But sometimes the current ACBL policy is justified. I think the ACBL Board of Directors should be willing to explain and justify all of their policy, which of course they can't. But it puzzles me why they don't even justify the parts that can be justified.
Anyway, these are answers to issues people have raised in their letters published in the ACBL Bulletin.
How would it be implemented? The idea is that there should be some cut-off for performance, and scores below this cut-off do not deserve awards. Assume a club game for now.
A Cut-Off for A Awards. This sounds like a suggestion from an A player. For some odd reason, B and C players rarely suggest ways their points should be reduced. In any case, if it was right to have a cut-off, there should be a cut-off for the A awards too.
What should be the Cut-off? It does not make sense for the cut-off to be the same for both B and C players. Simply put, a 50% game by a C player is, comparatively speaking, better than a 50% game by a B player. Following this logic, the cut-off for the A awards should be even higher. For example, if the cut-off was 50% for the B awards, it might be 53% for the A awards and 47% for the C awards.
I looked at 18 club games. Note that at least 40% of the players finish high enough to receive an A award. Therefore, the lowest A award is often close to 50%. In fact, the lowest was 50%, but half of the lowest scores were under 52%.
The percentage of players receiving awards is purely a judgment. You could question that judgment -- perhaps you think only 30% of the players should receive masterpoints. But to add a criterion of 52% would be incompatible with the current system or awarding 40%. If the criterion was 51%, six of the scores would have been eliminated, which still seems large, though reasonable. If the cut-off was 50%, that would rarely take away any masterpoints.
If we adopt a 50% cut-off for the A players, it seems we should also adopt a lax cut-off for the B and C players, so that their awards too are rarely eliminated. And if you do that, there is little point in having the policy. If you adopt a 51% cut-off for the A awards, then corresponding cut-offs for the B and C awards might be around 48% and 46%.
Evaluation. What would be gained by adding these cut-offs? To me, it is part of the excitement that sometimes I get an award even though I have a 51% game. The idea of basing masterpoints on both rank and score can be extended to say that people do not get first-place awards for games under 60%. But part of the excitement is that I sometimes win with a game under 60%.
So, establishing a cut-off percentage for awards is justifiable, in terms of making awards more proportional to merit. But there would not seem to be substantial gains. Meanwhile, we would be tampering with some of the game, and we would be tampering with the basic idea that points are based on ranking, not percentage. There would also be difficult choices as to the cut-offs; we might sympathize with the Board of Directors in not wanting to make those choices.
Empathy. I hope I have helped A players see how the B and C players might view this proposal. If you are an A player, do you like it when this principle is applied to the A awards? The whole point of stratification is that it is easier for the B and C players to win awards. And these are small awards. A 48% game by a C player is not outstanding, but it is a credible performance, and it deserves a chance to win a (small) masterpoint award.
Tournaments. The situation is trickier for tournaments. The club formula punishes small fields. So in a stratified club game, the C players are in essence "punished" by the fact that their field -- the C players -- is small. It looks to me like the other factors balance this out well. The tournament formula rewards small fields, so for section awards, the C players are rewarded by their field shrinking in size. Furthermore, the tournament formula gives a large percentage of people awards when the field is small. That means that the top C players can win B awards and yet there will still be awards for the C players. I have no data on tournament section awards.
Tournament Overalls. For tournament overalls, the above considerations do not as strongly apply and the situation doesn't look bad. I looked at one stratified pairs in a regional. The lowest C award went to a 48.26% game. That might not seem very high -- the lowest A award when to a 60.19% game. But the highest C pair was 51.27%. The point -- 48% is credibly good for a C pair.
I also looked at a stratiflighted Open Pairs at a Regional. The B/C/D players played among themselves. 51.27% won an overall award of 1.37 for the D players. Meanwhile, a 53.27% game won an overall of 2.93 for the A players.
I think the formulas should be fixed to that masterpoint expectations do not change with field size. When they do, the landscape will change. But I think the idea of cut-offs is not warranted. It isn't a bad idea, and if it existed, we probably wouldn't want to change. But I do not think it would be worth the time and effort of change.
The ACBL is a large, successful organization. Large successful organizations tend to be conservative, for good reason. I suspect the ACBL is simply being conservative -- it is afraid to offer regular points over the internet, because internet points are new.
Being conservative means worrying that there are dangers you don't know about. Let's think about some of dangers. One is that an internet game can easily draw 120 pairs, making it much larger than most (if not all) club games. Ideally, masterpoints per person would not increase with field size. In reality, it does. The question is, how much will it increase? I don't know. But if masterpoints per person increases substantially, then the internet games will draw even more competitors and fields of a 1000, or even more, are possible.
You address the issue of cheating. The ACBL does not understand the issue of cheating on the internet, because the internet is new. It takes time to understand any issue. Therefore, it is right to worry about this issue. And if the award for first place is very large -- I do not know the award for first overall in a 500-table club game -- what does that do for the incentive to cheat? Another problem is that most cheating at ACBL tournaments is caught by the efforts of players and the ACBL. It is not clear how either players or the ACBL could catch cheating on the internet. I suspect that there are ways of doing it, but that they are not fully developed and they are done by the internet company running the game.
I do not understand the issue of number of hands. In theory, you could have a 2-board game and award masterpoints. Doing that, you could finish a lot games in a short period of time. It would be expensive, but your expected points per hour would be very high. Presumably the ACBL does not allow that, but I am not sure exactly what the ACBL does.
And looking behind the scenes, how much does the ACBL charge for internet games. If it is less than the club rate, the ACBL cannot offer the same masterpoints to internet games as it does for club games.
There is still one more point. In my opinion, a club game is much more enjoyable than an internet game. Before the ACBL grants the internet equal status to club games, it would want to make sure that that had no impact on the interest, enjoyment, and participation in bridge.
This gets into the issue of reliability/validity. To the extent that luck is eliminated, skill tends to predominate. That's what we want. We probably don't want luck to disappear completely -- then the results would always be the same. But we usually want to eliminate luck as much as possible.
The problem with not duplicating the boards across sections is that there is more luck involved. Some boards allow a team to exert their skill; some boards have little skill involved; and some boards actually punish skill. And from another perspective, some boards are flat and some are not. These factors start to average out over 24 boards, but my sense is that they do not come close to actually being even. So, when boards are not duplicated across sections, you have the problem that you might get flat boards and hence no chance of winning (bad luck) or swing boards and hence a chance of winning (good luck). Of course, this is just one luck factor; there are many others.
The ACBL, as far as I know, does not use masterpoints to reward the reduction of luck or to punish gratuitously increasing the effect of luck. Therefore, it would not be ACBL policy to award more masterpoints to a STaC that duplicates boards across sections than to a STaC that does not.
Maybe the ACBL should change its policy. But there are a lot of different ways that luck can be increased or reduced. None of the effects are presumably very large, and my guess it would be enormously tedious to try to consider them. (For more information see my essays on reliability.)
The ACBL does encourage directors to reduce the effect of luck as much as possible. It is not clear how much the ACBL can actually enforce. In this case, perhaps the ACBL should more actively encourage the duplication of boards across sections. But sometimes, for whatever reason, that doesn't get done. That is unfortunate, but it is not a tragedy -- it is just one more luck factor in a game already filled with luck factors.
Furthermore, there is nothing to be done about the fact that N-S plays different hands than E-W. If the E-W hands require skill and the N-S hands are either watching or making decisions requiring luck (e.g., whether or not to bid a 50% game), the winners will tend to be from the E-W direction.
The problem you mention is probably more severe for Swiss Team events scored on Victory Points. The fact is, some boards are swing boards and some are flat; if you have flat boards, you will not earn a large number of Victory Points. So it would be nice if boards were duplicated in a Swiss Teams. As you might imagine, the difficulty of doing that probably exceeds any advantage to be gained.