[Zelandakh]

 661_Pete, on 2016-October-15, 16:57, said:

How long are grudges supposed to last, in BBO?

If someone is obnoxious I write a comment on their profile. A few have the comment that they are permanently banned. Life is too short to play with obnoxious opponents or partners and there is no shortage of alternative players around.

[barmar]

 Zelandakh, on 2016-October-16, 06:31, said:

If someone is obnoxious I write a comment on their profile. A few have the comment that they are permanently banned. Life is too short to play with obnoxious opponents or partners and there is no shortage of alternative players around.

Some people just mark them as enemies. HMFAG will never put you at a table with an enemy.

[Zelandakh]

 barmar, on 2016-October-17, 08:00, said:

Some people just mark them as enemies. HMFAG will never put you at a table with an enemy.

I have never used HMFAG so strategies based around that are probably inappropriate for me.

[661_Pete]

As I've said before, I stick to the Acol Club which is a relatively small band of players, most of whom are really friendly and a pleasure to play with. The same partners and opponents crop up time and time again. I couldn't afford to mark too many as 'enemies' even if I wanted to.

And I can't use HMFAG since that seems to place you in the MBC/RBC, willy-nilly. No problem though: I simply go to the Acol Club and scan the 'open' tables. Or I kibitz a game and wait for someone to leave.

Under this system, by the way, 'enemies' are not necessarily excluded from playing at your table. I know this because once someone joined my table whom I had, at the time, marked as an 'enemy' (he isn't now). You can always get up and leave if you don't like the company, of course.

[barmar]

 661_Pete, on 2016-October-17, 10:48, said:

Under this system, by the way, 'enemies' are not necessarily excluded from playing at your table. I know this because once someone joined my table whom I had, at the time, marked as an 'enemy' (he isn't now). You can always get up and leave if you don't like the company, of course.

That's true, enemies aren't excluded when people search for and join tables by hand. If you're the table host, and mark the table as requiring permission to join or sit, you'll be able to tell if the requester is one of your enemies. And the person searching for a table can tell if any of the host or players at a table is one of their enemies (they'll be shown in black in the table listing), so they can avoid joining. But neither of these works the opposite way -- if you're looking for a table, there's no warning that the host has you marked as an enemy.

[OldPlayr]

Given that HMFAG is widely seen as useless, why is BBO not interested in making even minor improvements to that function?

[barmar]

 OldPlayr, on 2016-October-20, 08:04, said:

Given that HMFAG is widely seen as useless, why is BBO not interested in making even minor improvements to that function?

It's on our to-do list. Figuring out how to improve it is the hard part.

[fred]

 OldPlayr, on 2016-October-20, 08:04, said:

Given that HMFAG is widely seen as useless, why is BBO not interested in making even minor improvements to that function?

HMFAG gets hit something like 50,000 times per day so apparently there are an awful lot of players out there who don't see it as useless. My guess is that very few of these people care about being matched with people who play the same conventions that they prefer - they just want to get to a table as quickly and painlessly as possible and start playing bridge. Presumably either they don't think or care about conventions much at all or are willing to live with making agreements via chat and/or looking at each other's profiles.

That is not to suggest that there is no point in improving HMFAG in order to make it more appealing to a wider variety of BBO members by adding some filtering options, but in its current form it is meant for very casual players (who I believe make up the vast majority of our membership).

Fred Gitelman
Bridge Base Inc.
www.bridgebase.com

[661_Pete]

 barmar, on 2016-October-18, 08:42, said:

That's true, enemies aren't excluded when people search for and join tables by hand. If you're the table host, and mark the table as requiring permission to join or sit, you'll be able to tell if the requester is one of your enemies. And the person searching for a table can tell if any of the host or players at a table is one of their enemies (they'll be shown in black in the table listing), so they can avoid joining. But neither of these works the opposite way -- if you're looking for a table, there's no warning that the host has you marked as an enemy.

I should perhaps explain that, on the occasion which I referred to, I was not actually the table host, so I'd have had no say in whether the person in question should be admitted or not.

