conclusion about relative barrel velocity.

Even if Magtec ammo really is that bad, a larger sample will clearly show it and by using the more robust sample size, a much more valid average and a valid SD, you could then clearly determine what the outliers are and draw meaningful conclusions about which barrel is faster and by how much.

In this case looking at the outliers (more than 1 SD from the mean) and excluding them you get averages of 1716 fps for the 16" and 1755 fps for the 20" barrel and I suspect that would be in the ball park you'd find with a 30 shot sample of each. Even if the SD were still 36 fps, a 39 fps difference in averages would still be statistically significant supporting an argument that the 20" barrel is slightly faster with that load.