New Scaratings https://scaratings.com/newScaratings/ |
|
Measures of bar prowess https://scaratings.com/newScaratings/viewtopic.php?f=43&t=1119 |
Page 1 of 1 |
Author: | Rhino [ Wed Jun 22, 2011 5:15 pm ] |
Post subject: | Measures of bar prowess |
Being a numbers geek, I follow a lot of data that does not and will never relate to myself or where I play. Top bar rankings are one such, and all of us are familiar with the leaders there. But are we? Consider how business works. If your revenue is $5B and your competitor's is $3B you are the biggest in the market space. But if your COGS and SG&A ARE $4.5B and theirs are $2B, not only are they more profitable but perhaps more importantly, much more efficient. For the sports fans, the Yankees win a crapload more games than, say, the Pirates. But their payroll is many times greater - each W costs a lot more in Yankeeland (yes - I know they are also very profitable - just illustrating one point here). So I started looking at bar efficiency - how much input (points or games) creates how much output (medals and rankings)? Now obviously this is a bit skewed as the number of points needed to win a LT game and those needed to win SD are very different. But over a six month period and considering that BT/CD/LT make up the huge bulk of games I suspect the impact is not massive. Also we all know not all medals are equal for prowess, but they are for rankings. Turns out Bob's Place rules the roost on points/medal efficiency. They need only 100180 total pts per gold and 62016 per medal (with LT about 1/3 of them) Perhaps even more eyepopping is the game "quality" stat - the ratio of top 100 games which are golds. Here the shenanigans come into play with BWW Mesa needing only 1.7 top 100 games per gold (timezone assisted). We have to go to T Bonez in Louisville to evade the Mesa machinations at 3.88 t100/gold. My own place is middle of the pack at 828k/gold and maybe top quartile at 561k/medal, needing just under 14 t100/gold Perennial superpower 86th Street is 187.6k/gold, 150k/medal and golds once every 9.75 top 100 games. Their only close rival Demoris trails a bit in pts efficiency at 199.6k and 151.7k, but edges them out needing only 9.2 top scores to get a gold. |
Author: | Dante [ Wed Jun 22, 2011 8:46 pm ] |
Post subject: | Re: Measures of bar prowess |
Rhino wrote: We have to go to T Bonez in Louisville to evade the Mesa machinations at 3.88 t100/gold. My own place is middle of the pack at 828k/gold and maybe top quartile at 561k/medal, needing just under 14 t100/gold . I played at T Bonez a few times. For a few months, almost every game was a repeat of earlier games in the day. I don't know if that affects anything at all. |
Author: | Akbar71 [ Wed Jun 22, 2011 10:18 pm ] |
Post subject: | Re: Measures of bar prowess |
Rhino wrote: Being a numbers geek, I follow a lot of data that does not and will never relate to myself or where I play. Top bar rankings are one such, and all of us are familiar with the leaders there. But are we? Interesting stats. Did you actually feed a whole lot of info into a spreadsheet or just run the numbers for specific bars of interest? If you do have a spreadsheet, I'm sure seeing it would interest many here, if it's postable.
Consider how business works. If your revenue is $5B and your competitor's is $3B you are the biggest in the market space. But if your COGS and SG&A ARE $4.5B and theirs are $2B, not only are they more profitable but perhaps more importantly, much more efficient. For the sports fans, the Yankees win a crapload more games than, say, the Pirates. But their payroll is many times greater - each W costs a lot more in Yankeeland (yes - I know they are also very profitable - just illustrating one point here). So I started looking at bar efficiency - how much input (points or games) creates how much output (medals and rankings)? Now obviously this is a bit skewed as the number of points needed to win a LT game and those needed to win SD are very different. But over a six month period and considering that BT/CD/LT make up the huge bulk of games I suspect the impact is not massive. Also we all know not all medals are equal for prowess, but they are for rankings. Turns out Bob's Place rules the roost on points/medal efficiency. They need only 100180 total pts per gold and 62016 per medal (with LT about 1/3 of them) Perhaps even more eyepopping is the game "quality" stat - the ratio of top 100 games which are golds. Here the shenanigans come into play with BWW Mesa needing only 1.7 top 100 games per gold (timezone assisted). We have to go to T Bonez in Louisville to evade the Mesa machinations at 3.88 t100/gold. My own place is middle of the pack at 828k/gold and maybe top quartile at 561k/medal, needing just under 14 t100/gold Perennial superpower 86th Street is 187.6k/gold, 150k/medal and golds once every 9.75 top 100 games. Their only close rival Demoris trails a bit in pts efficiency at 199.6k and 151.7k, but edges them out needing only 9.2 top scores to get a gold. |
Author: | Rhino [ Thu Jun 23, 2011 8:01 am ] |
Post subject: | Re: Measures of bar prowess |
All I did was cut and paste top 200 bars from BTs list into Excel and then add some simple arithmetic. Don't think I can post Excel here but will happily send to anyone who asks, |
Author: | Jim [ Thu Jun 23, 2011 8:55 am ] |
Post subject: | Re: Measures of bar prowess |
