[han]

jdonn, on Dec 10 2008, 01:19 PM, said:

platinum points earned
to
total platinum points given away in NABC events participated in

Instead of the total number of points given away it should really be the sum of the average number of platinum points given away in those events. Or more precisely for each event you enter you get the number of platinum points earned divided by the average number of platinum points earned in that event, and then you take the average over all the events you enter.

I think you'll agree that this is right, or at least close to right.

These are of course theoretical discussions because the ACBL does not seem interested in finding out how good somebody is. For the best players (used here in a very broad sense) it doesn't matter, the restriction that is currrently proposed is trivial for them. For those that struggle to win platinum points the restriction is such that those that enter lots of NABC events will have a much better chance to qualify for the platinum pairs. It seems to me that this is exactly how the ACBL wants it.

I also dislike the lack of open national events on some days of the NABC's in the proposed schedule.

[jdonn]

han, on Dec 10 2008, 01:44 PM, said:

jdonn, on Dec 10 2008, 01:19 PM, said:

platinum points earned
to
total platinum points given away in NABC events participated in

Instead of the total number of points given away it should really be the sum of the average number of platinum points given away in those events. Or more precisely for each event you enter you get the number of platinum points earned divided by the average number of platinum points earned in that event, and then you take the average over all the events you enter.

I think you'll agree that this is right, or at least close to right.

I thought the amount of points given out was already a function of how many people played? But if that's not the case then of course I agree.

[peachy]

jdonn, on Dec 10 2008, 01:59 PM, said:

han, on Dec 10 2008, 01:44 PM, said:

jdonn, on Dec 10 2008, 01:19 PM, said:

platinum points earned
to
total platinum points given away in NABC events participated in

Instead of the total number of points given away it should really be the sum of the average number of platinum points given away in those events. Or more precisely for each event you enter you get the number of platinum points earned divided by the average number of platinum points earned in that event, and then you take the average over all the events you enter.

I think you'll agree that this is right, or at least close to right.

I thought the amount of points given out was already a function of how many people played? But if that's not the case then of course I agree.

If there were a formula such as this, or something similar, then all the masterpoints a person holds could be calculated on the ratio of masterpoints available versus the masterpoints the person earned. It would then be easy to separate achievers from those who have masterpoints [or platinum, or whatever color] by "decades of attendance". The whole idea of total masterpoints, even total platinum points, is not worth much as an indication of success.

[han]

No but it works fine as an incentive to play more ACBL games which is what they are for.

[TimG]

jdonn, on Dec 10 2008, 01:59 PM, said:

I thought the amount of points given out was already a function of how many people played? But if that's not the case then of course I agree.

I'm pretty sure that all the NABC events have a fixed masterpoint schedule for the overalls, independent of size of field (though the number of places paid may vary by size of field).

Of course, you are right that how many total points are given out varies with size of field and the resulting number of sections and section awards.

But, should it matter that your section top was one section top out of 5 available or 6 available?

[han]

Maybe not, but when there are 6 entries the total number of masterpoints will be higher so when you divide by that you actually get less for winning when there are 6 contestants then when you win with 5 contestants. Clearly that can't be right.

[jdonn]

So we have

PP earned / (Total PP available / Total players)
=
(PP earned * Total players) / Total PP available

Which accounts for that problem han. Maybe it's simply irrelevant that

Total PP available = f(Total players)

for each particular event, since that function is not linear which I think would be a desirable quality for the formula we are trying to come up with.

Then instead of choosing an artitrary threshhold to meet, perhaps they should simply make it available to X number of players. (Yes, I know this would never fly in reality.) Either "the top X that want to enter can play" or "the top Y are eligible to enter, and the event will simply have whomever of that group chooses to play."

Now that I think about it, that last suggestion would only be good with an aging function applied to it, kind of like they do with seeding points in the Spingold. The last 10 years are mulitplied by 1, .9, .8, etc.

