Community - Forum
It is currently Sat Jan 20, 2018 6:35 pm

All times are UTC - 8 hours




Post new topic Reply to topic  [ 9 posts ] 
Author Message
 Post subject: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 9:10 am 
Offline
User avatar

Joined: Mon Sep 08, 2008 2:37 pm
Posts: 2794
I have heard it claimed many times that there are thresholds -- special Power or Skill levels that give you a larger than average increase in distance. I never believed this, but was curious to do a test to see if any results hinted at the possibility of there being these "magic" thresholds.

My test, albeit very limited, does not expose any such threshold.

Let's get right to the results...

Image

Observations

While the results do not show a perfectly smooth increase as the skill stat is increased, one can make the following observations:

  • The results are fairly smooth. There are no "magic" thresholds revealed by this particular test. That is not to say that if one tested 1 stat at a time, some threshold would not be revealed.
  • The results are smoother still, refuting further the notion of a skill threshold, if one factors in the variability of the percentage increase in the stat. For example, we see the last entry being 6.6% more, but the stat increase was 15.3%.
  • There may be some small tendency to get more payoff as you increase the skill stat. This would explain why skill pays off more at high levels than adding power.

To further test the notion that there are likely no magic thresholds, we did one last test. We set the skill stat at 187, keeping all other stats the same as above.

For this skill stat of 187, we should be able to predict the distance gained. We started with 65 skill and a 6 iron distance of 264 yards. We ended with 151 skill and a distance of 354 yards. That is 90 yards gained in 86 stat points added to our base of 65, or 1.047 yards/stat.

So, at a stat of 187, we are adding 122 points, which, at 1.047 yds/stat, should give us a distance increase of about 127.5 yards, for a distance of 391.5 yards for the 6i, full 3 hit.

I did the math before I did the test. The math predicted a 6i distance of 391.5 yards and the actual hit (average of perfect hits) was 387 yards, a difference of only 1.2%.

Notes

  • Irons were old Hurricane and the 3W was a Hybrid; both without any fittings.
  • I had planned to do the same test for Power, but lack of affordable F/Pedoras was the problem. I used 3 for this test.
  • 15% was added, but due to rounding to an integer (since you can not add like 3.5 stat points), the actual increase varied as shown in the last column.
  • Ball used during testing was a +1 Power 2piece (Helios)
  • Normal conditions apply: 25 degrees and average of 2 perfect hits.
  • Full 3 spin was set using Shift-<RightArrow>
  • There are other interesting tests one could do. If I get some F/Pedora donations, I will do more.


Cheers, A


Last edited by A_In_NS on Sat Nov 07, 2009 2:59 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 9:39 am 
Offline
User avatar

Joined: Wed Dec 31, 1969 4:00 pm
Posts: 1037
Interesting. I think you should make a topic linking all these studies you're doing and get it stickied. :mrgreen:

_________________
Image

~ Comprehensive Club Comparison Guide
~ Comprehensive List of Stat Clothes
~ In Depth Stat Guide
~ My Shot Online Goals


Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 10:39 am 
Offline
User avatar

Joined: Mon Jun 15, 2009 2:09 pm
Posts: 1302
This is great stuff.

I'd be interested in seeing setups that aren't necessarily a skill setup either.


Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 10:59 am 
Offline
User avatar

Joined: Sun Jul 20, 2008 11:04 pm
Posts: 925
Location: Norway
Interesting data collected there. Well done.

However I do not see the logic in comparing how much a % increase in skill does increase the total distance hit in percentage?

With that logic having 0 skill would hit 0 yards, which is obviously not the case. This also explains why you get a higher %-wise increase for each % increased at higher skill levels, simply because the assumption that an x % increase in skill would generate a certain percentage-wise increase in total distance is flawed, regardless of how many additional attribute points that percentage-wise increase may represent.

To me it makes more sense to compare how much extra does each additional skill attribute give at different levels (which is in fact what you use when you try to predict the distances hit with 182 in skill). If you do this you get the following yard per skill attribute for the ranges you have created by the data collection:

65-75 skill: 11 y/10 ski = 1,1 yards / skill attr.
75-86 skill: 11 y/11 ski = 1,0 yard / skill attr.
86-99 skill: 14 y/13 ski = 1,08 yards / skill attr.
99-114 skill: 15 y/15 ski = 1,00 yards / skill attr.
114-131 skill: 17 y/17 ski = 1,00 yards / skill attr.
131-151 skill: 22 y/20ski =1,10 yards / skill attr.

To me I would simply analyze this data to mean for each skill point added, you gain 1 yard on a full 3 o'clock 6 iron and same for 3w. Analyzing the data like this even strengthen the grip on busting the myth that there are certain skill thresholds.

Just look how close the approximation 200y+1.0y*skill is on your data for 6i and 252y+1,0y*skill for the 3w.

Amazing really I would say, it's more or less dead on. But, I am not saying this applies for skill levels outside this range, would be interesting to find out. How does the graph behaive when skill is very low, close to zero...? Is it still linear?

Would be interesting to see the same figures for a 1w.

I really appreciate the work you are doing.

The reason why so many people tend to advice to swap to skill build when you with reasonable levels of Power/Impact can allow 100 in skill, is probably because it's about at that point the extra distance you get from the skill build outweighs the shots lost due to the increased complexity of hitting shots from that build and not because there is such an amazing difference in distance ranges between 90 and 100 in skill.

For power though, there is just no way in **** the graph for such a test is linear for relevant ranges of comparison. I mean the difference between say pw20 imp50 ski30 vs pw25 imp50 ski30 appear to be waaay bigger than pw60 imp50 ski30 vs pw65 imp50 ski30. Would be amazed if this is not the case. Would be cool to see a full result for certain club (neutral spin) with say:

pw0 imp50 ski35
pw10 imp50 ski35
pw20 imp50 ski35
pw30 imp50 ski35
pw40 imp50 ski35
pw50 imp50 ski35
pw60 imp50 ski35
pw70 imp50 ski35
pw80 imp50 ski35
pw90 imp50 ski35
pw100 imp50 ski35

And then maybe on of the classic, how much in power, how much in skill for high skill builds.

Would be cool to see 1w/3w/6i neutral spin for the following stats (and also full 3 o'clock spin perhaps):

pw20 imp45 ski170
pw23 imp45 ski167
pw26 imp45 ski164
pw29 imp45 ski161
pw32 imp45 ski158
pw35 imp45 ski155
pw38 imp45 ski152
pw41 imp45 ski149
pw44 imp45 ski146
pw47 imp45 ski143
pw50 imp45 ski140

I don't know, just what I would like to know ;)

_________________
Character name: AnotherLevel (Camila)
Level: 85.197 (Tour-Pro) Freya-Lvl: 6,033 (9/270)
Formerly known as Norsken (retired lvl66 Belz since sep4-2009)


Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 1:24 pm 
Offline
User avatar

Joined: Mon Sep 08, 2008 2:37 pm
Posts: 2794
A cardinal rule of any testing is to change one variable at a time. Given a limited budget, I stuck to this rule.

There are many ways to approach this one, and if I had more F/Pedora, I would have done 2 further tests:

* increasing power alone
* increasing power and skill in fixed increments increments (%age-wise)

And, I would not mind doing your proposed test either.

Still, with F/Pedora going off at 7-8 million, one can not go too crazy.


Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 1:58 pm 
Offline

Joined: Wed Dec 31, 1969 4:00 pm
Posts: 812
i'll donate a pedora to the test maybe more ppl will also


Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 2:12 pm 
Offline
User avatar

Joined: Mon Jan 05, 2009 1:05 am
Posts: 1563
Location: Wimbledon, London, UK
Very interesting as always ainns.
I have always interpreted that there are (potentially) thresholds that are level-related rather than stat-related. You could argue that these are the same thing considering the stats for any given level are always known, but the 'hidden stats' would be the explanation for the sudden change in ball flight upon reaching a certain level (in my experience, level 78 is such a level threshold).

One way to check for such hidden stats might be to have two toons, one a true (e.g.) level 78 and one a level 102 de-stated to look like a level 78. If the latter shoots further then that would be due to the hidden stats accumulated between lv78 and lv102.

If this were true, then it is possible that the 1 stat you get for every level-up over level 100 actually equates to more than 1 stat when you factor in additional hidden stats.

Anyway, just hypothesizing :) Don't want to sound like a UFO chaser ;)


Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 3:04 pm 
Offline
User avatar

Joined: Mon Sep 08, 2008 2:37 pm
Posts: 2794
Thanks. I'll add CBird in game. If I get a few more, will post more tests. Would like to do a driver test too. If anyone has low durability Zyg clubs, particularly the following, I'll do the driver test:

* old Hurri driver
* new Hurri driver
* Hybrid driver

Cheers


Last edited by A_In_NS on Mon Oct 19, 2009 3:07 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: Test results: Are there "magic" thresholds?
PostPosted: Mon Oct 19, 2009 3:05 pm 
Offline
User avatar

Joined: Mon Sep 08, 2008 2:37 pm
Posts: 2794
Hmmm... nice idea. Would need to "borrow" a mid-level toon :). I only have mule toons for inventory. Nice idea though-- really like it.

Edit:
Test was performed. Results here:
viewtopic.php?f=36&t=45700


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 9 posts ] 

All times are UTC - 8 hours


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for: