dood, do you realize that no one is arguing if choking exists or not? the nxt guy said he believes there is choking. he just doesn't believe not choking=clutch.

man some people just love to argue regardless if there is one to be had or not.

Pulling marbles requires no skill. It is entirely based upon luck and the expectations your propose will play out perfectly.

However, A PGA professional 3 putts from 15' less than 1% of the time. To me the 3 putts are indicative of a loss of focus more than anything. They simply don't 3 putt from 15'. Thus, if the player happens to 3 putt from 15' when the tournament is on the line it is likely completely a result of choking mentally. If you are caught daydreaming on #4 the first round and 3-putt from 15’ oops, but hopefully you are a little more present on 18 on Sunday. Honestly, how is that not obvious?

We all get it NXT, it is not definitive due to one trial nature and you won’t ever allow yourself to think of anything abstractly. That’s what makes you great, run with it.

However, to say that a player having a 1% phenomenon occur on the last shot of the event isn’t choking is honestly amusing. This isn’t poker where the luck of the deck will 100% create suck outs and bad beats once the money is in. Maybe that is where your disconnect stems from by your trying to make live golf behave like poker.

The only non-choking scenario is actually what likely happened though. Derek played earlier than the leaders and likely hit his first putt a tad too hard (thinking he needed another birdie) and missed some sort of comebacker. While I still would consider this mentally choking as he altered his ideal play based on his perception of the moment I wouldn’t think of it as poorly as if he were in the last group with a clear cut 2 putt for the win.

Again, if you could think things through you should have proposed this was likely what happened instead of grasping to the idea that %’s are the only thing that matters as the event plays out. If you were Derek’s strategist you’d just give him a “man pro, run worse, you got coolered there!” and tell him there is nothing he could do about it. Rather, maybe you could look into what happened and help him learn from the experience.

Hamsterdam is right.

Jesus.

the great irony is that NXT posts stats so often yet has no idea how to apply Bayes' theorem

Last year PGA Tour players 3 putted on average 1.05%(median .8%) of the time from 10-15' so this % is likely skewed down bc the large majority of putts occurred from less than 15'. The range was many guys at 0% and all the way down to Geoff Ogilvy at 4%.

Then you can look at 15-20' and see Tour pros 3 putted 1.6%(median 1.4%). This ranged from a bunch of guys at 0% to 5.41% for Kyle Stanley.

The Tour average % of 3 putting from 15 feet likely lies somewhere in the middle of these numbers, so around 1.2-1.3%. Let's use 1.2%.

According to Ship, TOUR players NEVER 3 putt from 15 feet so long as they aren't day dreaming. How often are they day dreaming over a 15 foot putt? For fun let's say 10% of the time, which seems rather high.

So if they never 3 putt the 90% of the time they are focused, they would have to 3 putt 12%(!!!!) of the time when they are day dreaming. No way that is the case, so no way TOUR pros NEVER 3 putt from 15 feet when they're paying attention. Let's say they 3 putt 5%(more than 4x worse than average) of the time when they're being lazy. They would have to be day dreaming almost 25% of the time when standing over a 15 footer to get to a 1.2% 3 putt %. Neither of these scenarios seem possible.

You are right I don't think it's definitive, let's look at how you extrapolating small samples can get us in trouble.

Ironically Tiger Woods/"mr clutch"/"mr incredible focus" 3 putted more often from 10-15'(1.8% over 111 attempts) than he did from 15-20'(1.3% over 76 attempts) during the 2013 season. For you, those are astronomical sample sizes and I assume from that data you can be 100% confident Tiger 3 putts more often from 10-15' than 15-20'.

I don't use Bayes' a lot, as I don't do a ton of predicting, and haven't used it in quite a bit so correct me if I'm wrong. I assume you are referring to the 3 putting from 15 feet example.

With the following assumptions.
Chance of 3 putting from 15 ft: 1%
If you "choke" the odds of 3 putting from 15 feet: 100%
If you don't "choke" the odds of 3 putting from 15 feet: 1%

After 3 putting from 15 feet, Bayes' theorem appears to give us a roughly 50.35% chance of "choking". Still doesn't appear to be definitive.

What the audience likes is irrelevant to who is the GOAT at golf.

Why not? Society collectively gives meaning to our words. If the majority of people interested in golf think greatness is defined by performing under pressure at the majors then it's pretty relevant to the discussion of who is the greatest.

So Tiger 3 putted twice in 111 attempts at 10-15'. Tiger 3 putted once in 76 attempts at 15-20'. Don't think this qualifies as a large sample size.

I think that's his entire point.

I think Dinopoker's post is great.

Dinopoker's post would make a lot more sense if all golf tournaments in history had been played on the same course with the same equipment with the same weather and time of day. To someone as skilled as Ronnie OSullivan, snooker tables probably have a decent amount of variance in the cushions and felt, but it's not in any way comparable to difference between golf courses or even the same golf course from one day to the next.

I'm really stoned right now and this is making a lot of sense. Good post.

My counter though would be that scoring averages aren't something you can just compare across eras without considering additional factors and this can become quite messy.

Right, you'd want to compare it in a statistically sound way against the era's competitors.

If that shows you a result you don't like, you can then claim something intangible like the lack of heart or extra grittiness of players from one era or another to justify the result you have reached despite the clear statistical evidence. Or you can rely on a single statistic that may or may not be dispositive of the question at hand but does have historical or romantic allure.

Somewhat interesting news on Tiger's ties to a doctor linked to PEDs in baseball. I think this has been reported before, but I guess the new angle is that the links are more extensive than previously thought.

http://www.golfchannel.com/news/golf...76k-14-visits/

IMO the median is clearly the number to use to try and normalize the data at least some to eliminate one handers and when players take odd stances around others lines since neither of those issues are what led to Derek 3 putting. That gives us 1.1% 3 putt avoidance from 15’, ABSOLUTE MAX, likely sub 1% but whatever. The 2 data sets you used have 48,823 total putts and I actually believe more of the three putts could be attributed to one handers and funky stances, but I’m good with 1.1% for illustrative purposes.

Well, luckily for us we have this figure thanks to my work with Dr Lardon and the Second Scorecard I’ve detailed here before. A good Tour average for commitment is 93%.


Again, here you just jump off the boat without thinking through the entirety of the problem. PGA players make roughly 23% from 15’, leaving them with a second putt to be distracted on 77% of the time as well. Does that change your math at all? LOL.

93% focus rate on first putt, which tells us 7% of the time they hit an unfocused putt. Let’s call that a 7% BHR (Bone Head Rate).

77% of the time they will have to hit a second putt (due to making 23% of the first putts) and they will have a mental lapse 7% of the time again, adding another 5.39% BHR to the original 7% BHR. This means that 12.4% of the time they will have a mental lapse at some point while holing out from 15’.

Let’s say that instead of your absurd just to make an incorrect point 3 putt rate of 12%, they three putt the more likely 5% rate when having a brain fart. That yields .62 3 putts per brain fart (.124 BHR * .05) which leaves us with .48% (.011 - .0062) for the simple variance of occasionally 3 putting to make you happy. These numbers look good because clearly the likely cause of three putting from 15’ is a brain fart, not physical skills on Tour.

I am subtracting the random mental lapse portion as I don’t think there is any chance of their not being in the moment in this setting. If they aren’t then that is clearly mental choking, yay. So the remaining .48%, 1 in 208, is what I think is a reasonable 15’ three putt avoidance is for random occurrences.

Clearly a 1 in 208 event can happen at any time, I’m simply saying that in this scenario the likely cause is choking and not random variance. Yes I understand that isn’t definitive enough for you, again I say just tell your guy he did great and that was just a bad timing. I’ll look deeper. You do realize that IRL you don’t always get enough information and occasionally have to think creatively for yourself, right? I guess set mining low limits doesn’t require thought though (that’s a joke before you fly off the handle).


As always, it is simply amazing how many times you try to put words into someone’s mouth as though that is their position. I’ve never said anything short of “I agree the sample sizes are statistically insignificant, but I will use the only data we have to ask probing questions.” If the data is all you got, it’s all you got. Try and use some judgment to decipher a meaning. Apply your ideas and see if it helps out at all. Or keep doing what you do best and wait for others to figure out the tough concepts and explain them to you.

Furthermore, with regards to your incessant need to argue. You agree that choking exists, likely have a prime example of choking occurring, yet still insist that isn’t choking purely so you can argue. So odd.


