Thursday, December 8, 2011

Do World Rankings Really Tell Us Anything?

After my remark yesterday that Tiger only gained one point in his OWGR leap from #52 to #21 and so it wasn't really a big deal, I found myself wondering if the world rankings really mean that much. Let me show you what I mean.

The OWGR for this week lists 998 players. The last man on the list, Mark Williams of ZAF (which I assume is Zambia in South Africa) has a grand points average of .07. That's right, seven one-hundredths of a point. No player below #865 has more than one-tenth of a point, #595 has a quarter of a point, and #384 has a half point. Soren Kjeldsen of Denmark, sitting at #188, has .99 points and Benjamin Hebert of France finally breaks the single-point barrier at #187. He has 1.01 points.

That means the bottom 810 players in the ranking don't even have a single point. That's over 80% of the OWGR! Pardon me for being dense, but does this really tell us anything about how good these players are?

At the other end of the scale, world #10 Matt Kuchar has 5.00 points and #1 Luke Donald has 10.20 points. The Top 10 in the rankings are separated by more points (5.20) than the gap between #10 and #998 (4.93)!

A few other notable points milestones:
  • 4.00 -- Sergio Garcia, #17
  • 3.00 -- Miguel Angel Jimenez, #41
  • 2.00 -- Anthony Kim, #74
Every year at the Accenture Match Play, which takes the top 64 from the OWGR, players say that there isn't really that much difference between #1 and #64. And yet the only measurable differences seem to be among the top handful of players. If that's the case, then why bother with the OWGR at all?

What other options do we have? Most of the "measuring sticks" I can think of aren't much better:
  • If we just count total majors, we can only assess a few players. After all, there are only 4 majors each year.
  • Add the TPC, the BMW, and the WGCs and we're only up to 10 tourneys a year... and some of those don't have particularly long histories.
  • Total wins adds another layer, but how do we rate players who have never won (which is most of them)? And how do we equate wins across tours?
  • And each of these are cumulative over years. The big complaint about the OWGR now is that 2 years is too long for a ranking period.
Do I have a solution to this conundrum? No, I guess not. But it makes me wonder -- is there really any accurate way to determine the best players in golf?

Still, I guess anything is better than the BCS. They might as well draw numbers from a hat. ;-)

No comments:

Post a Comment