@BDanner Yeah, the hint of fruit in here is good, but sadly not enough to compensate for the problems. Would be interested to try after several months open, but no way am I buying a bottle just to check.

I went back and checked my notes. Peaches and cantaloupe on the finish.

I was also disappointed with thus one. I did find it slightly better after a few months open. Picked up a melon note on the finish that I actually quite liked.

@1901 Haha! I'd actually have a lot of fun reading a variation looking at slang. I mean, I wouldn't understand understand most of it, but how does fleek compare to amazeballs? The world must know.

@ContemplativeFox i thought that was strange for the US list, but I put it down to some sort of national nuance. So, interesting that you align more with the UK ranking. I do too. Thankfully no dench, wicked and totes amazeballs in the list.

@1901 That's really cool! I think I identify more with the UK here because reading down the US one I was shocked that the comparatively minor "very bad" was at the bottom while "abysmal" had 5 below it. It was also interesting to see that "quite good/bad" is perceived to be more extreme than just "good/bad" in the US, whereas in the UK the opposite is true.

@cascode Thank you for sharing your rating system.  It's fascinating how we encountered the same problem and then went in such different directions to solve it, yet produced similar results! I like the premise of your system as well and even see some surprising parallels in it, such as the max 100/100 indicating perfection and being essentially unattainable, yet the max 5/5 being not wholly uncommon.  I actually toyed with something a bit more like your system for a while where I came up with set of positive and negative words that related to several aspects of a drink (e.g. "complex", "harsh", "intriguing", "vibrant").  For each tasting, I'd select any number of them.  From there, I was going to develop a regression that applied weights to each word based on whether it was chosen or not and gave me a score for the dram. I found the whole thing to be too much effort though (particularly in trying to backfill for past drinks), so I abandoned it.

@cascode funny you should mention grand -that would be a good one for the list . Fairly common usage here and can mean anything from average to excellent

@1901 Thanks for that, I've not seen it before. I think Australian speakers would be more in line with UK usage than US. The UK list certainly lines up a little more closely with my own. Interestingly, three of the words in my list that I use quite often (inferior, adequate and acceptable) don't appear on those listings at all. On the other hand mediocre, fair and satisfactory are words I don't use. Ain't language grand 😄

@cascode given your categories you might find this interesting https://www.visualcapitalist.com/word-sentiment-scale/ Don’t know where Australia fits in :)

@ContemplativeFox It's interesting that we both encountered the same issue in rating systems during our respective journeys and needed to come up with different approaches. It's even more interesting that our ratings here often coincide although we ended up finding very different solutions. Yours is objective, while mine is subjective and would probably seem the wrong way round to some people. By the mid 1990s I had about 500 tastes in my old hand-written tasting journals and the arbitrary system based on comparison that I had used up until then was breaking down, for the same reasons you noted. I tried to make an objective evaluation system, but it never really worked. Often the score just didn't seem right, so obviously my algorithm was flawed. In the end, I decided to review purely on gut instinct by using a list of 44 "summary terms" from which I could select (eg "Poor", "Adequate", "better than just Acceptable, but not quite Average", "Good, almost Very Good" and so on). The terms each equate to a rating on a scale of 56-100. I went back and re-rated everything in my journals that I could still accurately remember, and I've used it for everything since. I allow myself a little latitude and there are occasional off-the-chart descriptors that I make up on the spot to use here on Distiller like "As Close To Perfection As Humanly Possible" and "Spurn It As You Would A Rabid Dog" but these are always just alternatives to "Outstanding" and "Undrinkable", which are the highest and lowest scores respectively. I have no hesitation in awarding something 5/5 here if I think it is justified because 5/5 does not mean "Perfect" in my rating system, it means "Outstanding" which is the equivalent of a score of 90+/100. "Perfect" is specifically a score of 100/100, which I have never given to anything.

@Soba45 It's also a big wall of text that I just typed in one go rather than a well-edited explanation with proper terminology, graphs and citations.

@Jan-Case Ah, yes, you're right. In most cases, it turns out that there's a clean translation to the Distiller scale. Only 2 cases where my scores round to the same number of stars :)

@ContemplativeFox Wooah...i think that explanation literally just blew my mind..haha. Very interesting read and thinking! Maybe I shouldn't have read it at 6am though..a bit to early for deep thinking :-)

@ContemplativeFox no, nothing special with Distillers rating system. It is just that the 0 to 5 stars with their 0.25 increments are in total 21 ratings to chose from.

