Pellet Head Measurement For Precision Shooting

In the second part of his investigation, Matt Coulter goes to the range to prove the results of pellet head measurement for precision shooting.

The results are startling. The groups reduced in size by as much as 36%!!!

You can read Part One here.

Out On The Range

One of the first questions to address was what distance is appropriate for this head size and accuracy testing?

I chose 50 Yards for the Air Arms and H&N pellets and 35 Yards for the Crosman pellets. These seemed reasonable for each pellet’s cost and intended purpose.

I wanted a range that was suited for the quality of pellet – not too long where wind would unduly affect results – but long enough to amplify what could be subtle difference between the pellet sub-sizes.

Shooting for this test was split between two days. The first day was a remarkably calm and this made for ideal 50-Yard shooting! A couple of dry patches were pulled through the barrel to establish a base line between each session.


An older (2005) single shot Air Arms S400 MPR rifle fitted with an Athlon 6-24×50 scope was used in testing.

While the Air Arms MPR is not a regulated gun from the factory, my gun has a larger air tube that has been regulated with an Altaros regulator. It delivers 80 (or more) shots with an extreme spread less than 20 FPS.

The second day of testing was slightly more windy, but still much better than a typical day where you’re battling the wind. It was on this second day that I tested the Crosman Premiers at 35 Yards.

For each pellet and head size, four 5-shot groups were shot and measured. Then the average group size was calculated.

Note: While the wind was minimal, I did this shooting outside so there was some air movement.  To try to minimize the effect wind would have on any one sized pellet, I cycled through the sizes when shooting groups.  For example, this is the order I shot the H&N pellets:
Unsorted group 1
4.50 mm group 1
4.51 mm group 1
Unsorted group 2
4.50 mm group 2
4.51 mm group 2
… Repeating for groups 3 and 4

Test Results And Data

1. H&N Field Target Trophy 8.64 Grain “4.50 mm” Group Sizes:


[table “624” not found /]

2. Air Arms Diabolo Field 8.4 Grain “4.52 mm” Group Sizes:


[table “625” not found /]

3. Crosman Premier 7.9 Grain Hollow Point Group Sizes


(This was just one of the four test targets shot for this investigation).

[table “626” not found /]


Looking at the average group sizes, 9 out of 10 groups shot with sorted pellets were better (smaller) than those shot with unsorted pellets.

The greatest improvement in group sizes were with the Air Arms pellets which also happened to be this gun’s “favorite” pellet where the average 50-Yard group size shrunk nearly a centimeter. That was a 36% reduction in group size. WOW!

So I would call that a win!

The H&N Field Target Trophy pellets which were sized closest to their advertised size (5.51) and had the least variation in actual size, saw just under 2 millimeter improvement in 50-Yard group size.

The Crosman Premier Hollow Point pellets saw a nearly 6 millimeter group size improvement in their 35-Yard groupings. And it was only the 4.53 mm sorted Crosmans that performed worse (on average) than the unsorted pellets. Even then, one of the four 4.53 groups was a beautifully tight 9.1 mm group.

This data shows that sorting pellets clearly DOES improve (shrink) group size. Overall, pellet head measurement for precision shooting reduced the group sizes by an average of 11.3%.

However, it would not be reasonable to think that by sorting pellets, that you will displace whichever pellet your gun chooses as its favorite. In this test, the best sorted H&N group is roughly equal to unsorted Air Arms in this test.

The main advantage seems to be once you find “THE PELLET” for your gun – as I did with these Air Arms Diabolo Field pellets –  you can further shrink the groups you are able to produce by dedicating the extra effort needed to measure and sort your pellets!

So do you need pellet head measurement for precision shooting? I’d say it’s essential!

For those who are interested, here’s the full data set from Matt’s testing. Thanks again Matt!!!

[table “627” not found /]