In the process of make preliminary tests, some variability was seen in the results due to the dimples on a golf ball so a Bald Eagle Ball was used which, while it is dimpled as is typical of golf balls, it has un-dimpled areas which we can use for putter impact so that this dimple issue has no bearing upon this study.
The surface of any golf green has a myriad of variables which are extremely difficult if not impossible to quantify, and these variables greatly affect golf ball movement. So the test were conduct this testing in a laboratory environment on a perfectly level pool table surface which has a reliably hard and smooth and level surface which is covered by a consistently textured cloth. The 1st test was on a fast billiard cloth an the 2nd was done with a stimp 11 putting carpet. Thus all the myriad of golf green unknowns that would prevent us from drawing conclusions from the testing are factored out.
Whether to use a human or a robot to stroke the putter was a difficult decision. A robot would give absolute consistency of stroke, but no robot yet made can replicate the stroke of a human. Since this testing would be observed in ultra slow motion and and has the ability to precisely measure all aspects of the stroke, so a human was used for stroking and simply reject those human strokes that exhibit any human error in stroking. Thus there is the benefit of an actual human putting stroke with all its complexity, yet one can exercise the control necessary for good scientific testing. Obviously we need a human who has well developed putting skill so we do not have to reject so many putts due to observed human error.
In testing a consistent stroke strength which is about what it would take to roll a ball on a level golf green about 10 feet. Putts under 5 ft are usually made by any kind golfer using any kind of putter; putts over 15 feet are usually missed by any kind of golfer using any kind of putter; but putts between these two extremes (10 ft putt plus/minus 5 ft) is where the kind of putter that a golfer uses can make the difference between making or missing the putt. So it seemed logical that this is the stroke strength of interest. Stroke strength can be precisely measured by counting the number of frames of slow motion video it takes for the ball to move a given distance, so any stroke strength significantly greater or lesser than that desired will be considered human error and rejected.
The objective of this test is to observe golf ball movement after putter impact to observe: 1) Down-Line quarter roll and half roll. Down-Line Roll can be: a) topspin, b) backspin, or c) no spin (skid); and most would agree that topspin is desirable over the other two. Putting is a rolling ball game and obviously an airborne launch is the cause of ball bounce, which most would agree is undesirable in comparison with rolling bounce-less launch.
To perform a loft test the ball movement was observed resulting from impact by three different putters: 1) a putter having a negative loft angle, 2) a putter having no loft angle at all, and 3) by a putter having a positive loft angle. Since the effective loft angle incident upon the ball at impact consists of a combination of the static loft angle of the putter (that is its published loft angle or the angle as it is measured statically), the press of the golfer in his putting stance (the angle of the putter shaft at impact as viewed out perpendicular from the desired line of the putt with a forward press reducing effective loft and a backward press increasing effective loft), and the placement of the ball forward or backward from the point of tangency of the stroke arc with the green (backward placement causing a decrease in effective loft and forward causing an increase in effective loft). In order to eliminate the factors of press and placement and so to concentrate only on the factor of static loft angle, human to stroke was made such that the ball was impacted at the low point of stroke path tangent and with the putter shaft perfectly vertical such that the press and placement were factored out, and the only static loft is effecting the ball motion; The human tried to always impact the ball with the stroke strength approximating a 10 ft putt so stroke strength could also be factored out of the test. Many strokes are to be made with each putter and then after analyzing the slow motion videos, all those with any significant human error will be discarded so that hopefully we end up with three objective representations of ball movement (one for each of the three putter types) upon which we may be able to make some sound scientific conclusions.
What about all the variations in putter design, don’t they adversely affect any conclusions that might be drawn from these tests? Yes maybe, but that simply points out the need for further future testing. In this testing, we have done our best to isolate on the factors that we sincerely believe to be relevant for the test which was intended to be performed and leave the testing of all other factors for another time. For example, do groves or pads in the impact area of the putter head effect ball launch?.. Maybe yes, maybe now – we think not significantly, but that is the subject of another test. For another example, does the mass of the putter head effect ball launch?.. Of course mass is a function of putter power so that would that would effect the distance the ball will travel, given equal stroke velocity, but does it affect the spin or bounce of the ball?.. We think not, but the proof of this will have to be the subject of another study. For yet another example, does the shape of the putter behind the striking face have an effect upon ball launch?.. We think not unless the shape alters the perfect vectoring of the shaft into the center of mass of the putter head, in which case we believe that shape does matter, but again that will have to be left for another study.