All in all I think your complete lack of experience in high level golf has you underestimating how hard it is to focus correctly in the most extreme pressures and select the correct shots. This is what makes the great caddies great. They have the experience and balls to remain present in the moment and think for their player when they simply don’t have the ability to.

Additionally, having the ability to simply perform to your average under the greatest pressure IS CLUTCH. That is the unquantifiable separator that Jack, Tiger, and all the greats possess. It is what allows people to win 5+ majors. Performing better than your competition under the gun is what clutch is. Doing it routinely is what makes you GOAT.

Derek needs to work on his mental approach, it failed him last week. How you can argue against that is truly amazing.

Wonder how many like myself that did consider him the GOAT and still do but if he doesn't win another Major than no longer give him that title.

That's the case now, but not in the future. We keep way better stats now with respect to course ratings and player performance. In the years to come this will make evaluating the overall skill of golfers from future eras against those of today much easier.

Does this mean that maybe Nicklaus and Hogan can't be properly evaluated against Tiger or against the 'next Tiger' whoever he (or she) may be? Probably. But so it is with Nicklaus vs Bobby Jones vs Old Tom Morris. (I mean, if you're using majors win to compare Nicklaus to Old Tom how is that even close to fair, when OTM only had one or two tournaments a year to play in and you had to travel to them on horseback!)

No, we will probably never know who was 'better' between Tiger/Jack/Hogan/Jones/Morris because there just aren't enough stats. Oh well. But in the future they will know whether 'their Tiger' is better than 'our Tiger' -- and my guess is that the comparison will be based on scoring, because nothing else matters.

Here's another, why does longevity matter? Who gets to decide how long you have to be great for?

Longevity matters because you have to prove you simply didn't outrun variance for a 5 year career. IMO though, once somebody has achieved what Tiger has over more than a decade we can rest assured that he didn't simply run hot. The GOAT isn't simply whose body holds out the longest, it is who played the best golf.

Again, Major GOAT = Jack, GOAT Player = Tiger.

Your math... I don't even... Let's go piece by piece

No, because my math is not breaking it down putt by putt. Mine is much simpler.

Mine assumed that when standing over a 15 footer a player has a 1.2% 3 putt rate, 90% focus rate and 0% 3 putt % when players are focused, since you said a Tour player could not 3 putt if they were focused.

So how often do Tour pros 3 putt when they are not focused if you believe the above assumptions? We just need to solve for X.

3 putt % over 15 footer = (90% * 0%) + (10% * X) = 1.2%

Since the focused make % is 0, the equation can be simplified to

10% * X = 1.2%

SO

X = 1.2%/10%

X= 12%
I understand the above may be some complex math for you to grasp, but please try. 12% is not some number a made up.

Let's run the above with your 93% focus rate and the 1.1% 3 putt % you are good with for illustrative purposes.

1.1% = (93% * 0%) + (7% * X)

1.1%/7% = X

X = 16% 3 putt % when unfocused. That seems like a lot especially when you seem to think the "real" unfocused 3 putt % is around...
Ok, now lets try the equation again adding your 5% number.

3 putt % = (93% * 0%) + (7% * 5%) = .35% 3 putt %

Notice how this number is nowhere near the 1.1% number that you were good with as a real 3 putt % from 15 feet? That's really weird.

LOL at this. How can you try to multiply their combined "brainfart" chances on their 1st and 2nd putt to their 3 putt % from the first location??

So you appear to believe that a Tour player now 3 putts from 15 feet .48% of the time when he is focused, and the other .62% of the time it is because he is unfocused?

Because .48% + .62% = 1.1%

LOLOLOL

You can't just take your average 3 putt number(1.1%) and subtract the "unfocused" 3 putt %(.62%) and get the "focused" 3 putt % of (.48%)

LOLOLOLOLOL

Do you even understand why? They happen at significantly different frequencies(93% vs 7%).

Let's ignore how wrong it was to get to this number and just plug .48% into the 3 putt % equation.

3 putt % = 1.1% = (93% * .48%) + (7% * X)

X= 9.3%

So in conclusion, based on your incorrectly calculated .48% focused 3 putt %, a Tour pro would have to 3 putt 9.3% of the time they are unfocused.

