New Rating System Feedback

Talk Discussion in 'BeerAdvocate Talk' started by sd123, Dec 18, 2014.

Thread Status:
Not open for further replies.
  1. drtth

    drtth Initiate (0) Nov 25, 2007 Pennsylvania
    In Memoriam

    Well, if nobody has informed the mods/admin of that
    A third way to think of it is that the average is what the numbers say it is and not some mid-point of an arbitrarily defined rating scale. Thus the average can differ depending on which group is being considered. For example, when I was a freshman at the U. the average student in an intro course general ed requirement class was drawn from all over the University. By the time I became a senior, the average student in an upper division class was a major who'd survived multiple prior courses and learned a lot about their major along the way. So average meant something different in each case. To insist that the most frequent grade had to be a C in both classes would be unrealistic. Similarly, if I buy and consume only the best of the beers available to me to insist that my average rating should be roughly 3.0 is unrealistic.
     
    TMoney2591 and ExVashonGujy like this.
  2. drtth

    drtth Initiate (0) Nov 25, 2007 Pennsylvania
    In Memoriam

    Which may well be one of the motivations of this most recent change. Those who inflate or deflate their score relative to the norm are being asked to explain themselves.
     
  3. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    According to the last comment (by Todd) on this thread (which I just discovered this morning), "there's no significant differences" between review scores and quick ratings. Now knowing that, I no longer feel separating the two different types of scores as I suggested earlier is necessary. At this point, I'd just argue to switch back to a single score for quick ratings so everybody can still get what they want out of this site.
     
    Patches826 and MLaVioletteJr like this.
  4. kevanb

    kevanb Pooh-Bah (2,705) Apr 4, 2011 Illinois
    Pooh-Bah Trader

    I still don't understand how clicking 4 extra things, which are located right next to each other, which takes all of maybe 5 seconds, is drastically impeding your ability to review. Especially when it provides better data.
     
  5. Bierman9

    Bierman9 Grand Pooh-Bah (5,313) Dec 20, 2001 New Hampshire
    BA4LYFE Society Pooh-Bah Trader

    No chance of the attributes being in 0.05 increments instead of 0.25? I find a big diff between a 3.75 and a 4.00 when I want something more in between....

    Prosit!!
     
  6. Todd

    Todd Founder (13,518) Aug 23, 1996 Finland
    STAFF Mod Team Society Pooh-Bah

    No chance. Adding 0.25 increments was a compromise we made ages ago based on user feedback, but at some point we need to stop making compromises.
     
  7. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    I don't enter my beers as I drink them. I enter a bunch of them all at once based on my notes. Sometimes that's 30-40 beers, sometimes it's 90-100. Take that x5 and now maybe you can comprehend why it's an inconvenience. It would literally take 5x longer to enter beers for me and for people who use this site like me.

    Again, I'm arguing for a rating system that caters to all users. I don't see why you are arguing that everyone should switch to this new/old/different system when it seems like there is/was already tons of information on beers. People who did full reviews never stopped doing full reviews. People who used all 5 attributes never stopped using all 5 attributes. But now, a lot of people who did quick (one-number) ratings are going to stop doing ratings altogether. Get it now?
     
    Patches826 likes this.
  8. MLaVioletteJr

    MLaVioletteJr Initiate (0) Jun 12, 2013 Massachusetts

    @Todd. When rating an Imperial Stout do you consider mouthfeel to only be 10% of the overall score? How about an American Pale ale? Oud Bruin? All 10%? Doesn't make sense to me. What if you had a cloudy Helles Lager, no big deal just as long as it tastes right? Arbitrarily assigning weights to each one of the attributes and then applying them across the board to all styles can't be right. Can you shine a little light on this? Thank you.
     
  9. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    See? This is a great example of another reason against making everyone rate on all 5 attributes. I just don't want to have to think this hard about beer. I've had enough to know whether it's good or bad and about where it should fall on the 1-5 scale, but goddamn, I just don't want it to feel like homework! There is already a plethora of information out there for others to go off of, why does anyone (other than me) care what I put down for a beer?

    EDIT: I should have pointed out, though, that it's a good question.
     
