TUGR Golf Rankings - How is this not the only accepted method of ranking?

[King_Slender]

This is not a LIV vs. PGA thread and I have no connection to TUGR.

 

If you aren't familiar with TUGR, here is a brief run down of how they do their rankings using a mathematically elegant formula:

 

https://tugr.org/method

 

As someone who is into mathematically sounds solutions (e.g. if you're someone who understands the mathematical elegance of Bitcoin, you'll likely get TUGR right away), I don't see how this isn't the ONLY method by which we rationally rank golfers.  

 

OWGR, which sets up various criteria in order to run a tournament that would qualify, setting up essentially arbitrary strength of field criteria and making exceptions when mother nature gets involved to allow shortened tournaments to get output some OWGR points.

 

TUGR seeks a mathematical elegant solution by simply asking: How can we compare Golfer A to Golfer B by assessing their performance against the world of professional golf? 

 

We have various Tours and there is always some crossover - PGA guys will play in tournaments with LIV guys or Champions Tour guys on the DP World Tour and the majors, so we do have the ability to compare their performance in common fields.  Those results can then be used to create rankings against guys who are not in those fields the same way you have a golf handicap and know that when your 0.9 index shows up against a 11 index, you're the better golfer.  The only real difference is that your handicap is based on your performance "against the course" and TUGR creates rankings based on common opponents.

 

Open to arguments against TUGR, but it seems OBVIOUS these are the real world golf rankings. 

[TLUBulldogGolf]

Or just use DataGolf rankings.

[SnowbirdTom]

Or meet the OWGR rules for competition in their tournaments?  It’s not rocket science.

[King_Slender]

7 minutes ago, TLUBulldogGolf said:

Or just use DataGolf rankings.

 

I like data golf too, but to me basing everything on what is essentially strokes gained is overly complicated and presents some confounding factors.  If you look at their rankings there is some shuffling of the Top 10 guys, but Scheffler is #1 in both and they have essentially the same Top 10 guys - I prefer the simplicity of TUGR's methodology.

[King_Slender]

11 minutes ago, SnowbirdTom said:

Or meet the OWGR rules for competition in their tournaments?  It’s not rocket science.

 

The argument against that is that it does not answer this question: Given the current state of the game of golf, how do we properly rank all of the players in the world?  

 

OWGR worked at a time when all of the top players were playing on the PGA or Euro Tours, but today's game is very different, including guys who play on the Champions Tour who can still compete in any given week during a major or tour event.

 

To stick to that only answers the question: Who are the top players playing in events that meet OWGR specific criteria.  

 

That necessarily omits a large portion of golfers from the data set, given the current state of the game.  I'm not one who is beholden to times gone by - just as equipment has made today's game vastly different from the game played years ago, so should the tools change that we use to assess who the best players are.

[bcjim]

36 minutes ago, King_Slender said:

This is not a LIV vs. PGA thread and I have no connection to TUGR.

 

If you aren't familiar with TUGR, here is a brief run down of how they do their rankings using a mathematically elegant formula:

 

https://tugr.org/method

 

As someone who is into mathematically sounds solutions (e.g. if you're someone who understands the mathematical elegance of Bitcoin, you'll likely get TUGR right away), I don't see how this isn't the ONLY method by which we rationally rank golfers.  

 

OWGR, which sets up various criteria in order to run a tournament that would qualify, setting up essentially arbitrary strength of field criteria and making exceptions when mother nature gets involved to allow shortened tournaments to get output some OWGR points.

 

TUGR seeks a mathematical elegant solution by simply asking: How can we compare Golfer A to Golfer B by assessing their performance against the world of professional golf? 

 

We have various Tours and there is always some crossover - PGA guys will play in tournaments with LIV guys or Champions Tour guys on the DP World Tour and the majors, so we do have the ability to compare their performance in common fields.  Those results can then be used to create rankings against guys who are not in those fields the same way you have a golf handicap and know that when your 0.9 index shows up against a 11 index, you're the better golfer.  The only real difference is that your handicap is based on your performance "against the course" and TUGR creates rankings based on common opponents.

 

Open to arguments against TUGR, but it seems OBVIOUS these are the real world golf rankings. 

I searched for Nick Dunlap and he doesn't seem to be ranked.  If,  true, and maybe im doing it wrong,  that would seem to be a bug. He is 71 owgr.

[SnowbirdTom]

Simple question back atcha - why should we care about certain folks that don’t want to be rated under a stable and fully accepted system?  It’s worked for quite a while and many tours.

[Cliffy2020]

13 minutes ago, SnowbirdTom said:

Simple question back atcha - why should we care about certain folks that don’t want to be rated under a stable and fully accepted system?  It’s worked for quite a while and many tours.

 

22 minutes ago, King_Slender said:

 

The argument against that is that it does not answer this question: Given the current state of the game of golf, how do we properly rank all of the players in the world?  

 

OWGR worked at a time when all of the top players were playing on the PGA or Euro Tours, but today's game is very different, including guys who play on the Champions Tour who can still compete in any given week during a major or tour event.

 

To stick to that only answers the question: Who are the top players playing in events that meet OWGR specific criteria.  

 

That necessarily omits a large portion of golfers from the data set, given the current state of the game.  I'm not one who is beholden to times gone by - just as equipment has made today's game vastly different from the game played years ago, so should the tools change that we use to assess who the best players are.

Kinda sums it up.

[idrive]

How do you rate a closed shop where the same players play each other every time?

 

These guys could be much better or much worse than the rest of the Tours. Who knows, how can you rank them until they are on equal grounds?

 

They delegitimize themselves when bringing in a player who hasn't played professionally in 12 years. No qualifying, just lots of money for showing up in an attempt to grow viewership.

 

 

[grochol17]

I went to their website and when talking about their simple methodology they say:

 

"Let a powerful computer calculate rankings with incredible precision."

 

Oh, well, if there's a powerful computer involved then it must be true.

[King_Slender]

43 minutes ago, bcjim said:

I searched for Nick Dunlap and he doesn't seem to be ranked.  If,  true, and maybe im doing it wrong,  that would seem to be a bug. He is 71 owgr.

 

I'll shoot them an email and see what they say. 

[King_Slender]

20 minutes ago, idrive said:

How do you rate a closed shop where the same players play each other every time?

 

These guys could be much better or much worse than the rest of the Tours. Who knows, how can you rank them until they are on equal grounds?

 

They delegitimize themselves when bringing in a player who hasn't played professionally in 12 years. No qualifying, just lots of money for showing up in an attempt to grow viewership.

 

 

 

15 minutes ago, grochol17 said:

I went to their website and when talking about their simple methodology they say:

 

"Let a powerful computer calculate rankings with incredible precision."

 

Oh, well, if there's a powerful computer involved then it must be true.

 

 

Both of these questions are answered the same way - you aren't just comparing them to the 54 guys they play every week.  Yes, that helps inform the rankings of those 54 guys, but when those 54 guys then play in tournaments on the DP World Tour or in majors, they then get baselined (and vice versa) against the rest of the world.

 

If LIV players in the majors consistently performed in the bottom of the tournament, then every guy in LIV would lose ranking under the TUGR system and be behind almost everyone on the PGA Tour.  However, if their top guys are performing well head-to-head, then they should be ranked on equal footing.

 

Just as an example from last year's majors:

 

The Masters: Rahm won and 4 of Top 10 (all 4 finished in the Top 6) were from LIV

PGA: Koepka won and 3 of Top 10 were from LIV

US Open: 2 of Top 10 were from LIV

British Open: 1 of Top 10 were from LIV

 

Clearly the best LIV players aren't ranked in the 100's under any objective measure, so TUGR allows that baselining when there are tournaments that they play together.  

[King_Slender]

52 minutes ago, SnowbirdTom said:

Simple question back atcha - why should we care about certain folks that don’t want to be rated under a stable and fully accepted system?  It’s worked for quite a while and many tours.

 

See above - I answered this.  And we can be objective and say that OWGR is no longer acceptable given the current state of the game. 

 

I have no loyalty to any Tour - I simply want to see the best players playing together as much as possible in the majors, above all.

[geauxt]

1 minute ago, King_Slender said:

 

See above - I answered this.  And we can be objective and say that OWGR is no longer acceptable given the current state of the game. 

 

I have no loyalty to any Tour - I simply want to see the best players playing together as much as possible in the majors, above all.

Seems to be a simple disagreement as to whether OGWR is still “acceptable.”

Im in the camp that it is. 

[cassel191919]

3 minutes ago, King_Slender said:

 

See above - I answered this.  And we can be objective and say that OWGR is no longer acceptable given the current state of the game. 

 

I have no loyalty to any Tour - I simply want to see the best players playing together as much as possible in the majors, above all.

