[Veritas]

So after reading the complaints about the hotlist and a lack of data. I thought maybe we should run some of our own experiments. We have the golfers to do it and the are I'm sure scientifically inclined people on this board would be willing to handle the analysis. I for one am definitely willing to do some analysis. One of the questions that I got interested in was how much a premium ball versus a "cheap" ball affects your scores after reading the topic on the question.

In order to run the experiment we would need some golfers who play a lot of golf who would be willing to play a significant number of games with both types of balls. We would want to be able to detect at the minimum a 2 stroke difference provided by the balls. As a result of all the factors that go into scoring, and after doing a power analysis we would need at least someone who could play 30 rounds with each ball. A total of 60 rounds. This is a lot so I don't know if anyone would be interested in doing this, but I just wanted to put it out there that I'm willing to do the analysis and share the results here if there are some golfers willing to do the testing. The more willing to participate the more interesting the data would be. Handicaps of all varieties can participate as it would be interesting to see whether the effect or lack of an effect varies among skill ranges.

[ultra45]

i think the idea sounds pretty good... but it is pretty tough to judge how a ball performs based on score alone... and there are a lot of individual preferences that could affect how the ball "performs" that don't really make a ball better or worse, just different

[8thehardway]

I don't know much about stats and testing, but here's some input anyway.

Play 9 holes with each ball to duplicate temperature, weather and course conditions as much as possible

Alternate the starting ball every round to allow for lack of warm up, 1st tee jitters, etc

Note course conditions. Rain may give the value ball an edge as spin rates are minimized or not a factor in holding wet greens.

Is 30 rounds/ball (540 holes) the minimum needed for relevant data? If it is, and you might want to eliminate testers from states where golf can't be played the entire year. It would smooth out the data by eliminating the opening rounds of a season which could be weeks apart and produce atypical scores.

[Rhod]

Keep in mind that I myself will play shots different ways depending on how much a ball spins

[Diesel]

probably hard to pull off unless you played 2 balls for an entire round...

your swing might not be the same another day and conditions could be different which affect ball characteristics...

i can tell you while i was out today there is a HUGE difference between a MOJO and a 330s as far as spin into and around the greens... i've been playing MOJO's i have laying around cause it's only about 45 degrees, but go sick of nothing holding on the green, and releasing through... i hit one from 120 with a PW and the thing hit took a forward bounce and rolled through the green... i then hit a 330s from the same spot and the thing hit, took one little hop and rolled back off the green... pretty telling story...

both shots hit into a slight upslope of the green...

[Boba]

DieselMp32, on Dec 28 2007, 09:38 PM, said:

probably hard to pull off unless you played 2 balls for an entire round...

your swing might not be the same another day and conditions could be different which affect ball characteristics...

i can tell you while i was out today there is a HUGE difference between a MOJO and a 330s as far as spin into and around the greens... i've been playing MOJO's i have laying around cause it's only about 45 degrees, but go sick of nothing holding on the green, and releasing through... i hit one from 120 with a PW and the thing hit took a forward bounce and rolled through the green... i then hit a 330s from the same spot and the thing hit, took one little hop and rolled back off the green... pretty telling story...

both shots hit into a slight upslope of the green...

I doubt playing two balls would give the results you are thinking about. For the same reason a mulligan is often better than the prior shot.

I'm not a statistics major, but acquiring a simple random sample of golfers who can crank out 60 rounds will be far more difficult than finding 60 golfers who can play two to four rounds.

[Veritas]

All good thoughts guys. I am definitely open to suggestions and putting together a test that most of us can agree on. One thing that I hope people also get out of this is how difficult it is to put together a useful golf experiment. There are a lot of tradeoffs you make due to the variablitiy and it is not as easy as some people suggest when complaining about the ones put out my golf digest. I will add my thoughts about everyones response below.

ultra45, on Dec 28 2007, 05:00 PM, said:

i think the idea sounds pretty good... but it is pretty tough to judge how a ball performs based on score alone... and there are a lot of individual preferences that could affect how the ball "performs" that don't really make a ball better or worse, just different

Yeah there are definitely preferences. It is well documented that these balls perform differently with different characteristics. But the question I am interested in asking is how big of an effect does this have on a players score. The reason why I choose score is because that is the goal of golf and changing your equipment is so you will play and score better. It can give a good insight into how much you are getting from playing that premium ball over a "rock flight". I would think we would have people pick there two balls and allow them to pick balls they prefer. It's not a test of one specific ball verse another specific ball but more of a premium ball verse value ball test.

8thehardway, on Dec 28 2007, 06:32 PM, said:

I don't know much about stats and testing, but here's some input anyway.

Play 9 holes with each ball to duplicate temperature, weather and course conditions as much as possible

Alternate the starting ball every round to allow for lack of warm up, 1st tee jitters, etc

Note course conditions. Rain may give the value ball an edge as spin rates are minimized or not a factor in holding wet greens.

Is 30 rounds/ball (540 holes) the minimum needed for relevant data? If it is, and you might want to eliminate testers from states where golf can't be played the entire year. It would smooth out the data by eliminating the opening rounds of a season which could be weeks apart and produce atypical scores.

I think this playing setup is a very good idea. We would definitely want the person to switch the ball they start up each round. In fact, thinking about it we could cut down the need to 30 rounds this way.

DieselMp32, on Dec 28 2007, 08:38 PM, said:

probably hard to pull off unless you played 2 balls for an entire round...

your swing might not be the same another day and conditions could be different which affect ball characteristics...

i can tell you while i was out today there is a HUGE difference between a MOJO and a 330s as far as spin into and around the greens... i've been playing MOJO's i have laying around cause it's only about 45 degrees, but go sick of nothing holding on the green, and releasing through... i hit one from 120 with a PW and the thing hit took a forward bounce and rolled through the green... i then hit a 330s from the same spot and the thing hit, took one little hop and rolled back off the green... pretty telling story...

both shots hit into a slight upslope of the green...

Yes your swing will be different on different days but that is the reason for the large sample size. You will expect to if you just played to have a normal distribution of scores (good and bad days surrounding an average score). We need to sample enough to get a good representation of these scores and distribution. By having a lot of samples we will get good days and bad days with both balls. And there is no doubt that the spin characteristics are different between balls. I just think it would be interesting to see how big of an effect this has on a players scores.

Boba, on Dec 28 2007, 08:51 PM, said:

DieselMp32, on Dec 28 2007, 09:38 PM, said:

probably hard to pull off unless you played 2 balls for an entire round...

your swing might not be the same another day and conditions could be different which affect ball characteristics...

i can tell you while i was out today there is a HUGE difference between a MOJO and a 330s as far as spin into and around the greens... i've been playing MOJO's i have laying around cause it's only about 45 degrees, but go sick of nothing holding on the green, and releasing through... i hit one from 120 with a PW and the thing hit took a forward bounce and rolled through the green... i then hit a 330s from the same spot and the thing hit, took one little hop and rolled back off the green... pretty telling story...

both shots hit into a slight upslope of the green...

I doubt playing two balls would give the results you are thinking about. For the same reason a mulligan is often better than the prior shot.

I'm not a statistics major, but acquiring a simple random sample of golfers who can crank out 60 rounds will be far more difficult than finding 60 golfers who can play two to four rounds.

Yeah you would have the affect of adjusting after the first swing. You could possibly address this by alternating the first ball on each hole. The difficulty here is getting someone who plays a lot of golf and has time to play two balls every time out. Yeah it would be easier to get more players but there are a couple of difficulties of doing this. The first one is that we would expect different levels of golfers to be affected differently by the balls so we would really need to get groups of golfers that are similar in playing level, which would not be that easy. Secondly, by adding different golfers we would produce a larger range of scores, which makes us need even more samples. We would need around 500 golfers to play a round with each ball to get the statistical power to make the same determination of the 60 rounds by one golfer. And these golfers would all need to be around the same handicap. So as you can see this is not easy. The advantage of one golfer is that you start to limit variables. But I'm open to anything we are capable of doing.

I am also interested in doing any other experiments people are interested in

[Veritas]

Rhod, on Dec 28 2007, 08:09 PM, said:

Keep in mind that I myself will play shots different ways depending on how much a ball spins

Yes that is great. That is what we want. One thing that might be interesting to see is that maybe you just adjust to the different ball and end up scoring around the same regardless of ball. Or maybe no matter how much you adjust you just have a much better score with the premium ball due to its characteristics. As a result, I would want you to play your best with each ball and change your playing to each ball.

[ultra45]

i think the test would have to go away from score and more to specific shots... because there are too many variables that can affect score that won't be related to ball performance specifically... i think the best scenario would be to try and hit the same shots side by side with different balls and see what happens... of course if you're playing a busy course you may not have the time/ opportunity to do that... but hit a few diff balls of each tee and see if/what happens.. same thing for say 100 yards and 50 yards... maybe even try to shape shots with different balls and see if some are easier to "work" than others...

even if you read different posts on here already... some people absolutely love a certain ball and the next person can hate it and wouldn't play it if they were paid...

IMO... the best test would be to get good ol iron byron... get a bunch of diff balls... and see what happens with diff clubs... and then you'll get great result if you could swing exactly the same every time...

for myself... when i pick a golf ball.. i want it to stop on partial wedges... as long as it's not significantly shorter than what i'm used to... brand does not matter to me...

[Veritas]

"i think the test would have to go away from score and more to specific shots... because there are too many variables that can affect score that won't be related to ball performance specifically..."

This is very true but once again that is why we have to obtain a large sample size. With the large sample size we should be able to see the effects of the ball on the score if there is one. All these various effects on the scoring that don't have to do with the ball are what cause your range of scores. Or statistically speaking this will affect the standard deviations of the population.

The score test is designed to look at how your preferences actually affect your score and not just your perception of the affect on the score.

I in no way think your idea for testing is a bad idea...I just think it answers a different question than the original question I proposed. Yours answers which balls can be worked, which balls go further, which balls spin more. My question asks how do those abilities (working the ball, spinning more, etc.) affect your score. I would be interested in seeing both experiments.