  10. Todd

    Todd Founder (13,518) Aug 23, 1996 Finland
    STAFF Mod Team Society Pooh-Bah

    We've done enough to determine that the difference between a rating and a review rating isn't significant, however, it was the feedback from both users and industry friends paired with our desire to provide more thought behind the numbers that drove the update. And alienating some users was a known risk that we were willing to take.
    Correct. We had a lot of breweries that aren't open to the public scratch their heads at ratings from people who never visited; because they can't. So we're back to allowing the beers speak for production-only breweries.
    Yes, it was certainly the motivation behind outlier detection.
    We're not reverting to single scores; that would defeat the entire purpose of the update. And we're not attempting to cater to everyone.
     
  11. Todd

    Todd Founder (13,518) Aug 23, 1996 Finland
    STAFF Mod Team Society Pooh-Bah

    Yes. And it's not arbitrary. http://www.beeradvocate.com/community/threads/new-rating-system-feedback.239545/#post-3078821
    After reading all of your posts, it simply sounds like the system isn't for you. And that's okay. Or maybe you just need more time to adjust to it; after the initial reactionary phase subsides.
     
  12. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    I understand that reverting back to single scores would defeat the entire purpose of the update. My whole point is that, imo (which I realize stands for zippo in your mind), it's a bad update. When people make changes and they are bad changes, sometimes they change back or change to a third option altogether. It doesn't mean they're dumb or giving in, it means they're open-minded and willing to do what's best or right. I get that you see this all differently and that this method IS better and right, I just don't agree with it (and I suspect many, many other users don't either). Whatever, it's your site and you can obviously do what you want with it.

    The part about not catering to everyone, though. That seems a little perplexing. You guys run off ad revenue, right? Doesn't more users equal more revenue?!? Hey, again, it's your site.
     
    Fargrow, Patches826 and iTunesUpdates like this.
  13. kevanb

    kevanb Pooh-Bah (2,705) Apr 4, 2011 Illinois
    Pooh-Bah Trader

    Again, it's as simple as this. You want to this site to work for you. Except, this site is intended to work for the community. It's obvious that there is going to be a clash. There are plenty of other services, namely Untappd, that allow you to easily track and score beers, either as you drink them or at a later date. It's just a philosophical difference.

    Todd says it best, and it's been their mantra since the beginning of the site.

     
  14. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    Well of course YOU'RE okay with this change because it doesn't require YOU to make any changes to the way you use this site, but for everyone arguing against it, THEY DO have to change the way they use it (and it's a pretty big departure for people who use the single score method). I'm pretty sure you would try to stand up for your side if it were the other way around.
     
    Patches826 likes this.
  15. MLaVioletteJr

    MLaVioletteJr Initiate (0) Jun 12, 2013 Massachusetts

  16. Todd

    Todd Founder (13,518) Aug 23, 1996 Finland
    STAFF Mod Team Society Pooh-Bah

    Revenue isn't everything. And our new users and traffic actually increased this week.
     
  17. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    I'm glad you added that last sentence. It sounded kinda harsh without it. And I hope you're not taking my feedback the wrong way. I've been on here for almost 2yrs (lurking for probably another year before that) and in that time I've participated actively on the boards, I've entered a ton of new beers, I've rated over 1100 beers and 40 places, hell , I even did a trade with someone through here. I also suggest a lot of edits to beer info when I see it's missing or incorrect. And to have you say, "maybe this system isn't for you," is kind of a slap in the face (it's also my entire point, though). That last sentence softens that slap, though.
     
    ExVashonGujy likes this.
  18. kevanb

    kevanb Pooh-Bah (2,705) Apr 4, 2011 Illinois
    Pooh-Bah Trader

    Perhaps the time you joined is a large factor here, keep in mind the way it is now, was the way it was for over 10 years, and there was plenty of dissent when the single score update happened.
     
  19. SoCalBeerIdiot

    SoCalBeerIdiot Pooh-Bah (2,191) Mar 10, 2013 California
    Pooh-Bah Trader

    Exactly to my point that you'd argue if the tables were turned and you had to change they way you did things. Thank you for helping me make my point.
     
  20. kevanb

    kevanb Pooh-Bah (2,705) Apr 4, 2011 Illinois
    Pooh-Bah Trader

    I guess it was a counter to when you said, sometimes you have undo changes and updates when the community doesn't approve or it's not working as intended. This is whats happening. You're timeline is just skewed.
     
Thread Status:
Not open for further replies.