Jump to content

Best bars of 2005


khunsanuk

Recommended Posts

Hi,

 

Here are the results of the GoGo ranking over the last year (actually since mid July).

 

Only those bars with a minimum of 10 votes were counted.

 

Worst bars of 2005:
3  Fantasia            BKK-NEP 	4.40  10
2  Deja Vu             BKK-SC   4.00  13
1  Red Lips            BKK-NEP 	2.60  10

 

Best bar: Nanaplaza
1  Rainbow 4           7.70  33
2  Angelwitch          7.56  45
3  Rainbow             7.25  20

 

Best bar: Patpong
1  Camelot Castle      8.20  10
2  King's Castle I     7.18  11
3  King's Castle II    7.00  12

 

Best bar: Soi Cowboy
1  Sheba's             7.69  42
2  Suzy Wong           7.55  38
3  Shark Club          7.46  24

 

Best bar: Walking Street
1  Peppermint Palace   7.77  35
2  Diamond             7.42  26
3  Happy               7.32  28

 

Best bar: Pattaya other
1  Kitten Club         6.33  15
2  Misty's             5.17  12
2  Tahitian Queen 1    5.17  12

 

The top bars in either Bangkok or Pattaya are:

 

1  Camelot Castle      BKK-PP    8.20  10
2  Peppermint Palace   PTY-WSt   7.77  35
3  Rainbow 4           BKK-NEP   7.70  33
4  Sheba's             BKK-SC    7.69  42
5  Angelwitch          BKK-NEP   7.56  45
6  Suzy Wong           BKK-SC    7.55  38
7  Shark Club          BKK-SC    7.46  24
8  Diamond             PTY-WSt   7.42  26
9  Happy               PTY-WSt   7.32  28
10 Rainbow             BKK-NEP   7.25  20

 

Even though the first Soi Cowboy bar in the top 10 of best bars is only on the 4th place, Soi Cowboy is still considered to be the best bar area, although NEP is a very close second. Strangely enough these two areas are also the most expensive.

 

Sanuk!

Link to comment
Share on other sites

Soi Cowboy is still considered to be the best bar area, although NEP is a very close second. Strangely enough these two areas are also the most expensive.

 

It may be that those who go to NEP and SC don't visit PP too often, and those that go to PP don't visit SC or NEP. So they can't compare the two areas.

 

It may also be true that those who go to the BKK bars don't visit PTY and vice versa. Again the two places are not compared with each other.

Link to comment
Share on other sites

It just goes to show what is paradise for one is hell to another.

 

I like Fantasia. It is a fun place with no drink pressure. But one of my friends works there and that probably makes the difference.

 

Unfortunately, it looks like one of my favorites, Shark Bar, is going to be crowded. :doah:

Link to comment
Share on other sites

Hi,

 

eyebee:

"It may be that those who go to NEP and SC don't visit PP too often, and those that go to PP don't visit SC or NEP. So they can't compare the two areas."

 

Correct. NEP and SC got lots more votes than PP, which likely helped influence the scores.

 

shygye:

"It just goes to show what is paradise for one is hell to another."

 

Absolutely, everybody's tastes are different. The ranking shows the average ratings.

 

Sanuk!

Link to comment
Share on other sites

I recommend that you perform a bayesian estimate on the data, to get the true rankings, due to the errors of having a low sample size:

 

weighted rank (WR) = (v / (v+m)) * R + (m / (v+m)) * C

 

where:

R = average for the bar (mean) = (Rating)

v = number of votes for the bar = (votes)

m = minimum votes required to be listed (i.e. 10)

C = the mean vote across *all* entries (dunno what this, guessing 5)

 

therefore using top 3 in NP:

 

1 Rainbow 4 7.70 33

2 Angelwitch 7.56 45

3 Rainbow 7.25 20

 

gives you:

 

Angelwitch WR 7.095

Rainbow 4 WR 7.072

Rainbow WR 6.500

 

(Right, that should confuse everyone! :))

Link to comment
Share on other sites

AhTeckLim said:

55555555555555

 

Yehtmae's formula : weighted rank (WR) = (v / (v+m)) * R + (m / (v+m)) * C

 

My formula : fav bar = wherever I am drinking in

 

 

Reminds me: Where does an Irishman go for a Vacation (holiday)??

 

"A different bar". ::

Link to comment
Share on other sites

Yehtmae said:

I recommend that you perform a bayesian estimate on the data, to get the true rankings, due to the errors of having a low sample size:

 

weighted rank (WR) = (v / (v+m)) * R + (m / (v+m)) * C

 

Outstanding! Would have made a great word problem for my stats class!

 

"If x number of punters..."

 

Hahaha!

 

To quote Ali G, "Respect, mon!" :bow:

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...