Well, in this case the RG, cover checmisty, and surface will be kept relatively constant. So, by your theory, all three of these balls will react identically. And thats possible. Thats why it's a theory and you test it.
Why on earth do you think the cover chemistry is a constant between three different balls made by three different manufacturers? other than the fact the covers are all resin and you don't have a urethane or plastic in there the covers are not a constant. The surfaces are all different for each of those balls also not a constant in the arsenal. I never said the balls will react the same I expect the Reax to be earlier and the Defiant to be later with more backend I have no idea what that innovate will be like. That's the issue here is your making assumptions. You assume that because you have three pearls the coverstocks are the same they are not. You have three polished balls so the surface is the same they are not.
True, but as we've argued about before...how successful surface changes are depend on two rather important questions.
1) How long does a X Abralon surface last?
Depends on the surface low grits ie 500 starts changing within a game higher grit surface last much longer on average 500-1500 should be refreshed by hand after every series. 2000-4000 refresh by hand every other series. Polished surfaces are a little more difficult but I redo them when the ball reaction changes usually about every 40 games or so.
How does the surface change each time the ball is thrown? Surface grit always changes towards matching the lane surface around 4500 to 5000 grit. There have been numerous studies done on this I'm not just making it up
2) How effective are hand applied surface changes?
Very effective for unpolished surfaces not so much with polished surfaces
If surface changes are your strategy, you need to know roughly how the use of the ball changes the surface. If you know that a 3000 Abralon ends up being a 3130 Abralon after 1 series...then you can make adjustments to surface accordingly. And, just as important, you have to be able to change surface by hand. If not, you need a personal ball spinner. Most inexpensive models are poor quality. The more professional units are very costly. And paying a pro shop $25 per ball every 3 games...thats essentially $75-$100 per week of surfacing cost. For a pro that gets this service for free by their rep out in the truck...not a big deal. For the 99.4% of the rest of the bowlers...it's an absurd expense.
No one suggest you have your ball completely resurfaced every 3 games maintain lower grit surface by hand and redo polished surfaces every 40 games or so. I know even in Cali you can find a proshop to resurface for $15 mine lets me use there machine for free as long as I buy my pads and polishes from him. $15/40=$.37 per game think you can afford it.
I don't understand this comment. You seem to be saying RG is irrelevant and manufacturer differences are myths. I can tell you that PBA professionals...more than 1...have stated there is a BIG difference between one company and another. Radical is believed to hook the soonest...followed by Brunswick. Ebonite/Track/Columbia tend to hook the latest. The rest of the brands fit in between.
I am in no way stating RG is irrelevant not sure where you got that from but brand is completely irrelevant when a ball starts to hook is determined by Surface, Cover Stock, R.G, and maybe somewhat the differential although that's really more how not when. Notice brand is nowhere included in that list. Anybody who thinks all Radical balls are early have never thrown the Ridiculous or the Reax Gusto. Storm is well known for balls that are late and have big backend take a look at the Alpha Crux that ball isn't late. The ball manufactures try to make a complete line anymore, thinking that particular manufactures are only making one type of ball motion is just a fallacy.
Are you sure? How did Motiv cheat? Answer = Differential. Why would a company choose differential as the way to cheat if differential didn't really matter much?
One of the reasons why you and I argued about how serious the Motiv Carnage thing really was. On the Pro Tour I doubt to many of them are troubled by the difference of whether a ball has a differential of .61 or .59. For bowlers at our skill level I don't think we would notice if the ball was .61 or .41. Differential primarily effects flare potential most of us don't have releases powerful enough to cause 7+ inches of flare anyway and even if we did it really doesn't matter as long as the ball is flaring enough to expose fresh ball surface it doesn't matter if the gaps between the flare rings are .25 or .75 the effect is the same.
Poor PerfectScale...nobody likes it. : ( The bottom line is you need a quantitative way to compare balls from numerous manufacturers. There is no method that is currently available that is more likely to be worth a darn. It's not "Perfect"...but nobody else has a better system...so until they do, we have PerfectScale.
Not to mention, I've seen a decent correlation between PerfectScale and overall ball reaction. Lets look at an example:
Arsenal #1.
When I put Arsenal #1 together, a couple years back, I thought the progression would be:
Hammer Rhythm, 900Global Bullet Train, Columbia300 Encounter, Brunswick Slingshot. That is putting together an arsenal based on cover material. Solid, Hybrid, Pearl, Pearl. But when I went to put this together, the ball driller that was going to drill the Bullet Train said the Bullet Train cover (S79) would be too strong to be used as a ball down option and he would suggest the Bullet Train be the 1st ball in the progression for medium-heavy oil.
So the new progression was Bullet Train, Rhythm, Encounter, Slingshot.
Had I used RG, the progression would have been Rhythm (2.50), Encounter (2.51), Bullet Train (2.55), Slingshot (2.586).
Had I used manufacturer strength, it would have been Slingshot, Bullet Train, Rhythm, Encounter.
Had I used PerfectScale, it would have been Bullet Train (222.8), Rhythm (211), Encounter (203.8), Slingshot (157.6). <----Ding Ding Ding!!
In other words, PerfectScale predicted the exact arsenal and progression that the ball driller eventually suggested. Why? How? Luck? Randomness?
It's actually quite simple. PerfectScale takes into account all of the above specs, data, and how balls behave when tested...to give a number that has a lot of information that goes into it's make-up. I'm not trying to convince anyone to use PerfectScale. I promise, I get no royalties...I don't work for bowlingball.com...if I did, Bowl1820 would probably resign and we'd lose about 1/3 of our community.

But so far...I have seen no actual data nor testing that shows PerfectScale to be any less reliable a system for arsenal and progression development...than any other spec. The great thing about the arsenal and progression tool I'm working on is it takes into account ALL of these items. PerfectScale is just 1 of 7 variables. Currently all values are weighted equally...but that would probably change once I get the testing complete.
I think you just proved my point about the importance of who the manufacturer of the ball is
I meant by having all the cores assymetrical, that variable is held constant. I didn't mean symmetry didn't mean anything.
I know what you meant and it's still wrong.
Radical Reax RG Int. .013
RG Defiant Edge RG Int. .017
Ebonite Innovate RG Int. .008 (actually should be considered symmetrical)
Neck not even close to a constant
Again...not saying you can ever isolate just one variable...when comparing balls from numerous time periods and manufacturers. But you also will have a difficult time trying to understand the effects of specs on a progression or arsenal...if you have a large number of variables. Every time you reach a conclusion...someone can just point to one of the many variables and say your conclusion is wrong.
The trick is, with 7 variables...potentially 8 or more...to truly isolate a variable...would require a much larger pool of equipment to test. I am interested...but not that interested...nor that wealthy. But if a manufacturer wants to furnish me the equipment, I'd gladly accept the challenge.
Consider the surface, rg., cover strength, manufacturer intent, and especially what you need the ball to do and make your decision. It doesn't require 7 or 8 data points to build an arsenal. If you want to make it easier decide on a brand preference and it makes it even simpler.
Like I argue with Rob. At the end of the day you need a system you can use to build an arsenal and are likely going to be building an arsenal of a lot of different manufacturers. That means you need a way to slot balls in your arsenal. Of course there are nearly 10 variables to consider...and thats the trick. Which of these variables MEAN something? Which of them MATTER? If they all matter, then we'll never have a true system to create an arsenal.
Just might as well buy 3 random balls and use constant surfacing to try an manipulate ball movement.
It makes more sense than using brand as a basis in putting your balls in a progression
Bookmarks