For reference a 90-golfer 3 putts 5% from 15 feet, so sorry if I don't believe that even when they are not paying attention a Tour player 3 putts almost twice as often as a 90s-golfer from 15 feet.

This whole debate started with someone saying that the example from the Web.com tour was...
I simply refuted that he could be so sure after witnessing 1 event.

Sometimes the data you have is not enough to come to any sort of definitive conclusion and that is my point. A lot of times it can get you going in the right direction though, but there are other times when it steers you completely wrong. You want to use data from small samples to prove your point, but then turn around and disagree with a conclusion that would be drawn from a different small sample. This is the exact problem with using small samples to try and make definitive claims. If the sample is really small "using some judgment to decipher a meaning" is not enough to overcome the issues with dealing with really small samples.

I always love when you put these condescending statements into post that are riddled with mistakes, especially when the mistakes you make aren't related to anything even remotely "tough".

Can you show me where I said for sure that Derek didn't choke? I just am pointing out that nobody can be sure what he did based on a sample size of 1. After looking at his 1 15 footer I would never definitively say he did 1 thing or the other.


If you think it is unquantifiable, why have you even been arguing or trying to use quantifiable examples like

54 hole closing record overall
54 hole closing record in majors
Performance in 4 straight major championships

as proof that clutch exists?

Without being anywhere near 100% sure Derek choked, it seems like it could be more detrimental to tell him that he did choke if he actually didn't, than to tell him that he didn't choke if he actually did.

Stumbled across this gem on Nate Silver's website www.fivethirtyeight.com today. It's a pretty decent read but there were some great, applicable takeaways for the discussions that have occurred in this thread.

http://fivethirtyeight.com/features/...aybe-not-pele/

The author is attempting to compare players from different eras. With regard to that he has the following to say...
He goes on to compare 3 players using multiple metrics.

Player A as the "largest" sample, followed by player B, then player C has the smallest.

Over the samples, A out performs B to the point where the author can be quite sure A is better than B despite issues with the size of the samples. However C actually outperforms A over his small sample. That prompts the following

Your timing could be better on trying to make an argument that modern players would dominate those of past years, considering Angel Cabrera won this week and Tom Watson tied for 35th.

First of all, I get it, I said “Tour players simply don’t 3 putt”. I fully recognize that they miss 3’ putts and thusly three putts will happen. I’m sorry if you have no gauge, again, for reasonable interpretation. I think that saying they simply don’t three putt is reasonable based on it being a once a year occurrence. They aren’t walking around in fear on 15’ putts.

Having two opportunities to daydream makes your math incorrect. You can’t 3 putt without hitting a second putt.

I don’t think it is made up, I only know it is wrong due to not controlling the data in some fashion to eliminate the one handers as well as using incorrect concentration %’s and their application.



What you fail to seem to grasp is that I am getting the 3 putts that are a result of lack of focus out of the data. I think you know that, but need to run with the 0% number in order to keep an argument alive.

Controlling for attentive efforts gives us a true expectation for how often a PGA player can expect to 3 putt from 15’ based on skill and bad bounces. I get it, you need to use the zero percent in order to make your math look sweet, but that is not what I think reality is. Can you move back to a discussion based in reality? I’m removing the brain farts as there is no chance a brain fart occurred from day dreaming in this specific instance.

Again, do you even understand what I did to the data? I used the expectation of daydreaming occurrences, in conjunction with an appropriate three putt figure, in order to control the data removing brain farts from the expected three putt rates. This leaves us with a .48% chance of three putting that is attributable to skill once mental errors are removed.

That is why I say that I understand a 1/208 event can occur, but in my opinion that is more likely the result of a choke than bad luck.

A poor mental performance on either putt can result in a three putt. They are independent trials, like flipping a coin more than once, and should be viewed independently.

If you are able to concentrate 93% of the time and have to hit 2 shots you only have an 86.5% chance of concentrating on both shots. I even reduced the number of second attempts based on how often a player makes the initial 15’ putt.


Why not? If we agree that 1.1% is the true average 3 putt rate and we have controlled the data to establish that .62% is attributable to mental lapse. The remaining .48% is left to skill and bad bounces. I know you won’t get this though.

We have already removed those differences in frequency when drilling down to their respective expectations, kind of like a pro-rata share.