[TimG]

Shouldn't you just make it total points available to a single pair? Every pair in the event has the potential to win whatever 1st place is worth. 1st place award does not vary with number of participants.

[jdonn]

TimG, on Dec 10 2008, 02:59 PM, said:

Shouldn't you just make it total points available to a single pair?

I don't think so. Maybe the amount of places that get paid should be incorporated though. But now I just don't feel like working it out.

Actually that makes sense. Now I like

PP earned / ((Total PP available/Total places paid) / Total entrants)
=
(PP earned * Total places paid) / (Total PP available * Total entrants)
=
(PP earned / Total PP available) * (Total places paid / Total entrants)

The function in that last form looks very logical to me. It would account for something like this. Take two events with 10 pairs, which use these scales to award platinum points.
Event A: 100, 75, 50, 25
Event B: 70, 50, 20, 10
The way I had it before, a pair coming in 3rd in event A would get the same seeding as a pair coming 2nd in event B, which doesn't seem right. But this way a pair coming 3rd in event A earns (50/250)(4/10) = .08, whereas a pair coming 2nd in event B earns (50/150)(4/10) = .13, which seems right.

Hmm but then the problem is you get just .12 for coming second in event A. I guess that's because I picked a bad example by making 2nd worth a higher % of the available PPs in event B than in event A, which probably wouldn't happen in real life events of the exact same size. I think.

[TimG]

Isn't 2nd 75% of 1st, 3rd 75% of 2nd? After that the % goes up, but it follows a set progression regardless of field size (I think).

Your event A would be 100, 75, 56.25, 44.44.
Your event B would be 70, 52.5, 39.37, 31.11.

Not so sure about this, but isn't "total places paid" simply a percentage of "total entrants" so that (total places paid)/(total entrants) is a constant?

Anyway, if you use my method of (masterpoints won)/(masterpoints available to individual), the first four places (in both events would be): 1.00, 0.75, 0.56, 0.44.

A quick look at a couple of events from Boston shows that the last place overall would be worth in the neighborhood of 0.07-0.10.

[TimG]

jdonn, on Dec 10 2008, 03:07 PM, said:

(PP earned / Total PP available) * (Total places paid / Total entrants)

Doesn't this mean that as the event gets bigger and bigger and more PP are available, winning the event (a fixed award) is worth less and less. At least assuming that (Total places paid / Total entrants) is constant or nearly constant.

[cherdano]

Do you really want to make someone ineligible for the Platinum pairs because besides doing well twice with his regular partner, she also entered the LMP twice with a decent advanced partner and made it to day two, but not further?
Coming up with a rating system isn't exactly what you should try to do while posting on BBF from work

[jdonn]

cherdano, on Dec 10 2008, 04:13 PM, said:

Do you really want to make someone ineligible for the Platinum pairs because besides doing well twice with his regular partner, she also entered the LMP twice with a decent advanced partner and made it to day two, but not further?
Coming up with a rating system isn't exactly what you should try to do while posting on BBF from work

I admit it was just for sh!ts and giggles.

[jdonn]

TimG, on Dec 10 2008, 04:06 PM, said:

jdonn, on Dec 10 2008, 03:07 PM, said:

(PP earned / Total PP available) * (Total places paid / Total entrants)

Doesn't this mean that as the event gets bigger and bigger and more PP are available, winning the event (a fixed award) is worth less and less. At least assuming that (Total places paid / Total entrants) is constant or nearly constant.

True, but I don't think it should be a fixed award. One day the world will go the way I want it to.

[kenrexford]

maggieb, on Dec 8 2008, 01:13 AM, said:

TimG, on Dec 7 2008, 09:41 PM, said:

it means that a partnership made up of two males under 55 won't have any NABC events to enter those two days.

Come on, is it really that hard to play with a woman for a few hours?