Regarding HMFAG, it wasn't I who claimed it was 'useless': just that, as it stands, it doesn't meet my own specific preferences. If the BBO team are working on it, well and good. I'm OK with what I do, which is scanning the 'open tables' list - though it does get frustrating when you see the 'perfect' table to join - and, just as you click the empty seat, someone else leaps in before you!

[OldPlayr]

Good to learn that BBO may improve HMFAG. Clearly that is a challenging task.

My thoughts would be to start simple and improve incrementally over time. Unlikely that an attempt to make everyone happy in the first version will succeed. A good first step would be to modify user profiles to allow specification of a very general bidding system played (ACOL, SAYC, Precision, etc.). Not a lot of detail or gadgets in the first version. See how simple matching by basic bidding system works. It would go a long way toward better partner matching. Obviously, this will help only those who take the time to edit their profile to specify budding system(s) played.

Adding basic bidding systems to profiles is a big enough step for a first version. If this works and is catches on with a sufficient number of users, subsequent versions can expand and refine profiles to be more specific on gadgets and bidding options. One could envision very detailed profiles that enable partner matching, but an incremental path toward that seems more prudent and manageable.

After a long career in information technology, I recognize that coming up with suggestions is a lot easier that coding them!

[Kaitlyn S]

 661_Pete, on 2016-October-20, 16:04, said:

though it does get frustrating when you see the 'perfect' table to join - and, just as you click the empty seat, someone else leaps in before you!

It is more frustrating to see the "perfect" table to join, click on the locked empty seat, not get it, and see it still there after you refresh! This was the whole reason I started this thread, and quite honestly, the person might be waiting for a regular partner or a "friend". Still, I would be interested to know how often either my self-rating or my gender plays a part.

[Zelandakh]

I think you will find that your gender is more of an advantage than a disadvantage. Certainly my female account never had any trouble at all in finding acceptance and was always warmly received, more so than either of my main accounts.

[aguahombre]

 Zelandakh, on 2016-October-21, 05:02, said:

I think you will find that your gender is more of an advantage than a disadvantage. Certainly my female account never had any trouble at all in finding acceptance and was always warmly received, more so than either of my main accounts.

Interesting. Hope this thread doesn't degenerate into thoughts about cross-gender impersonation.

[hrothgar]

 barmar, on 2016-October-20, 08:25, said:

It's on our to-do list. Figuring out how to improve it is the hard part.

If it were me, I'd go in a very different direction: Create a permanent Indy with a built in ladder component.

Each round of the Indy lasts for some short number of boards (three or so feel right)
You automatically get paired with some random partner who is of (roughly) the same rank in the ladder.
You get matched against some other random pair of roughly the same rank.

Teams who do well move up the ladder
Teams who do poorly move down the ladder

Players who quit mid-round get banned for a day or two

From my perspective, this has a bunch of benefits

1. It provides an easy way for players to find a quick pick up game
2. Over time, players should rise or fall to their appropriate level. You should get matched against players of about the same skill
3. Players will (hopefully) find people who they enjoy playing with
4. Folks who are obsessive about ratings can use their ladder rank as a proxy

(Oh yeah, it should be pretty easy to implement)

[diana_eva]

 aguahombre, on 2016-October-22, 10:35, said:

Interesting. Hope this thread doesn't degenerate into thoughts about cross-gender impersonation.

gender impersonation thread is here: http://www.bridgebas...43-hmfag-trial/

[Kaitlyn S]

 hrothgar, on 2016-October-22, 10:52, said:

If it were me, I'd go in a very different direction: Create a permanent Indy with a built in ladder component.

Each round of the Indy lasts for some short number of boards (three or so feel right)
You automatically get paired with some random partner who is of (roughly) the same rank in the ladder.
You get matched against some other random pair of roughly the same rank.

Teams who do well move up the ladder
Teams who do poorly move down the ladder

Players who quit mid-round get banned for a day or two

From my perspective, this has a bunch of benefits

1. It provides an easy way for players to find a quick pick up game
2. Over time, players should rise or fall to their appropriate level. You should get matched against players of about the same skill
3. Players will (hopefully) find people who they enjoy playing with
4. Folks who are obsessive about ratings can use their ladder rank as a proxy

(Oh yeah, it should be pretty easy to implement)

While I would suspect in this day of P/C and everybody deserves a trophy, that this would never be implemented, I actually love this idea. For while there is a lot of merit in being able to find partners that play the same system, I think it's equally important to play with players of the same level.

