Luke Donald Back to #1

[Zwick]

I think you misunderstood me there.

They shouldn't only count the last month, just update the rankings once a month and not once a week. The running 2 year system is fine.

Originally Posted by Zeph

They got the FedEx Cup and money list for the week to week changes. When you make a list of the worlds best athletes in any given sport, you want more than the last month as reference.

[k-troop]

Originally Posted by Zwick

I think you misunderstood me there.

They shouldn't only count the last month, just update the rankings once a month and not once a week. The running 2 year system is fine.

That wouldn't make the relative rankings any less arbitrary.  At any given calculation, there are oddities within the formula which might cause relative rankings to change in a manner inconsistent with recent play.  It's feasible that, given which day of the month you choose as your update, Rory would have never made it to #1 at all.  (i.e. say he's #1 for two weeks in Feb, two weeks in Mar, two weeks in Apr, etc, but never on the magic "recalculation date".)

Agree with Shorty's previous post, that who is #1 according to the OWGR doesn't mean a whole lot these days.  When the numbers are really close, they're apt to change from week to week, even if the affected players don't play.  It still gives you a good idea of who's who in golf.  You can argue whether #2 is better than #1, but no one is going to argue that #1 isn't better than #30.

[brocks]

I think you misunderstood me there. They shouldn't only count the last month, just update the rankings once a month and not once a week. The running 2 year system is fine.

The problem is not how often the rankings change. Baseball fans are perfectly fine with first place in the standings changing every day, and they can understand that a team that won its first 50 games is going to stay on top for a while, even if it loses its next ten games. The problem is simply that people can't be bothered to take ten minutes to understand how the system works. Determining the strength of field is a bit arcane, but the rest is just fourth grade arithmetic. The only significant difference from other sports is that you don't get any points for finishing second in a football or baseball game, while golf tournaments typically award points to the top 30-60 players, depending on the strength of field. That said, I do think that in our "what have you done for me lately?" society, a two-year window is too long. IMO they should shorten it to 18 months, or maybe even a year, and cut off the depreciation at no less than 30%, rather than zero. It would increase the volatility of the rankings, but it would also result in a much closer correlation between the rankings and the public's perception of who is currently playing well.

[k-troop]

Originally Posted by brocks

The problem is not how often the rankings change. Baseball fans are perfectly fine with first place in the standings changing every day, and they can understand that a team that won its first 50 games is going to stay on top for a while, even if it loses its next ten games.

The problem is simply that people can't be bothered to take ten minutes to understand how the system works. Determining the strength of field is a bit arcane, but the rest is just fourth grade arithmetic. The only significant difference from other sports is that you don't get any points for finishing second in a football or baseball game, while golf tournaments typically award points to the top 30-60 players, depending on the strength of field.

That said, I do think that in our "what have you done for me lately?" society, a two-year window is too long. IMO they should shorten it to 18 months, or maybe even a year, and cut off the depreciation at no less than 30%, rather than zero. It would increase the volatility of the rankings, but it would also result in a much closer correlation between the rankings and the public's perception of who is currently playing well.

I like the 1-year idea, maybe with bonuses for wins, or wins in majors or something like that contributing to your ranking a little longer.  Unfortunately, this would lead to a lot more "should XXXX get a special invite to...." threads, but it would make the rankings more reflective of who is really the best.  It's definitely absurd that points for a win can be depreciated to the extent that, after about 90 weeks or so, they bring your 2-year average down.

[brocks]

I like the 1-year idea, maybe with bonuses for wins, or wins in majors or something like that contributing to your ranking a little longer.

Bonuses for wins and majors are already built into the system. It's very heavily top-weighted. If you get 100 points for a win (which you do for a major), you only get 60 points for second, 40 for third, on down to 14 for 10th, and 3 for 45th. And in the weakest PGA events, you get less than 30 points for a win, so winning a major is better than winning three Reno-Tahoes. You can get as little as 6 points for winning on the Canadian or South American tour, so you'd need to win 16 of those in a row to equal one major, and even then you would already be losing points on your first few wins from depreciation. You can get a complete explanation and downloadable point chart here: http://www.officialworldgolfranking.com/about_us/default.sps?iType=425

[sean_miller]

Originally Posted by brocks

Quote:

Originally Posted by k-troop

I like the 1-year idea, maybe with bonuses for wins, or wins in majors or something like that contributing to your ranking a little longer.