I think its acceptable, I tend not to care about guys who choose to play exhibition golf for a lot of cash. When they venture away from the circus they tend to get some ranking points.

[Pops99]

The LIV stans sure go to some lengths to make everyone else change for a closed-shop, buddy, invite only, hit and giggle controlled by Third Leg Greg league that chose to disrupt golf. 

 

Actions, consequences, reaping, sowing, etc. YMMV. 

Edited March 4 by Pops99
.

[TLUBulldogGolf]

1 hour ago, King_Slender said:

 

I like data golf too, but to me basing everything on what is essentially strokes gained is overly complicated and presents some confounding factors.  If you look at their rankings there is some shuffling of the Top 10 guys, but Scheffler is #1 in both and they have essentially the same Top 10 guys - I prefer the simplicity of TUGR's methodology.

 

They basically seem to be doing the same thing. I don't know what's complicated about strokes relative to the field.

 

TUGR's methodology is plainly stated as 'Let a powerful computer calculate rankings with incredible precision.' Ok.

 

I think LIV is dumb and I find it funny that they are faced with the consequences of not meeting very clear requirements for rankings but it would probably be better at this point for the OWGR to just figure out how to incorporate their events.

[PracticeSwinger]

1 hour ago, King_Slender said:

This is not a LIV vs. PGA thread and I have no connection to TUGR.

 

If you aren't familiar with TUGR, here is a brief run down of how they do their rankings using a mathematically elegant formula:

 

https://tugr.org/method

 

As someone who is into mathematically sounds solutions (e.g. if you're someone who understands the mathematical elegance of Bitcoin, you'll likely get TUGR right away), I don't see how this isn't the ONLY method by which we rationally rank golfers.  

 

OWGR, which sets up various criteria in order to run a tournament that would qualify, setting up essentially arbitrary strength of field criteria and making exceptions when mother nature gets involved to allow shortened tournaments to get output some OWGR points.

 

TUGR seeks a mathematical elegant solution by simply asking: How can we compare Golfer A to Golfer B by assessing their performance against the world of professional golf? 

 

We have various Tours and there is always some crossover - PGA guys will play in tournaments with LIV guys or Champions Tour guys on the DP World Tour and the majors, so we do have the ability to compare their performance in common fields.  Those results can then be used to create rankings against guys who are not in those fields the same way you have a golf handicap and know that when your 0.9 index shows up against a 11 index, you're the better golfer.  The only real difference is that your handicap is based on your performance "against the course" and TUGR creates rankings based on common opponents.

 

Open to arguments against TUGR, but it seems OBVIOUS these are the real world golf rankings. 

 

I'll try and chime in and refrain from the 'just comply', 'we're better than you are' road : looking at it from a pure stats point of view.

 

The 'danger' I see from their method is that it puts a lot of emphasis on the cross-over results to extrapolate the rankings, from their correlation matrix. I play in my beer hockey league and you play in yours all year long, we meet once a year for a tourney and your team beats mine - then it carries a lot of weight on the rankings of every team in my and your league... Niemann wins the Masters this year, would drag Pat Perez ranking up a lot compared to Jake Knapp and others, even though we don't have any comparables between these guys - inversely, Niemann missing the cut would flip it. I understand that it's the only 'numerical' way to compare everyone, just my opinion that it draws a lot of conclusions from a 'few data points' (few compared to intra tour data points).

 

There's no perfect way, that's for sure : why 18 months, weighted if the result was 16 months ago compared to last week? (if my only interraction with you was 16 months ago and you played against Steve last week - what's the correct implication on my ranking compared to Steve?)

[grochol17]

Just now, cassel191919 said:

I think its acceptable, I tend not to care about guys who choose to play exhibition golf for a lot of cash. When they venture away from the circus they tend to get some ranking points.

 

I think this is the crux of the issue.  The OWGR is basically saying that LIV isn't real golf, so rounds on LIV do not count for rankings.  Therefore, you only get points for official tournaments and if you don't play in those, then that is on you.  The LIV guys all knew this before they left for LIV, so I don't see why we should all be subjected to Taylor Gooch's attempt at inventing a new golf ranking system.

[jimecherry]

2 minutes ago, grochol17 said:

 

I think this is the crux of the issue.  The OWGR is basically saying that LIV isn't real golf, so rounds on LIV do not count for rankings.  Therefore, you only get points for official tournaments and if you don't play in those, then that is on you.  The LIV guys all knew this before they left for LIV, so I don't see why we should all be subjected to Taylor Gooch's attempt at inventing a new golf ranking system.

how is taylor gooch involved with tugr?

