2019 MGS Ball Test

[iacas]

11 minutes ago, JxQx said:

It is kind of odd that the test shows almost identical numbers for the longest and shortest balls with the only difference being 3mph ball speed. Are the aerodynamics really that bad on the shortest ball to cause that big of a distance loss?

No. More and more it seems like a badly conducted test.

[JxQx]

10 minutes ago, iacas said:

No. More and more it seems like a badly conducted test.

Yeah it definitely seems like it. It would be nice to see an open dialogue between the TXG guys and the MGS guys about the testing as they are having conflicting results, but MGS just brushes all comments off that ask for clarity.

A test procedure along with the data would have been really nice.

[saevel25]

Here is a comment from the comment's section in the MGS article, in particular towards a comment about the video @iacas posted. The comment is from an employee at MGS. 

Quote

The fact is that their setup couldn’t have been much more different from ours. Any reasonable person would expect a different result. Anybody with knowledge of testing (specifically ball testing) and basic critical thinking skills should be able to understand as much. 

We tested with a robot Why? Because it ensures consistent contact from swing to swing. A golf ball is round. Short of laying the blade into the upper half, you don’t hit it the ball off-center. Anything else that happens…high face, low face, heel, toe, the force is applied proportionally and will impact every golf ball mostly the same.

Not to get sidetracked – all of a sudden, the forgiveness benefits of low compression balls are being promoted largely because they lose less speed when you miss the sweet spot. This is true, but it’s mostly a clever use of math meant to confuse golfers. Consider this scenario: Your ball speed is 150 with a high compression ball and 148 with a lower compression ball. An off-center club strike costs you 5% of ball speed. With the high compression ball you’re now at 142.5 having lost 7.5 mph. At 148, that same 5% only costs you 7.4 mph. YOU’VE SAVED BALL SPEED (.1 mph in this scenario – because of the softer ball). Granted, you’re still almost 2 mph slower, but you’ve lost less speed – you’re a tick closer than you were before. Now, it is true that because of the low spin properties of soft balls, the relationship is not exactly 1:1, but according to one golf ball expert I spoke with about “forgiveness”, soft balls do retain ever-so-slightly more speed, but not to any degree that he would feel comfortable telling golfers that a soft ball is more forgiving.

Back on topic…we tested with a robot, outdoors, full flight. Why? Because it’s what every expert we talked to (+/- half a dozen over the last 5 years) told us was the right way to do it. It’s almost certainly the methodology any ball company would have recommended before the test. The full flight aspect is critically important for ball testing as dimple design can have a significant impact on downrange performance (the number we’ve been given is “up to 8 yards” due to dimples alone). Again, this is what everybody in the ball industry knows (just like everyone in the ball industry already knew that softer is slower). It’s exactly why, despite having the means to do an indoor ball test for the last 5 years, we waited until we could use what the industry says are the right tools.

The difference between TXG and MGS is the following as far as I can tell;

1. TXG was indoors.
2. TXG used a different launch monitor versus MGS using Trackman 4. It looks like a GC-Quad that TXG used. 
3. TXG used an actual golfer versus MGS using a Robot.

I am not sure you can compare the two tests. Does the indoor launch monitor provided close enough real world (70 degree, no wind) comparable results? 

Also from TXG, the same employee, 

Quote

This is the hardest point to convey to those who haven’t seen it first hand. Over the course of the test, there were several “holy shit, did that just happen” moments. Balls go wildly offline. Wind was mild during testing, and we still saw several balls fly significantly offline (in some cases AGAINST the general direction of the wind). When that happens, you check the robot, you check the data, and invariably the only thing left is the golf ball. I’ve mentioned it in previous comments, the ball experts we’ve spoken with have all seen it themselves. The cause could be one of a number of things – ball out of round, uncentered layers (most notably the core), aerodynamic issues (dimple irregularities, raised seams, that sort of thing).

Golf balls are not 100% consistent, not perfect, and it’s pretty disconcerting when you realize that through no fault of your swing, a near perfectly struck ball can still go OB.

Not specified in their article, there was mild wind, what ever that means. That is something I think should tried to be isolated out. They could have posted results noting the wind direction and speed. I would think you'd want the test done on a pretty calm average day to make sure the data has less variables.

 

[boogielicious]

13 hours ago, saevel25 said:

Here is a comment from the comment's section in the MGS article, in particular towards a comment about the video @iacas posted. The comment is from an employee at MGS. 

The difference between TXG and MGS is the following as far as I can tell;

1. TXG was indoors.
2. TXG used a different launch monitor versus MGS using Trackman 4. It looks like a GC-Quad that TXG used. 
3. TXG used an actual golfer versus MGS using a Robot.

I am not sure you can compare the two tests. Does the indoor launch monitor provided close enough real world (70 degree, no wind) comparable results? 

Also from TXG, the same employee, 

Not specified in their article, there was mild wind, what ever that means. That is something I think should tried to be isolated out. They could have posted results noting the wind direction and speed. I would think you'd want the test done on a pretty calm average day to make sure the data has less variables.

 

This is where a proper Design of Experiments method comes in. By randomizing the test, blocking certain variables (like club type) and gathering a statistically significant number of samples, the variation will be shown as either significant or not. The OEMs do all their testing this way because it would support claims. To prove your ball is longer, has more spin, etc, it has to be able to be defended in court. A crappy test could be ripped apart.

[JxQx]

17 hours ago, saevel25 said:

Not specified in their article, there was mild wind, what ever that means. That is something I think should tried to be isolated out. They could have posted results noting the wind direction and speed. I would think you'd want the test done on a pretty calm average day to make sure the data has less variables.

 

Yeah the wind speed/direction definitely should have been recorded especially since they are saying the golf ball aerodynamics can have up to an 8 yard impact on distance. That should be made even worse in the wind. 

 

Also just as a general question to everyone, are there any places you can get fitted for a ball that is full flight with a trackman? I've only really seen indoor fittings for balls with maybe the exception being manufacturer fitting locations. It seems like it would be impractical to do because of having to separate balls at whatever range you are at.

[dennyjones]

9 hours ago, boogielicious said:

Design of Experiments

Haven't heard that phrase since my many years in the quality lab at GM.   I completely agree with your logic

 

9 hours ago, boogielicious said:

This is where a proper Design of Experiments method comes in. By randomizing the test, blocking certain variables (like club type) and gathering a statistically significant number of samples, the variation will be shown as either significant or not.

+1

[boogielicious]

16 hours ago, dennyjones said:

Haven't heard that phrase since my many years in the quality lab at GM.   I completely agree with your logic

 

+1

I’ve done dozens of them for all kind of product and process testing. It takes patience, but you know the data is robust.