Again, you have two trials both with 7% fail rates. You need one more (7% * X) in your equation above and then solve….what do you know, that results in 5%, like it should. As for the 90’s shooter, you are the one who has said a million times “putting simply isn’t that hard”. So for the EV of three putting from a mere 15’ for a 90’s shooter to line up with a distracted Tour pro on significantly more penal greens seems about right. It doesn’t take much of a mental lapse to bang one 4’ by.



Fully agree, and that is where experience and ability come into play.

Question, how would it be possible to get more than one data point for one golf shot? I understand that if Derek had a 2 putt to win every week that in a couple years we could see what his conversion rate is. But for this one specific occurrence we can only ever have one data point. Instead, in an individual sport we often need to look at general statistics and then relate our own small set of data and make a few inferences.

Justin Rose needed to hear that he was a good wedge player and change his perception of himself. What on the other hand if he had that hunch and the data said he was merely ok? Should those two ideas start to form an opinion and practice initiative? Heck, it’s even possible that he is a bad wedge player and the data set was too small to represent that yet due to being early in the year. There’s that variance again.


I do too, especially when you continue to defend your position incorrectly.


Agree with this too. I certainly wouldn’t go to him as my student and say “WTF, nice choke.” But I would ask what in his estimation happened. Derek knows if he choked or not and to dodge the question would not serve him in the long run. Why not address the 800 lb gorilla and see what it has to say? The toughest conversations are often the most fruitful.

That's what I was thinking. Watson's success shows how golf may be a little different in this regard.

You have to look at the 15 footer only. You cannot do any calculations based on the 2nd putt because you have no idea how long the 2nd putt someone has will be. DUH.

You did no such thing to control the data. You think you did, but you did not.



Again, you didn't. I coincidentally don't think pros 3 putt 0% of the time even when they are focused, so the number isn't doing me any favors.

Again just wrong.

You did none of this. Hilarious you think you did, sadly it will take someone else to convince you otherwise.

If you really wanted to solve for this, if you think Tour pros focus 93/7 respectively and when they are unfocused they 3 putt 5% of the time here is how often they would have to 3 putt when focused to get to an average of 1.1%

1.1% = (93% * X) + (7% * 5%)

X = .81%

I have a feeling there is no way you think Tour pros 3 putt .81% of the time from 15 feet when they are focused though. If that is the case, one of the other numbers has to change.

And explain to me how you are able to do calculations on the 2nd putt when you have no idea of the length of said putts?

Well I won't get it because you are using this .48% interchangeably for 2 conceptually different things at the same time.

1. The 3 putt avoidance number for focused pros from 3 feet.
2. The amount of the 1.1% overall 3 putt % that focused putts account for These are different things. Maybe you don't realize it, but they are not the same.

Since you believe all of the following, we can plug it into the equation and solve for the unknown.
1.1% overall 3 putt rate. .48% 3 putt rate when focused. Focused 93% of the time, unfocused remaining 7%.

1.1% = (focused rate * focused 3 putt %) + (unfocused rate * unfocused 3 putt %)

1.1% = (93% * .48%) + (7% * unfocused 3 putt %)

Solving for the unfocused 3 putt % gives us 9.34%. Again, no way Tour pros 3 putt that often even when they aren't paying attention. So one of your assumptions must be wrong.

Now lets look what % focused/unfocused make up of that 1.1%

1.1% = (.45%) + (.65%)

Brain farts .65% divided by overall 3 putt % 1.1% = 59%


Or lets use all of your numbers to calculate "your" player's hypothetical 3 putt %
All of the same assumptions from above except we now know you think they 3 putt 5% of the time when they are unfocused, so we can solve for their overall 3 putt %.

3 putt % =(focused rate * focused rate 3 putt %) + (unfocused rate * unfocused 3 putt %)

3 putt % = (93% * .48%) + (7% * 5%)

3 putt % = .45% + .35%

3 putt % = .8%

Still not the 1.1% you agreed on when doing calculations.

And their, brain farts would account for only 44% of the overall 3 putt %(.35/.8)

This is just a perfect example of how you have not "controlled" the data at all nor removed the "mental errors"


See above

More wrong. Again you cannot do the 2nd trial because you have no idea how long of a putt will be left.

<3