Bonuses for wins and majors are already built into the system. It's very heavily top-weighted. If you get 100 points for a win (which you do for a major), you only get 60 points for second, 40 for third, on down to 14 for 10th, and 3 for 45th. And in the weakest PGA events, you get less than 30 points for a win, so winning a major is better than winning three Reno-Tahoes. You can get as little as 6 points for winning on the Canadian or South American tour, so you'd need to win 16 of those in a row to equal one major, and even then you would already be losing points on your first few wins from depreciation.

You can get a complete explanation and downloadable point chart here:

http://www.officialworldgolfranking.com/about_us/default.sps?iType=425

Do they split the points like they do with prize money? I mean if 5 guys tied for second place, don't they split 2nd through 6th place money, then the next guy gets 7th place money?

Does a guy finishing T2 three times get only 20 fewer points than a guy with 2 wins and a MC in those same events?

[k-troop]

Originally Posted by brocks

Bonuses for wins and majors are already built into the system. It's very heavily top-weighted. If you get 100 points for a win (which you do for a major), you only get 60 points for second, 40 for third, on down to 14 for 10th, and 3 for 45th. And in the weakest PGA events, you get less than 30 points for a win, so winning a major is better than winning three Reno-Tahoes. You can get as little as 6 points for winning on the Canadian or South American tour, so you'd need to win 16 of those in a row to equal one major, and even then you would already be losing points on your first few wins from depreciation.

You can get a complete explanation and downloadable point chart here:

http://www.officialworldgolfranking.com/about_us/default.sps?iType=425

I get that.  My comment was that if the points trail only lasts for one year, then maybe wins contribute to your ranking a little longer.  Like if Rory makes it to August without winning, his US Open win would drop off (under a modified OWGR system).  It seems like that win should contribute to his OWGR for more than a year.  As I type it, I realize that it doesn't make a lot of sense.

[brocks]

Do they split the points like they do with prize money? I mean if 5 guys tied for second place, don't they split 2nd through 6th place money, then the next guy gets 7th place money?

Yes, they do it exactly like that. [quote] Does a guy finishing T2 three times get only 20 fewer points than a guy with 2 wins and a MC in those same events? [/quote] Assuming you mean solo second rather than T2, and assuming all three events are majors, yes. For non-majors, the point differential would be less, in proportion to how many points the winner got. For example, if it's a 40-point event (about the same strength of field as the Sony Open), then solo second gets 24 points (60% of what the winner gets), so a guy with three solo seconds would get 72 points, and a guy with two wins and an MC would get 80. That may not seem like a huge premium for winning, but the difference between first and second might be just one shot on the fifth playoff hole, while the difference between solo second and MC is usually pretty big.

[MEfree]

Originally Posted by Zeph

I've yet to see someone come up with a better system than the current one.

Originally Posted by k-troop

It's definitely absurd that points for a win can be depreciated to the extent that, after about 90 weeks or so, they bring your 2-year average down.

As I have stated in other threads and to members of the OWGR, I agree with you 100% AND IT IS VERY EASY TO FIX THIS PROBLEM WITHOUT ANY FUNDAMENTAL CHANGE TO THE CURRENT OWGR SYSTEM! Simply use the sum of the event weights as your numerator instead of the number of events played.  This is how weighted average is computed mathematically.

Here is an example using 3 hypothetical events where a player earned 10 points in each with the first event counting full value, the second event counting 50% and the 3rd event counting 10%.

Event	Points	Weighting	Weighted Points
1	10	1	10
2	10	0.5	5
3	10	0.1	1
		OWGR Avearge Points	16/3.0 =  5.33
		Weighted Average Points	16/1.6 = 10.0

If the OWGR Committee adopted this approach, there were be very little change to the current rankings and it would eliminate the bias against players who played more events towards the older part of the 2 year period.  The OWGR could still have a minimum numerator (I`d say 22.5 instead of the current 40 to be in line with the way the points are currently depreciated) to avoid a player doing well in an event or two and then hiding to protect his ranking.  There is no downside to this approach other than the OWGR Committee having to admit there is a mathematically better way of doing things.