Interesting stuff, but stats can be so misleading. It still comes down to whether you like the bar and other players. A lot of players don't regularly play at the location that gets credit for the medals and wins. You can't differentiate between team golds and individual golds. And over the years I've found it amusing that some very very good players have few if any medals while some very average players have lots of medals ( maybe it's the games and time of day they play?) and now there are a few strutting about with lot of iphone medals. Consider Indigo Joe's in Wildwood Missouri with 43 golds and only a couple of them earned not using the iphone scam. Two very important stats to me that Buzztime no longers displays are the locations activity rank and the last time the site reported. |
Author: | gavv [ Fri Jun 24, 2011 8:49 am ] |
Post subject: | Re: Measures of bar prowess |
I still wish they'd keep track of 'site wins' as a stat in addition to the current medal count, it would be another (and i think) more interesting metric. It would be more interesting to see a site get a 'gold site medal' for say a 9900 BT bar score win that just happened to have 1 question where all five were .2 sec too late and 'unsynced' and yield only 1 regular gold (if that), versus just that split second yielding 5-10 golds. gavv Also, teams with heavy percentages in showdown, sports trivia challenge, six, spotlight, playback would skew the numbers pretty well given the total points possible and the near impossibility to obtain more than one gold in a game. |
Author: | Rhino [ Fri Jun 24, 2011 11:23 am ] |
Post subject: | Re: Measures of bar prowess |
gavv wrote: ......Also, teams with heavy percentages in showdown, sports trivia challenge, six, spotlight, playback would skew the numbers pretty well given the total points possible and the near impossibility to obtain more than one gold in a game. I know what you mean, but a bar that heavily slanted in medal counts is unlikely. Think of it this way - if a bar won every single gold for six months in SD, it would barely break the top 200 I'm looking at, and that's pretty much impossible anyway. A bar at the top of the tree like yours could win every single premium game for 6 months and still have that be less than 18% of your golds. |
Author: | gavv [ Fri Jun 24, 2011 3:21 pm ] |
Post subject: | Re: Measures of bar prowess |
Rhino wrote: gavv wrote: ......Also, teams with heavy percentages in showdown, sports trivia challenge, six, spotlight, playback would skew the numbers pretty well given the total points possible and the near impossibility to obtain more than one gold in a game. I know what you mean, but a bar that heavily slanted in medal counts is unlikely. Think of it this way - if a bar won every single gold for six months in SD, it would barely break the top 200 I'm looking at, and that's pretty much impossible anyway. A bar at the top of the tree like yours could win every single premium game for 6 months and still have that be less than 18% of your golds. well, that was purely in the 'points per gold' metric. Where if a site has 8-10M in points in a couple of premium games where its maybe a handful of golds, and only one per game at that, it skews that number. Any site that's mostly premium game-playing doesn't really fit well into any of bt's existing metrics. Kinda wish they'd also break it down into top sites per game instead of just for players. it's all ebb and flow too. Premium-wise 86th st primarily plays just STC on a regular basis, and the occasional tournament-time playing of Six. Steve and Turbo are daily grinders, and we have some pretty solid weekend afternoon sportswatching play. During non NFL weeks and STC is at 9Eastern, we play solid from 5-9 and make some hey, but during the football season, we play STC at 730 and leave, those extra 6 BT games during the offseason does net us a lot more extra medals. gavv |
Author: | MitchWolf [ Fri Jun 24, 2011 10:45 pm ] |
Post subject: | Re: Measures of bar prowess |
Jim wrote: Interesting stuff, but stats can be so misleading. It still comes down to whether you like the bar and other players. A lot of players don't regularly play at the location that gets credit for the medals and wins. You can't differentiate between team golds and individual golds. And over the years I've found it amusing that some very very good players have few if any medals while some very average players have lots of medals ( maybe it's the games and time of day they play?) and now there are a few strutting about with lot of iphone medals. Consider Indigo Joe's in Wildwood Missouri with 43 golds and only a couple of them earned not using the iphone scam. Two very important stats to me that Buzztime no longers displays are the locations activity rank and the last time the site reported. Thanx Jim for mentioning Indigo Joe's there. ROMEOT n ROMEOD is the reason the medal count is ridiculous high. He never plays a box at the site he is at and logs in to another site. He does that to my home bar but yet don't know where he is REALLY at. I used to play a lot at places where I am a visitor. But on workday, I go to my home bar after work to play and at times I go somewhere else on my days off just to play LT due to Baxter's don't open til 4p CST. |
Page 1 of 1 | All times are UTC - 5 hours |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |