[fred]

bd71, on Oct 31 2009, 12:59 AM, said:

fred, on Oct 30 2009, 03:54 PM, said:

6) The possibility of cheating through communication with partner via phone or MSN or similar is not present in these tournaments. The possibility of easy cheating is one of the traditional objections that some ACBL officials have had to issuing masterpoints for online bridge. One would think that Robot Duplicates would alleviate some of these fears.

Fred, I almost included a comment similar to this in the message to my district director which I sent you. But as I was thinking about it, I'm not sure it holds water. I doubt that the ACBL board is specifically worried about "cheating through communication with partner via phone or MSN". Rather, they are worried about cheating, period. And cheating in the ACBL robot games would still be relatively easy if you had two accounts and previewed some hands with one account so that you could bid or play defense double dummy with the other account.

Regardless of the cheating threat, I think these tournaments are excellent and should be ACBL-sanctioned, so I want to be sure you have the strongest set of arguments. I fear that this specific argument is relatively easy to rebut, and would be better left out. Instead, recommend you marshal the best empirical evidence you have on why you think cheating, while possible, is minimal in these tournaments.

Best of luck.

Thanks - you make a very good point.

Cheating is one subject that I do not like to talk about at all and, when I do, I normally try to be very careful with my words. I did not do a good job of that in my previous post. In retrospect, probably I should not have mentioned cheating at all.

I don't want to make things any worse, but let me just say there is more to this subject than meets the eye (or perhaps even the imagination given some of the unimaginable things that go on). For example, the subject of detection of cheating is arguably at least as important as that of ease/methods of cheating.

I am sure you can appreciate that I would rather not have a detailed public discussion on any of these subjects. But the unfortunate reality of the situation is that I won't be able to avoid talking about this in private (to the ACBL BOD). These are subjects that both BBO and the ACBL need to care about. This is not something that we can simply sweep under the rug.

I do appreciate your advice and your concern. I will certainly be careful about what I choose to say during my presentation.

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[TimG]

fred, on Oct 30 2009, 03:54 PM, said:

For example, the text of the motion itself suggests that the robots play very poorly. This is not true.

I think you should enter a robot in every robot tournament. This would at least give an indication of how well the robots play relative to the players entering the robot tournaments.

[gwnn]

I think the robot would get more than 50% from the limited experience I acquired playing with them. After finishing when I kib most (60+% instead of the a priori 25%) of the retarded things that I see definitely come from the human. This is assuming I kib a random player.

[mike777]

I think the robot motion ( 093-36) is important.



I think from a marketing viewpoint motion (093-34) is vastly more important and in fact being overlooked.


If I was a part owner of BBO I would gladly lose the battle over 093-36 to win 093-34. I can always come back and refight.....-36.


If online points count the same as f2f points, that is the Golden Goose.

[JoAnneM]

Hi Fred, I initially came to the robot tournaments because a friend told me about them, and she was outraged that people were earning masterpoints playing with robots. I decided to check it before forming an opinion.

Well, I was smitten immediately. It took me several games to form a partnership with the GIB, but that is the case with any new partner. And since GIB can't change I am just adjusting to him/her. I enjoy the games so much, and they are not easy.

My husband is not a bridge player and we have an active life outside of my bridge games and the club I own and manage, so it is nice to be able to sit down and play whenever I can fit it in, and not worry about finding a compatible partner.

I am going to write you a separate email. I am going to ask our District Director to vote to continue these tournaments, but if something happens between now and the meeting and he can't attend, since I am First Alternate District Director you certainly have my vote.

Jo Anne Murdock
District 20
First Alternate

[Jlall]

gwnn, on Oct 30 2009, 08:51 PM, said:

I think the robot would get more than 50% from the limited experience I acquired playing with them. After finishing when I kib most (60+% instead of the a priori 25%) of the retarded things that I see definitely come from the human. This is assuming I kib a random player.

Yeah it would average well over 50 % and it's not even close.

[barmar]

I definitely enjoy the Robot Duplicates, although I like Robot Rewards more (they're more intense because of having to play quickly, I can pass out minimum hands so I don't waste time on part scores as much, and the prospect of winning $$ is exciting). I don't particularly care about the masterpoints; the main reason I play the ACBL Robot Duplicates is because they're 12 boards instead of just 6, and I think they attract better competition than the BBO Robot Dupes. But I don't see a problem with awarding this masterpoints, for the reasons others have given: all the human players are in the same seat, so we're competing with each other, not the bots.

[shyams]

Fred,

So far all responses appear to be from ACBL members. So I thought I will share my views...

I am based in the UK. I am not an ACBL member and am unlikely to become a member in the future. I guess there are loads of people playing ACBL robot tourneys like me viz. non-members (now and in future)

Yet, I play robot MP tourneys on BBO because I like the higher standards of competitors (vs. other tourneys). I think the ACBL robot MP tourneys actually help me become a better player.

In summary, I contribute to ACBL revenues (albeit a small amount) without demanding any services in return. I do so because I strongly believe:

1. The robots play as good as (often better than) many humans in ACBL online tourneys
2. The ACBL MP tourney on BBO generally attracts a high standard of players
3. Robots dont mind if I take a mini-break for the occasional "real-world" disturbance (e.g. 5 min phone call). In contrast, I have had humans (ACBL members) being rude when I was away from my seat for 20-30 seconds!