What about variations in fitting, doesn’t that have an adverse effect on any conclusions you might draw? No actually not, because fitting factors affect human performance, good fitting makes it easier to perform the putting stroke perfectly, bad fitting makes it more difficult. The human performing all the putting strokes in this testing is actually using poorly fitted putters for his putting style and stature, and that does not affect the test because upon analysis of his putting strokes, we reject any of the strokes that do not meet testing requirements. So the result of poor fit is that more of those putting strokes are rejected than would otherwise be the case if he were perfectly fitted.
What about variations in Gripping and Back-weighting? No these factors are not important to this test for the same reason stated in the previous paragraph. Of course these factors may improve putting performance in some way or by some degree, but that is a subject of another test; in this test whether they make a difference in performance is irrelevant because we reject any significant human error during our analysis anyway.
The 300 frame per sec video of each test putt was reviewed and the result were measured. The video files were each named to have the measurement recorded in the file names. For those that want to see the raw data, the following is the file naming convention.
The 1st token in the file name is related to the manufacture and or model name.
The 2nd is CHS or club head speed measured in the number of frames it takes for the head to traverse 3 inches prior to impact.
The next 3 fields in the file name were to describe the loft.
ML: means the manufacturer’s published loft angle,
SL means the push/pull (press) angle of the shaft which ideally for the test is zero,
DL which means dynamic loft which is the sum of ML and SL.
Then the next field or token was QR which means the distance in inches it took for the all to achieve quarter roll. HR is the measurement of the distance in inches it took for the ball to achieve half roll. In some cases the total weight was specified as the value after the characters GR.
The results show that increasing the negative loft reduces the forward roll distance. The following are the plots of quarter roll and half roll vs loft that summarize the findings of the two surface types.
BILLIARD CLOTH TESTS
TMSPID Chs18 SL-0.5 ML4 DL3.5 QR9.2 HR12.2
Yhan CHS19 SL0 ML0 DL0 Low1.5in before LA 1deg QR8 HR11.2
Yhan Chs20.5 SL-1.4 ML0 DL-1.4 Low2inBeforeImpactPint LA5deg QR 6.75 HR 10 at 600fpsView
Yhan Chs21 SL0.5 ML0 DL-0.5 Low2inahead LA-1deg QR 7.5 at 600fpsView
CARPET TESTS (11 on Stimpmeter)
Whether to use a human or a robot to stroke the putter was a difficult decision. A robot would give absolute consistency of stroke, but no robot yet made can replicate the stroke of a human. Since this testing would be observed in ultra slow motion and we have the ability to precisely measure all aspects of the stroke, so we can use a human for stroking and simply reject those human strokes that exhibit any human error in stroking. Thus we have the benefit of an actual human putting stroke with all its complexity, yet we can exercise the control necessary for good scientific testing. Yes, we could have used a robot for putter loft and impact surface testing; but when testing for the significance of head twist on ball launch we could not because a robot would fasten to the shaft in such a way that head twist is prevented from twisting, while a human would hold the grip in such a way that head twist may be very significant.
Which human do we select for this job? Obviously we need a human who has well developed putting skill so we do not have to reject so many putts due to observed human error. So we selected Rick Malm, an avid golfer who is also an Artistic Pool Champion (trick shot champion for those who don’t follow the sport of pool that much). Artistic pool requires extraordinary skill, excellent hand-eye coordination, and precision muscle control; Rick has all that not only for the sport of pool but also for the sport of golf, so he is the perfect human to use for this testing. Rick is also a retired IBM Researcher with 30 years experience in computer architecture, graphics and data analysis. Rick has the high speed camera equipment that he uses for analysis of elite athletes in golf (Long Drive and PGA tour pros).