I would go one step further and say that you should be able to HMFAG as a PAIR with a pair rating (which is presumably higher than your individual rating because of a pair's synergy) where you would meet tougher opposition to make up for the fact that your pair does have less misunderstandings and defensive accidents than two random players at the same level.

I could see it working better for newer players also - they are less likely to be called an "idiot" if their partner doesn't notice their mistakes. The downside is that the advice received by newer players from their peers is frequently wrong, and talented card players will learn to overbid against newbie defenses (although advancing higher in the rankings will solve that problem quickly!)

While it would be good for me to play with players at my own level, I'm not sure it would work so well for hrothgar. For most players at his level probably are coming on with established partnerships, so if he wants to play a few hands for fun, he jumps in and HMFAG's, and would be put at the table with the three best players looking AT THAT INSTANT. Oddly enough, with not that many really good players looking, it wouldn't surprise me if the three best individual players looking got down to my level. While he would avoid the really new players, or the really bad players that have been playing a long time, he would not avoid players that make mistakes that are obvious for him to avoid.

EDIT to add: It almost seemed as if he was suggesting voluntary entry into this entry. I would instead suggest using this system for all HMFAG requests, without penalty for leaving earlier than "the end of a round" since some players log into play a few hands for an undetermined time that might depends on when a spouse or S/O comes home, a call for work is received, a call from a serviceman comes, etc where it could be 1 hand or 30 depending on sheer randomness. However, if one leaves during the middle of a hand, they get the result for the hand as played by a robot or a fillin.

One drawback is that sometimes this will put the obnoxious bad player who complains when you take more than a second to play the card might be put into the Relaxed Room. I can see a few ways around it - the easiest being that people that are marked as enemies by more than 'X' people just don't go to the relaxed room (I'm assuming that you are set as an enemy for rudeness more often than for bad play.)

[hrothgar]

 Kaitlyn S, on 2016-October-22, 11:02, said:

I would go one step further and say that you should be able to HMFAG as a PAIR with a pair rating (which is presumably higher than your individual rating because of a pair's synergy) where you would meet tougher opposition to make up for the fact that your pair does have less misunderstandings and defensive accidents than two random players at the same level.

I think that the random matching is a feature

if folks can enter as a pair and you have ratings, then your going to need to deal with a lot more cheating accusations

[Kaitlyn S]

 hrothgar, on 2016-October-22, 11:26, said:

I think that the random matching is a feature

if folks can enter as a pair and you have ratings, then your going to need to deal with a lot more cheating accusations

If nobody including you knows your ratings, and the only reason for the hidden ratings is to get an equal match, I don't think there is much incentive to cheat.

[hrothgar]

 Kaitlyn S, on 2016-October-22, 12:06, said:

If nobody including you knows your ratings, and the only reason for the hidden ratings is to get an equal match, I don't think there is much incentive to cheat.

Cheaters gonna cheat (and we're going to need to listen to other folks complaining about it)

[Kaitlyn S]

 hrothgar, on 2016-October-22, 12:58, said:

Cheaters gonna cheat (and we're going to need to listen to other folks complaining about it)

But that's possible now. Why would the introduction to an invisible rating system that only gives you tougher competition for getting better results make cheating more prevalent?

In a way, this might be an improvement, as any real cheaters are going to rise to a level where they will play against people who notice their unusually good results, can record them, and use them as evidence to get the cheating to stop, rather than to have them continue to play against weaker pairs that will only complain about the cheating but not know what to do about it.

But not allowing the pair to be rated as a pair will give each player a higher than deserved invisible individual rating, causing other players to be disappointed when either of them joins a table as an individual.

We're talking theory here, for we all probably know that the chances of this ever happening are lower than the chances of a random "expert" on BBO being an expert (somebody that would be sought after as a professional partner.)