Tickers versus Reviewers

A) Not sure why this is in Mid-Atlantic

B) It sounds like you need to find a reviewer that you trust and follow them. There are plenty of outstanding BA reviewers, just lurk in the "What are you drinking" thread and you can usually find a few of 'em. Then, head over to their reviews and find ones that speak to you.

C) I like having the "public" vote to compare my own taste to, but I never look at the votes before tasting and voting myself. It's exciting when I go high on a beer that the general public doesn't love or vice versa, but more often than not I'm fairly close to the average.

 

Some people tick, some review. Some people do both. Why is this even a thread?

 

* this has been moved to site help section, thought you might find better answers here *

 

Its definitely to your credit that you'd worry about this, so good on you.

One thing its easy to miss on here is the explanations given behind certain decisions. As I understand some things I've seen in the Announcements forum and some comments made along the way in various threads it’s a deliberate choice to merge the two sets of numbers.

When ratings only were first introduced the two sets of numbers were kept separate. After a reasonable period of time the folks running the site did some statistical tests and found that there were no consistent significant differences between the numbers produced by reviewers and the numbers produced by raters, so the two sets of numbers were merged into one, if only because keeping them separate didn't change anything. (Basically it seems that the raters as a group do try to provide realistic numbers representing their reactions to the beer.)

Also there's also something that happens as the number of ratings/reviews increases. Basically the extreme cases come to have less and less impact on the average score as the number of ratings/reviews done by those who care to do a reasonable job of picking their numbers increases. This means there's a general tendency for those who are just throwing out any old number to have a score to their credit to be wasting their time, if only because if they hit the mean nothing changes, if they go the high extreme where its not realistic, others will go low extreme and the two sets of extreme numbers effectively cancel each other out of the mean score.

That said, I recommend that you take some time to try writing reviews, even if only for yourself. Over my time here I've found that with reviews (once I felt comfortable enough with what I was writing to post them) do a much better job of helping me remember that I've had a beer and what I liked or didn't like about it. And if I actually go back and re-read the older ones I can see both how my tastes have changed and how my knowledge/enjoyment of beer has improved.

Cheers!

 

Pretty much for the same reason you took the trouble to post and ask a question. Some people like to ask questions about things they don't understand.

Enjoy your day.

 

This is just an observation rather than a mathematical analysis but I find that tickers are more inclined to give very high or low scores than are reviewers. While this may not affect the ratings of beers that have a lot of ticks/reviews, I think it does impact those with a relatively low number. I rarely see reviews above 4.5 or below 3.0 but I have seen lots of ticks outside that range. For example, I recently came across a beer with about 10 ticks, most of which were in the 3.75-4 range but just a couple of very low ticks dragged the overall score down to about a 3.25.

 

I'd say that your observation is pretty much spot on and say that you've described pretty well both what can happen with small numbers and what typically will happen as the number of ratings increases. I'd bet that if we could go one step further and look into the future to see the overall score for that beer when it has 50-100 ratings, there's a good chance that the score would be within shouting distance of 3.25. This is one big reason why personally when looking at a beer with a low number of ratings I start with the overall score (rAvg) and how it compares to others in the same style, look at the variability (pDev) to see how large that is, and then, for beers I'm interested in, start looking through the reviews to see what the reviewers I trust have to say about the beer.

 

I used to review (not on BA). But then I realized I wasn't actually enjoying the beer, simply critiquing it. You are no better or worse as a BA whether you review or tick (though I do neither).