[grochol17]

1 minute ago, jimecherry said:

how is taylor gooch involved with tugr?

 

I was making a smart a** comment because it wasn't that long ago that he made his "Masters asterisk" remark...

[cassel191919]

3 minutes ago, grochol17 said:

 

I think this is the crux of the issue.  The OWGR is basically saying that LIV isn't real golf, so rounds on LIV do not count for rankings.  Therefore, you only get points for official tournaments and if you don't play in those, then that is on you.  The LIV guys all knew this before they left for LIV, so I don't see why we should all be subjected to Taylor Gooch's attempt at inventing a new golf ranking system.

I will add to this I also don't think they should get ranking pts for these stupid 70 player designated events with a 20 person cut. Like you said they knew what they signed up for, so stop the fing crying.

[grochol17]

5 minutes ago, Varry_Hardon said:

I'll try and chime in and refrain from the 'just comply', 'we're better than you are' road : looking at it from a pure stats point of view.

 

The 'danger' I see from their method is that it puts a lot of emphasis on the cross-over results to extrapolate the rankings, from their correlation matrix. I play in my beer hockey league and you play in yours all year long, we meet once a year for a tourney and your team beats mine - then it carries a lot of weight on the rankings of every team in my and your league... Niemann wins the Masters this year, would drag Pat Perez ranking up a lot compared to Jake Knapp and others, even though we don't have any comparables between these guys - inversely, Niemann missing the cut would flip it. I understand that it's the only 'numerical' way to compare everyone, just my opinion that it draws a lot of conclusions from a 'few data points' (few compared to intra tour data points).

 

There's no perfect way, that's for sure : why 18 months, weighted if the result was 16 months ago compared to last week? (if my only interraction with you was 16 months ago and you played against Steve last week - what's the correct implication on my ranking compared to Steve?)

 

Don't you know they have powerful computers with lots of precision? 

 

But, seriously, you're spot on... these kind of comparisons are prone to all sorts of issues when extrapolating from undersampled data and unless the TUGR people publish a paper explaining their method in actual detail with discussions of how they handle giving too much weight to one off events, I won't believe that they are doing this a particularly accurate way.

[Rapidcat]

Home tour bias needs to be factored into any world rankings.

 

US nationals playing the PGAT have an advantage over foreigners and so they get an easier run in the OWGR.
 

[Hawkeye77]

Of course this is yet another LIV thread - OWGR is it, LIV folks can beat their head against the wall all they want. 

[jimecherry]

6 minutes ago, grochol17 said:

 

Don't you know they have powerful computers with lots of precision? 

 

But, seriously, you're spot on... these kind of comparisons are prone to all sorts of issues when extrapolating from undersampled data and unless the TUGR people publish a paper explaining their method in actual detail with discussions of how they handle giving too much weight to one off events, I won't believe that they are doing this a particularly accurate way.

alot of ml is smoke and mirrors where they cant explain every result so i'm not shocked there's no white paper on exact methodology.

i also wouldnt say the sample is to small if they have every score of every hole a person plays plus course data that can add up real quick if you have enough tournaments. 

[jimecherry]

1 minute ago, Hawkeye77 said:

Of course this is yet another LIV thread - OWGR is it, LIV folks can beat their head against the wall all they want. 

and all alternatives to owgr should be shutdown ;}

[Rapidcat]

3 minutes ago, Hawkeye77 said:

Of course this is yet another LIV thread - OWGR is it, LIV folks can beat their head against the wall all they want. 

 

Joaquin Niemann is the best golfer in the world right now and he is ranked #72….#OWGRbroken.

[Hawkeye77]

1 minute ago, Rapidcat said:

 

Joaquin Niemann is the best golfer in the world right now and he is ranked #72….#OWGRbroken.

 

LOL, OWGR works, as you and everyone else know he chose to play where he can’t get ranking points for the bro league. 
 

Same old complaint and just another LIV thread trotting out something discussed over and over. 

[jimecherry]

Just now, Hawkeye77 said:

 

LOL, OWGR works, as you and everyone else know he chose to play where he can’t get ranking points for the bro league. 
 

Same old complaint and just another LIV thread trotting out something discussed over and over. 

owgr works so well that augusta was like that guy sucks so we'll give him a spot in our tournament ;}