Is BBOSkill site down?

[doxa1998]

i couldn't access it since some days (event tried from different devices...): all i get is a message saying "Error Access denied for user 'skolabri'@'localhost' (using password: YES)"

dont know who 'skolabri' could be neither have anything strange in my hosts file...

Any suggestion?

[eagles123]

it hasn't been updated for years anyway and was a complete waste of time to start with.

[johnu]

At least there was some correlation between results and rating. The only thing worse is BBO's self rating system where somebody who learned bridge in the morning can self rate as world class in the afternoon (actually, that's wrong. The only thing you need to self rate as world class is to know how to log into the site whether you know anything at all about bridge).

[1eyedjack]

Better to have a useless self-rating system that everyone knows is useless and hence ignores, than a pseudo scientific system that risks attracting greater faith in the output than is justified.

[johnu]

The site is down now, but just for drill, if it ever comes back up, take a look at some of the top ranked BBO point winners, ie those with A, K, Q, J symbols in their ID's. I checked a number of them a couple of years ago and they all had well above average or better ratings. If the ratings meant nothing, you would expect a spread from novice to world class. Of course, there are ways that the ratings get skewed (e.g. if you play almost exclusively within a closed group) but for the average player who plays against a range of opponents, it's a decent enough rating system. As far as accuracy, it seems good enough for me for its intended purpose which I thought was to screen players to match the level of a pickup game.

[johnu]

Better to have a useless self-rating system that everyone knows is useless and hence ignores, than a pseudo scientific system that risks attracting greater faith in the output than is justified.

 

That view has been discredited many times on this forum by posters who say they inflate their self ratings so they won't be automatically rejected by (most?) tables. Apparently a lot of other users put a lot of faith in the self rating system.

[MrAce]

That view has been discredited many times on this forum by posters who say they inflate their self ratings so they won't be automatically rejected by (most?) tables. Apparently a lot of other users put a lot of faith in the self rating system.

 

John, I am very fine with people inflating their skills compared to people who will cheat their @$$ of to inflate their board results so BBO skill or whatever shows them as expert or world class.

 

Compare BBO with OKB (and this is coming from someone who resisted leaving OKB for a long time and did not play in BBO untill 2005) BBO is relaxed. If we let this BS BBOskill get in to our lives, BBO will turn into chaos. Now when we play a board and one of the opponents or both of them leaves in the middle of it, we do not care, we skip. But we will be angry if that board was bringing us too many BBOskill points. We will call yellows and...omg...you name it, we will have it, all kind of things that none of us want in our life. And I do not think yellows, most of them are working voluntarily, will find any time to deal with tons of incidents everyday.

 

Even when your opponents play honest but having a hot day, it will make you paranoid. Once this paranoia starts, it will make chain reaction and will create more cheaters who has this stupid idea of getting revenge. We do not need any of this *****! Believe me you will regret it if that happens.

[johnu]

MrAce - You are making the assumption the only reason some players will cheat is to improve their ratings. This is actually a moot point since BBoSkill hasn't been updating ratings for a couple of years now, but some people will cheat even if there is no ratings to improve. As long as a score is kept, some people will cheat or behave badly. Others will leave the table to avoid a bad result, berate or boot partners who make a mistake, reject anybody without an expert rating, etc. This is all without a result based rating system currently in place. You don't have to read far into the forums and to see people complaining about cheating and bad behavior.

 

And even without ratings, BBO keeps score. There are pair and individual tournaments, team games, even the normal main bridge club where there is a running score kept. Cheating is ridiculously easy online and ratings are likely the least important reason for doing so.

[1eyedjack]

That view has been discredited many times on this forum by posters who say they inflate their self ratings so they won't be automatically rejected by (most?) tables. Apparently a lot of other users put a lot of faith in the self rating system.

I hardly think so. We get a handful of posts of that nature, perhaps a dozen or so (if that) over the course of a year. Meanwhile the site usually has over 10K users logged in simultaneously at some time in the day.

 

There is an element of validity to your point, in that someone who self-assesses beginner status is statistically more honest than someone who self assesses expert. Those who assess themselves as beginner and find themselves inconvenienced having done so soon learn to amend their profiles. Possibly in conjunction on rare occasions with a moan to the forums.

 

Meanwhile, the toxic consequences of Lehman ratings (which I understand were more accurate than BBOSkill) to the OKBridge site demonstrate that the proposed cure is worse than any perceived disease.

 

I have an alter ego that I sometimes use on BBO. I do so only because the software denies us the opportunity to conceal our BBO master point status from public scrutiny (for no justification that has yet been explained), and I do not use that other account to play in qualifying events. I self assess myself as "private" in my alter ego, which setting has, if reports to the forum are to be believed, as much a disincentive to finding a game as "beginner". Strangely, I do not find it as problematic as others profess, and when faced with a choice between reports of others and my personal experience I side with personal experience.