[newtogolf]

It's actually a good point.  If a guy is playing poorly he'd do better in the rankings to withdraw rather than play and not make the cut or finish poorly.

Originally Posted by bplewis24

I find this whole thing hilarious.  It changes every two weeks while one guy is sitting at home sipping tea.

Brandon

[brocks]

If the OWGR Committee adopted this approach, there were be very little change to the current rankings and it would eliminate the bias against players who played more events towards the older part of the 2 year period.  The OWGR could still have a minimum numerator (I`d say 22.5 instead of the current 40 to be in line with the way the points are currently depreciated) to avoid a player doing well in an event or two and then hiding to protect his ranking.  There is no downside to this approach other than the OWGR Committee having to admit there is a mathematically better way of doing things.

But why go through all that, if your goal is to let him keep his 10 point average? It would be much simpler, and achieve the same goal, if you just eliminated depreciation altogether. IMO you do need to have bias against old events, because winning last year's Masters is not as good an indication of a player's current game as winning this year's Masters. But the depreciation should stop at no less than 30%, to prevent situations where an old win actually hurts a player's average.

[brocks]

It's actually a good point.  If a guy is playing poorly he'd do better in the rankings to withdraw rather than play and not make the cut or finish poorly.

For WGR calculations, a WD is the same as a MC.

[newtogolf]

You're right I realized after I hit enter it would be the same, it would just be better for him not to enter a tournament at all.

Originally Posted by brocks

For WGR calculations, a WD is the same as a MC.

[brocks]

You're right I realized after I hit enter it would be the same, it would just be better for him not to enter a tournament at all.

But every week you don't play, you lose points to depreciation. And if you skip too many events, the minimum divisor kicks in. Your total points are divided by a minimum of 40 events, even if you've only played 30. Tiger currently has only 33 events in his two-year window. Believe it or not, he would be ranked 4th if they didn't use a minimum divisor.

[Brett]

some things just don't lend themselves to quantification....

[MEfree]

Originally Posted by brocks

But why go through all that, if your goal is to let him keep his 10 point average? It would be much simpler, and achieve the same goal, if you just eliminated depreciation altogether.

IMO you do need to have bias against old events, because winning last year's Masters is not as good an indication of a player's current game as winning this year's Masters. But the depreciation should stop at no less than 30%, to prevent situations where an old win actually hurts a player's average.

I think you misunderstood how weighted average works because I showed 10 points for all the events.  Here is a better example involving Player A and Player B who both earned a total of 40  points (pre weighting) in 3 events but with A having better results as his 20 point performance came in the most recent event while B`s 20 point performance came in the oldest event.  Notice that A has a higher average in both computations but that the OWGR "Average" is actually less than the lowest event points earned by either player.

Event	Weighting	Player A Pts	A Weighted Pts	Player B Pts	B Weighted Pts
1	1	20	20	10	10
2	0.5	10	5	10	5
3	0.1	10	1	20	2
Sum	1.6	40	26	40	17
	OWGR Avearge Points		8.67		5.67
	Weighted Average Points		16.25		10.63

[newtogolf]

True, but short term (to avoid minimum divisor kicking in) the points lost to depreciation could have less of an impact than a MC or poor finish?

Originally Posted by brocks

But every week you don't play, you lose points to depreciation. And if you skip too many events, the minimum divisor kicks in. Your total points are divided by a minimum of 40 events, even if you've only played 30.

Tiger currently has only 33 events in his two-year window. Believe it or not, he would be ranked 4th if they didn't use a minimum divisor.

[brocks]

True, but short term (to avoid minimum divisor kicking in) the points lost to depreciation could have less of an impact than a MC or poor finish?

Correct, a poor finish will always hurt a top golfer more, because the depreciation occurs whether he plays or not. The one exception is for players who are already below the minimum divisor (and unless I missed one, Tiger is the only player in the top 80 who is in that situation). For Tiger, there is no downside to playing, because any points he earns will increase his total, but his divisor won't go up until he's played over 40 events. Of course, if he does get up to 40 events, then those weak finishes will start hurting him.