One year, I played in the Mixed Pairs with a lady who went completely off the deep end, almost getting us thrown out of the event, having multiple 7-minute discussions with the TD's, and threatening to send a bomb in the mail to one pair we played against. We scored up about a 35% in the first set, and then the bomb threat set kicked in.

So,

As the one and only time I ever played in the National Mixed Pairs ended with me being on a watch list as an associate of a suspected domestic terrorist, I'd say, "Yes," to that question. LOL

[TimG]

kenrexford, on Dec 10 2008, 06:02 PM, said:

maggieb, on Dec 8 2008, 01:13 AM, said:

TimG, on Dec 7 2008, 09:41 PM, said:

it means that a partnership made up of two males under 55 won't have any NABC events to enter those two days.

Come on, is it really that hard to play with a woman for a few hours?

One year, I played in the Mixed Pairs with a lady who went completely off the deep end, almost getting us thrown out of the event, having multiple 7-minute discussions with the TD's, and threatening to send a bomb in the mail to one pair we played against. We scored up about a 35% in the first set, and then the bomb threat set kicked in.

So,

As the one and only time I ever played in the National Mixed Pairs ended with me being on a watch list as an associate of a suspected domestic terrorist, I'd say, "Yes," to that question. LOL

Your wife?

[xcurt]

awm, on Dec 10 2008, 01:30 PM, said:

I think the point of platinum pairs is to restrict the field to "people who usually qualify for day two of national events" and not to restrict the field to "people who usually place in the moderately high overalls of national events." Obviously these are different goals; the latter is a stronger (but much smaller) field.

Just for fun I checked my PP holdings. I didn't think I had any since I last played a NABC in 2001. Guess what, I have almost 5 PPs from session awards (and it would be more except that excludes the session awards I got in events where I made the overalls).

han, on Dec 10 2008, 01:44 PM, said:

These are of course theoretical discussions because the ACBL does not seem interested in finding out how good somebody is. For the best players (used here in a very broad sense) it doesn't matter, the restriction that is currrently proposed is trivial for them. For those that struggle to win platinum points the restriction is such that those that enter lots of NABC events will have a much better chance to qualify for the platinum pairs. It seems to me that this is exactly how the ACBL wants it.

Exactly.

I'll repeat my earlier assertion: The ACBL, having so devalued their existing currency the bridgeo^Hgold masterpoint, is trying to replace it with the new bridgeo^H^Hplatinum masterpoint. However, they are trapped by their need to prevent a revolt of the peasantry^Hrank and file so they aren't really able to devalue it enough.[1] Any number of textbooks on monetary policy will tell you an insufficient devaluation is bad policy, as they need to devalue at least enough to regain central bank^H^HACBL BoD credibility.

[1] Just to illustrate this assertion, I'm more than a quarter of the way to a lifetime exemption after only two NABCs (all my other NABCs predate the existence of platinum points). I consider myself a reasonably good player, but when the PPP was proposed I certainly envisioned something a little scarier than a field of clones of me.

[barmar]

I think you do yourself a disservice. If you regularly make it into the overalls in NABC+ events, you're a very good player.

What are you expecting, a two-section game containing just national and world champions? This is not intended to be as exclusive as an invitational event like that. My guess is that it will be about 1/3 to 1/2 the size of the Blue Ribbon Pairs. As someone else remarked, the initial field will probably look much like the 2nd day of the BRP, and by the 3rd day it will be down to the real cream of the crop.

[stacy]

barmar, on Dec 12 2008, 02:51 AM, said:

What are you expecting, a two-section game containing just national and world champions? 

Actually, yes. I'd love to watch an event like that.

[jdonn]

stacy, on Dec 12 2008, 08:54 AM, said:

barmar, on Dec 12 2008, 02:51 AM, said:

What are you expecting, a two-section game containing just national and world champions? 

Actually, yes. I'd love to watch an event like that.

Sounds a lot like the Cavendish, except not quite as exclusive.