[steve2005]

bboskill will always be dead it used up way too much bbo resources when it was calculating ratings

[johnu]

I hardly think so. We get a handful of posts of that nature, perhaps a dozen or so (if that) over the course of a year. Meanwhile the site usually has over 10K users logged in simultaneously at some time in the day.

 

The number of posters in the forum in a minuscule percentage of the players on BBO. Lots of reasons for not posting on the (English language) forum, English is not their native language, just lurking and not posting, no interest in posting on the forum, etc. Extrapolating percentages of the total population from forum members is hopeless without extensive sampling studies.

 

Meanwhile, the toxic consequences of Lehman ratings (which I understand were more accurate than BBOSkill) to the OKBridge site demonstrate that the proposed cure is worse than any perceived disease.

 

Have you actually played on OKBridge? I was a member in the past but it wasn't worth the annual cost. I didn't think OKBridge was any worse, or any better than BBO in terms of being a friendly place to play. Plenty of people still play on OKBridge at ~$100+ a year so they are voting with their money that OKBridge is worthwhile. BBO is free (except for ACBL tournaments, etc) so most people will put up with a lot more since it doesn't cost them any money.

[lycier]

BBO is great.

BBO is excellent.

BBO is wonderful.http://www.bridgebase.com/forums/public/style_emoticons/default/rolleyes.gif

[1eyedjack]

Have you actually played on OKBridge? I was a member in the past but it wasn't worth the annual cost.

Yes I was a member for many years, both before and after they started charging. Cost was a factor for me also in deciding to resign, but not the only one.

 

Despite that OKB retains a loyal core of followers, it has not prevented moans about Lehmans that are just as prevalent as the dribble of moans about self rating. The former moans (on OKB Lehmans) centre on the fact that players rely on them. It got to the point when players could apply to have them re-set periodically, somewhat diluting their effect and reliability The latter moans (on BBO self certification) centre on the fact that you cannot rely on them, which is a valid moan but countered by the simple solution: don't attempt to rely on them. At least not seriously, or perhaps only after playing a few hands (it normally only takes 2 or 3).

 

Personally I think that the self certification is pretty pointless and a waste of screen space. But also pretty harmless for all that once you have a few weeks' experience on BBO, so live and let live I guess.

 

To introduce the problems of beginners finding a game is to conflate two separate problems. Introducing an accurate rating system that confirms their beginner status is NOT going to solve their problems finding a game amongst a population of advanced players who do not want beginners at their table. The solution to that problem must lie elsewhere.

[MrAce]

Honestly, if BBO charged me more than OKB, I would still not hesitate to choose BBO over OKB. And this is coming from someone who resisted leaving OKB for a long time.

I am a bridge fanatic. I play about 450-500 boards per week. I would not mind paying for it if I have to. I would never choose something worse just because it is free or cheaper when it comes to bridge. In fact I probably pay to BBO more than I payed to OKB.

[Jinksy]

There is a fundamental question which BBOskill was trying to address and self-ratings are basically irrelevant to, which is 'how do I get a decent game with randoms?'

 

You can make some guesses based on their profiles, but it's not always that effective - and in any case, you can't (for eg) automatically filter the people you see when looking at tables/tournament registrations etc, to save yourself the hassle of looking at 15 different empty (/Gerber/Benji Acol) profiles before you get to a single one who looks plausible.

 

I don't know if there's any way of solving this without causing the issues Timo is talking about, and I'm not claiming it's a bigger problem than they would be, but I wish people wouldn't treat it as a non-issue. BBO would obviously be a better experience if it could just magically figure out our skill level and give us some options based on it.

 

For what it's worth, abandoned boards don't feel like they would be a showstopper - if dummy has been tabled at the point any player but dummy leaves, BBO could just ask robots to finish the hand (I think it already does this in some cases?), and assign the final score to the players who were there at the start. It wouldn't be perfect, but it should average out over time.

[helene_t]

BBO would obviously be a better experience if it could just magically figure out our skill level and give us some options based on it.

BBO could easily do this. No need for magic.

 

However, BBO has chosen not to do so.

 

We have had many discussions about this, see http://www.bridgebase.com/forums/topic/35525-rating-players/ for one of the better threads.

 

For BBO's official point of view, see http://www.bridgebase.com/help/3/topics/faqs/text/intro.html#ratings

[Jinksy]

Obviously rating players is eminently possible (the NGS does it reasonably well). I meant 'magic' as in, 'with no reference to actual table results, so that it doesn't have a social impact of the form Timo was describing'.

 

I'm claiming that having ratings is surely a net positive for the online bridge experience. Getting them is evidently a net minus. Weighing these things and deciding that the costs outweigh the benefits is reasonable, but ignoring the benefits altogether isn't, and seems liable to discourage seeking incremental improvements towards the ideal of having useable ratings without social disruption.