I have no problem paying ACBL for robot-MP tourneys; I find it odd they have a problem instead!

[hrothgar]

Fred / Uday / et al.

I decided to spend a bit of time playing around with the Robot Races this morning in the hopes that I might come up with some better informed commentary. Here are a few – somewhat random – thoughts. Hope that some of this is useful

First and foremost, the Robot Races felt fair. When I play a hand, I can generally tell “Wow, I just did something really stupid” or, more rarely, I just did something clever. When I look at more scores – bye and large – I seem to score poorly on hands where I blundered. I score well on hands where I think that I did something decent. I scored about even on hands where that felt down the middle. Associated with this: You best chance to convince members of the ACBL BoD that the Robot Races are real bridge/a real test of skill is to make sure that said members are able to participate in one or more events. I think that everyone would benefit if the audience for your presentation was as well informed as possible. I recommend trying to hold one or more tournaments specifically for BoD members. (Alternatively, set them all up with their own accounts and allow them to compete for free between now and the BoD meeting).

Next: From what I can tell, the overwhelming majority of ACBL sponsoring Robot Races are rigged so that South (the player) automatically has the strongest hand at the table. I have some very real problems with this. I might be a stick-in-the-mud, but you're introducing a lot of unnecessary distortions. When I am bidding / declaring a hand, I know with certainty that the other three pairs CAN'T have a better hand than me. This can significantly distort both the bidding and the play. When you are trying to hash this all out with the ACBL one of the broad themes that you need to advance is that the ACBL shouldn't discriminate against the Robot Races when awarding masterpoints because RR and F2F are both “bridge”. Introducing unnecessary deviations like always giving South the strongest hand dilutes this line of argument. (For what its worth, I understand that your customer base probably prefers that they always get the strongest hand. Simply put, I don't believe that the customer is always right)

This might seem hypocritical, however, I'm now going to suggest a much more significant change to the game. I don't think that players should ever be dummy.

* GIB is non deterministic
* Allowing players to play out all hands where their side declares (should) more closely correlate their scores with their performance
* The GIBs play the hands WAY too fast. On my PC the cards flew by too quick to see. This was a very jarring reminder that this isn't F2F bridge.

[JoAnneM]

When you join the tournament it states that South is going to be dealt the best hands.

[jdonn]

I would want to keep south with the best hand (I think this reduces variance relative to skill level by a lot, not to mention most people probably find it the most fun that way), but it would certainly be a good idea to rotate the hands and have the human declare whenever he would have been dummy (which would reduce that variance even further).

[hrothgar]

JoAnneM, on Nov 1 2009, 09:43 PM, said:

When you join the tournament it states that South is going to be dealt the best hands.

I'm well aware of this.

I was objecting to the structure and prevalence of this type of tournament, not the advertising.

[helene_t]

I think the argument "one should only get masterpoints for playing against humans" is flawed.

Robot duplicates provide a very pure comparison between humans: they all sit south, they all not only have identical cards but also identical partners and opps. None are advantaged by over-ethical opps and unethical partners, or disadvantaged by the reverse. None are advantaged by encountering the weak-NT opps at the boards where opps don't have weak balanced hands, because all opps play the same system.

I don't believe any f2f tourney format can provide an equally accurate assessment of player skills.

[JoAnneM]

I agree with jdonn. I often cringe when watching the GIB declare a hand. I would rather the GIBs cringe while I am playing it.

[Phil]

hrothgar, on Nov 1 2009, 08:05 AM, said:

* Allowing players to play out all hands where their side declares (should) more closely correlate their scores with their performance

This is a feature I've always wanted. I also agree it distorts the bidding when you know another hand can't be better than you.

Perhaps if the tourneys were 15 boards, that would make up for the pushes when the GIBs are in the bidding alone, and the length of time would take about the same.

[Vampyr]

I have never played in a robot tournament. It sounds like a lot of people enjoy them. So, if it is as much fun as people here are saying, would you/they really play less if they did not award ACBL masterpoints? If people want to collect masterpoints, maybe BBO should introduce their own.

[TylerE]

They do. And there are robot tournements that award them. And they get maybe 1/5th the business as the ACBL ones, despite identical conditions, so clearly a large part of the customer base for those tournaments IS playing at least in part for the masterpoints.

[barmar]

TylerE, on Nov 2 2009, 12:43 AM, said:

They do. And there are robot tournements that award them. And they get maybe 1/5th the business as the ACBL ones, despite identical conditions, so clearly a large part of the customer base for those tournaments IS playing at least in part for the masterpoints.

The conditions are not identical. The ACBL Robot Duplicates are 12 boards, the BBO Robot Duplicates are only 6 boards. Or has this changed?

[3for3]

I am not sure the notion that you are guaranteed the best hand is real bridge. It probably doesn't matter often, but can't one take inference about the opponents hands knowing you have as many or more HCP as they do?

As an alternative, just guarantee that the player gets 11 or more points. Then no such inference could be drawn.

Please note I have never played these events, just wondering

Danny

[barmar]

While that's certainly a legitimate objection, I'll bet the people who proposed the motion weren't even aware of that detail. The motion refers to all games with computer participants, not just ones where the deals are cooked in some way.

Yes, you can take inferences, but so can everyone else, so it's fair as long as the guarantee is made clear to all players.

Fred, if the motion were amended to only prohibit games where the hands were dealt non-randomly, would that be acceptable?