[MacMia]

rdbivyleagun, on Dec 28 2007, 03:29 PM, said:

So after reading the complaints about the hotlist and a lack of data. I thought maybe we should run some of our own experiments. We have the golfers to do it and the are I'm sure scientifically inclined people on this board would be willing to handle the analysis. I for one am definitely willing to do some analysis. One of the questions that I got interested in was how much a premium ball versus a "cheap" ball affects your scores after reading the topic on the question.

In order to run the experiment we would need some golfers who play a lot of golf who would be willing to play a significant number of games with both types of balls. We would want to be able to detect at the minimum a 2 stroke difference provided by the balls. As a result of all the factors that go into scoring, and after doing a power analysis we would need at least someone who could play 30 rounds with each ball. A total of 60 rounds. This is a lot so I don't know if anyone would be interested in doing this, but I just wanted to put it out there that I'm willing to do the analysis and share the results here if there are some golfers willing to do the testing. The more willing to participate the more interesting the data would be. Handicaps of all varieties can participate as it would be interesting to see whether the effect or lack of an effect varies among skill ranges.

Not to muddy the waters, but what the heck..

I think what I would like to see is for each participant to take samples of 6 types of balls and hit from 4 different shots / yardages- Driver, 150,100,60, 10 foot putt (hitting a total of 25 balls per type).

In terms of how you would conduct the test, I would also think that it would make sense to mix in some range balls in each shot type.

So, follow me on this...

Driver- hit 40 balls, 5 brand 1, 5 brand 2, 5 brand 3, 5 brand 4, 5 brand 5, 5 brand 6, and 10 range. Mix them all up in a shag bag and hit them as they come out. Track the distance, dispersion, etc...

You could also stratify your sample by handicap and swingspeed... I also think that by doing this, you could get the testing done over a period of days. 60 rounds is a lot...

[RunOfTheDill]

probably should black out logos because the ball can you give you a mental security.

[MacMia]

RunOfTheDill, on Dec 29 2007, 12:47 PM, said:

probably should black out logos because the ball can you give you a mental security.

Love that idea... Other than having different dimple patterns, it would be about as generic as you could get it.. Probably could do the test with a bunch of Logo Overruns or Even X Outs as long as they were blems and not defects...

[Veritas]

Once again, the experiments you are suggesting are more about the ball characteristics of each ball and mine was about how those characteristics affect your score. While the characteristic test could be conducted on a much shorter basis, it would not answer the original question that I proposed. If no one is interested in that question then no problem. I am a PhD student so my test design was not just pulled out of my butt. The sample size was calculated using a power analysis based on desired alpha level, desired Beta level, projected mean scores, and standard deviations. And I was definitely disappointed when I saw the large number of samples needed to answer the question, but I thought if there was any place where someone was willing to do a golf test over 60 rounds it would have been on golfwrx.

Now since people don't seem to be interested in the original question, especially due to the long term nature of the study, lets just focus on the test you guys want to run.

First off, you guys seem to be interested in an experiment about characteristics of the balls across different balls. What is the specific question you guys are trying to answer? If we figure that out first we can better design the experiment we want to run and get more out of it.

[ultra45]

rdbivyleagun, on Dec 29 2007, 12:03 PM, said:

Once again, the experiments you are suggesting are more about the ball characteristics of each ball and mine was about how those characteristics affect your score. While the characteristic test could be conducted on a much shorter basis, it would not answer the original question that I proposed. If no one is interested in that question then no problem. I am a PhD student so my test design was not just pulled out of my butt. The sample size was calculated using a power analysis based on desired alpha level, desired Beta level, projected mean scores, and standard deviations. And I was definitely disappointed when I saw the large number of samples needed to answer the question, but I thought if there was any place where someone was willing to do a golf test over 60 rounds it would have been on golfwrx.

Now since people don't seem to be interested in the original question, especially due to the long term nature of the study, lets just focus on the test you guys want to run.

First off, you guys seem to be interested in an experiment about characteristics of the balls across different balls. What is the specific question you guys are trying to answer? If we figure that out first we can better design the experiment we want to run and get more out of it.

no offense to the original question... i'd really be interested in the results as well... to be part of it... i don't think i play 60 rounds in a season... and i also "try" to play the same ball to try an be as consistent as possible.. or at least keep the variables as consistent as i can... i also like thinking about different set combinations... wedges... shafts.. yada yada... but when i bring them up to my boss to get her (head professional) opinion... her answer is "stop thinking and swing the club"

cause i thought about what experiment would i like to see... so since i like a ball that stops on partial wedges... i'd like that... BUT... what ball, what type of lie/grass... which wedge... what is it landing on (slope, flat, etc) and that's leaving individual differences out of it... now i'm no phd student and have non done stats for a very long time... but at the end of the day i think the experiment would be either too specific or too general

also i guess... is there a difference between different wedge/ball combinations from the fairway vs rough vs different length rough or grass types... maybe different combinations are "better" from the different conditions