[StevenG]

Perhaps the importance of skill level is overrated for BBO. I think there are two main distinctions between players, serious/social and chatty/silent. If a profile said something like serious/chatty and named a basic set of systems played (SAYC, Acol, etc.) without delving deeply into conventions, it would be much easier to find a suitable game. Extending this further, you could set up tables as social/chatty/SAYC and only allow people whose profiles matched.

[mycroft]

As we've discussed before, provided the skill level was *hidden* and *only used for "pair me with randoms"*, I'm happy with it. It'll probably be better in the long run. And of course, they may be doing it (remember what I said about "hidden"?) Anything else will be gamed, and will lead to "I can't play with you, it'll hurt my rating" or "I can't play a drunk-and-crazy speedball for fun". Which was my issue with Lehmans.

[hrothgar]

I said it before, I'll say it again...

 

I think that BBO would benefit from a permanent floating Indy

 

You enter the "event" and you get assigned a random partner and matched against another pair of randoms.

you play some short number of boards together (say three)

 

At the end of the round scores are totaled, the winners move up the ladder and the the losers move down.

For fun, allow players to partner with a robot and have a few robot pairs.

 

If folks rage quit in the middle of a round, ban them from the event for some period of time and treat their remaining hands as bottoms.

 

I see a couple benefits to this sort of system

 

1. This provides a format where people get to play with and against a wide number of people. it will give folks a chance to discover new players who they might be compatible with.

2. For those who really care, they can treat their rank in the ladder as a "rating". (If you're able to rise to the top, you are presumably playing quite well)

 

if this event proves popular, you can have 20-30 different virtual sections running at any given point in time.

(A section would be some ephemeral grouping of 20-30 tables whose scores would be compared for a single board)

 

It wouldn't be impossible to cheat with such a system, however, it would make it more difficult

[Jinksy]

Anything else will ... lead to "I can't play with you, it'll hurt my rating" or "I can't play a drunk-and-crazy speedball for fun". Which was my issue with Lehmans.

 

Maybe anything else is gameable, but there'd be nothing to stop BBO from implementing an opt-out system, or one partition game types into ranked and unranked.

[steve2005]

Maybe anything else is gameable, but there'd be nothing to stop BBO from implementing an opt-out system, or one partition game types into ranked and unranked.

I think BBO policy on ratings is pretty clear that it is a no go and has nothing to do with ability to implement a rating system.

[barmar]

Maybe anything else is gameable, but there'd be nothing to stop BBO from implementing an opt-out system, or one partition game types into ranked and unranked.

IIRC, OKbridge has (or had, when I was a member until around 10 years ago) partitioned games. They have competitive games that affect your Lehmans, and casual games that don't.

 

I don't think it helped much.

[mycroft]

s/I don't think it helped much/It was a disaster/

 

I'm not a bridge operator, I can afford to not be politic in my opinions (sometimes).

 

I used to run a mentor partnership on OKB. If I set the table, and it was ranked, we'd get players who were playing at my level and were less than thrilled at both the fact that partner was so much less experienced and that we wanted to discuss the hands - even if I marked it clearly a mentoring table. If she hosted, the same, except *I* was geting the side-eye. If it was unranked, we never got opponents.

 

Eventually we found another compatible mentor pair to play with, and then when that mentor left soon after, I ended up picking up all three as students. That's the only way it worked.

 

The best part about ranked games was that very quickly, the invites were in very narrow rating bands; as a result, nobody ever either moved much or learned anything. And what does a 55-45 pair do?

 

It's really simple. Everyone wants a partner slightly better than they think they are, and opponents that are just slightly worse. That'll just work, won't it?

[ladydoc]

As we've discussed before, provided the skill level was *hidden* and *only used for "pair me with randoms"*, I'm happy with it. It'll probably be better in the long run. And of course, they may be doing it (remember what I said about "hidden"?) Anything else will be gamed, and will lead to "I can't play with you, it'll hurt my rating" or "I can't play a drunk-and-crazy speedball for fun". Which was my issue with Lehmans.

 

 

It is not possible on BBO at the present time to have a perfect legitimate game, except in the ACBL tourneys, which are monitered (I hope).

 

For years the people with the highest skill ratings on BBOskill got them by playing with themselves as partners. If you wish the name (names) I can tell you who they were, and that's just the tip of the iceberg.

The reasons people would do this 1. economical, ... those are the 'teachers' who give paid bridge instructions and want to have a believable profile on BBOskill,

and 2. to appeal to a selected field target ie, the other sex.

 

Bridge base online is great the way it is now, with no ratings, except self ratings. Most true experts underrate themselves because they know each other.