[uday]

The New York Times , at the end of this article Jordan Chodorow and Karen Allison Play at Bridge Base Online said

Quote

One drawback of online play has been the adverse effect on bridge club attendance.

I presume they mean "ACBL bridge club attendance"

The ACBL was kind enough to supply data on annual tables from offline/online clubs

To me, this doesn't really demonstrate that online has been bad for offline ( or if it has, that the impact has been trivial ).

Am I right to interpret this data in that way?




Raw numbers are:

Quote

1999-12-31 2208217.0 590.0
2000-12-31 2175938.7 2587.5
2001-12-31 2376748.0 3320.0
2002-12-31 2513059.5 2768.0
2003-12-31 2423864.0 3480.0
2004-12-31 2455965.4 22396.0
2005-12-31 2439972.6 84251.0
2006-12-31 2424666.3 164079.0
2007-12-31 2336286.5 335974.5
2008-12-31 2364245.2 470854.0
2009-12-31 2448937.9 589520.0
2010-12-31 2409653.8 709333.0
2011-12-31 2401487.7 818929.0
2012-12-31 2392044.1 905890.0
2013-12-31 2370579.0 942040.0

[Bbradley62]

Your interpretation of the data certainly seems to be accurate. I doubt it's relevant that we don't know where the 778,721 in the article actually comes from.

[hrothgar]

Could quick comments

1. Stacked bar graphs aren't a good way to visualize this sort of data.

A pair of line graphs suggests that BBO table count is following a sigmoid and that ACBL table count is bouncing arround
A scatter plot doesn't suggest a strong correlation

2. Even if I saw a strong correlation, I'd want to see whether there might be other variables confounding the data

For example, what is happening to overall ACBL membership in North America

Personally, I think that any decrease would be related to aging population.

[whereagles]

A quick back-of-the-envelope correlation check over the data shows: (calcs: R software)

Pearson correlation online-offline: +0.129
Spearman correlation online-offline: +0.05

Figures mean correlation is nearly non-existant. If something, the plus sign even hints at more synergy between the two. (A minus sign would mean online might be stealing offline, but it would still require correlation to be over ~50% to be meaningful.)


Linear regression with year as independent variable shows:

Online data fits very well with linear growing trend.
Offline data does not fit well with a linear model.

Offline data has no clear increasing/decreasing trend under linear regression. A more elaborated model is necessary to predict future evolution.


Basically, preliminarily I would say online bridge is not handicapping offline. This is contrary to the NYT article's comment.

[winkle]

Yet another god awful column? Poor deal analysis accompanying by baseless comment; seems par for the course for that column.

[treytor2]

Uday is there a way to narrow the field by say USA only and or North American countries? I wonder if countries like Turkey, Italy, Spain or such where historically it was a huge playing field in the early 1900's to present and or where Tv was not a great consideration, so local clubs were filled and or bridge in family homes was played with regularity? If that is still so and they are also in ACBL I think some of the figures are skewed? I know locally we have gone from 10 tables to 3 and half to 5 is all in the last 10 years. One reason is as stated age and death. I see very little new blood coming in! But I know of some that do play on line. Another consideration as in my case a baby boomer still working. I cannot make local clubs as some do in retirement. I feel there is still a large group of us still working up to 67 or more of age. I get more playing time in BBO than local. My estimation is on line is the winner and local is declining. Perhaps ACBL needs to go online and ask poll questions in private to people asking them if they play and or why not? Also find out what is the problem as for me it is geographic? I am 2 hours from Las Vegas or Henderson the only place closest to get points or play good players. Can you give figures from Sectionals, such as Lake Havasu City I bet has gone from 100 or more tables to maybe 20 or 25? Is it the timing with other sectionals or ask the people why they don't support it? Certainly more research than this is needed for a better comparison and evaluation. Thanks.

[uday]

Quote

we don't know where the 778,721 in the article actually comes from

Those are the 2013 ACBL tables at BBO. The graph contains all online ACBL tables, from all online ACBL clubs.

[Vampyr]

uday, on 2014-April-21, 21:47, said:

Those are the 2013 ACBL tables at BBO. The graph contains all online ACBL tables, from all online ACBL clubs.

Are the ACBL BBO tables all for-pay tables? I would guess that more people play in free than paid games, but there is probably no way to tell how many of these are ACBL members (or would-be ACBL members; say if only live play were available).

[uday]

Vampyr, on 2014-April-21, 22:09, said:

Are the ACBL BBO tables all for-pay tables?

Yes. We ran a survey once, and this suggested that there are many thousands of ACBL players on the site who don't play in ACBL games, but are here nonetheless

[Siegmund]

Quote

Yes. We ran a survey once, and this suggested that there are many thousands of ACBL players on the site who don't play in ACBL games, but are here nonetheless

I can confirm that -- I play a lot of set games, and a fair few robot tournaments, but for a variety of reasons (sadly, mostly to do with the software and the directing staff) almost never play a human-vs-human tournament at BBO.

As to the original NYT article... anecdotally, I know many people who play both places, but know many people who learned online and later came to the club, and only a small handful who quit coming to the club in favor of playing only online. (Same thing with poker - states like WA and NV banned internet poker to "protect" their live casinos, but I and many other people learned poker online and later played in casinos... without online poker, I would have never given those casinos a dollar.)

[helene_t]

Richard and Nuno: simple correlations are meangless as you cant disregard tlime. You might want to do a time series analysis but imo the stacked bars are ok. Agree with Richard that line charts would be preferable.

Not much to conclude from these data. Breakdowns per regio and age group might be helpful. Or maybe some interviews with players who have increased or decreased their offline activity.

[MrAce]

helene_t, on 2014-April-28, 02:20, said:

Or maybe some interviews with players who have increased or decreased their offline activity.

I met online bridge at Ville De Lille - France, World Champ.

I used to go to bridge club everyday, I mean literally everyday since I was 18. And then I opened my own bridge club and had to go both clubs everyday. One of them was my business, and mostly my students, other one was where I give lessons free to college students - high school students and play with good players.

Once I met the online bridge, I immediately cut all my offline bridge activities except than professional causes and big events. Mostly I was fascinated to be able to play with different people from different cultures and nations. And bridge I play at the bridge club was never enough for me, even though I played at least 6-7 hours everyday and after the game we all just drink at the pub next door until we all get boozed up by alcohol and talking the hands. (This is the part I missed about Turkey and other European countries that I lived for a while) But the number of good players in my city was limited and we all knew each other from the way they breath. Playing almost same system etc etc...Online bridge gave me opportunity to play bridge about 15-18 hours a day, this was like heaven for me. Now I was able to play as pd or opponent with other good players from Turkey and then from other countries.


I have to also mention this; I was able to see at most 3-4 hands played by top celebrities in bridge magazines. Like Garozzo, Meckstroth, Rodwell, Hamman, Helgi.....all hands were hand picked by editor of course. First I was disappointed when see some of them go down in a contract when I thought they could easily make. I learnt that the people whom I thought were GODLIKE, were also making mistakes sometimes or perhaps they were just playing better line which happens to lose on this particular hand. Now I was able to sit and watch them. They were all over the place in OKBridge. This was like my dream that came true. 24/7 available games, kibitzing, learning....I fell in love with it and %80 decreased my offline bridge activity.

But that is just me