@Jan-Case Yeah, I realize that it's unfortunately by design a deliberately complicated rating system, which is why I translate the ratings into stars but also start my recent tastings with a line like "Rating: 8/23", so at least anyone reading has some hope of understanding when I ramble on about X spirit being maybe a 9 or even 10 that I'm saying it might be a little better than the rating I gave, but still not that great. I'm actually not familiar with Distiller's 21 step system, but I'm familiar with other step-based systems and I'd be interested to know the details of what Distiller does! I don't see it explained anywhere obvious though? Do you happen to know where I can find out more about it?

My short answer for Bushmills, Black Bush, then Red Bush, then Slane, Tulamore DEW, Redbreast

@ContemplativeFox the 21 steps rating Distiller uses I guess works nicely when applying / adapting your scoring to it.

@ContemplativeFox I as a German can definitely see why a systemic approach like this actually really sound like the proper way of doing this (as far as I could understand it) - but as someone who repeatedly failed maths in my final years of regular school the theory behind it is completely beyond me. I didn’t need maths for university but in situations like this I sometime have some regrets of being lazy when it came to numbers :)

@Jan-Case Great question. I'll answer it with way more detail than I imagine you want. TLDR: It's a prime number so it's hard to mentally convert to a standard scale (e.g. 100 points), which allows me to rank items into buckets moreso than actually rate them, which through reading I've found humans are bad at and through experience I've found I'm bad at. I keep reading that humans are bad at assigning ratings and I found that to be very true in another ratings like I made, where after the first couple hundred I realized I was essentially ranking things against each other to determine ratings. That ranking approach served me quite well though until I got to around 500, at which point it started breaking down as well because I'd get hotspots where I mis-rated something and would then over-or-under-rate numerous others because I'd say to myself "well, this is definitely better/worse than that other one". This proved to be particularly problematic at key numbers like 70 because even though I was using the full 100 point range (well, aside from the absolute extremes) and following a roughly normal distribution, I found myself fixating on certain "special numbers" that are meaningful on most 100 point scales (e.g. letter grade dividing lines) and creating these "hot spots" based especially often at those points. That list is now past 1000 and needs serious rework. For my alcohol spreadsheet, after reading a couple of books on behavioral economics and having a few conversations with UX researchers, I decided to try to hybridize the approaches to heuristically limit the problems I encountered in the past. To do this, I decided to use a bin-based ranking system from which I could infer ratings. It enables me to pick a few points of comparison from each of several bins and then classify each entry in the bin where the quality seems most similar in aggregate. To get to what I think was your actual question though, I came up with 23 bins (with the optional 24th bin of zero for Southern Comfort, I guess) for several reasons. I wanted to actually use the entire range all of the way out to 1 and 23, but I both knew that I would be very reluctant to actually give either a 1 or 23 because of the implication of perfection or lack of any value. If I went with too coarse of a range (e.g. 1-5), I was afraid I'd create a strange distribution that was way too bunched toward the middle. Too fine of granularity and I wouldn't have enough data-points to limit hotspotting and wouldn't ever use the extremes. I also needed a prime number so that I couldn't easily divide and estimate on a 100 point (or similar) scale, thereby applying ratings rather than rankings. Additionally, I didn't want a number that was really close to a round number (e.g. 19, 31) because it would be easy to just approximate as ratings by rounding the scale up or down. This meant I needed a number ending in a 3 or a 7. Listing my options starting from zero gave me 3, 7, 13, 17, 23, 37, ... 17 seemed a bit small and 37 seemed a bit large, so I landed on 23. With ~1200 ratings, I've mostly avoided my previous problems so far, though I might be starting to have some issues crop up in certain categories that have too many entries. I do get sloppy at times as well and start referring to "points" (e.g. in this tasting where I say I like this a point better than regular Bushmill's). Sometimes this means I've actually forgotten I'm supposed to be looking at samples from bins and comparing in aggregate. Other times, it means I looked at the aggregate and used shorthand, just calling out a couple of key points of comparison. Still other times, it means I was too lazy to actually do the comparison and just took a guess based on what I thought was in the surrounding bins rather than actually checking. Overall though, it's been working quite well.

@WhiskeyLonghorn Yeah, I really enjoyed the 16 and 21, so I had some moderate hopes for this...but they weren't met.

Agreed. For $10-$15 less, Black Bush is a better sipper than this, particularly once the bottle has opened up. The 16 Year is a huge step up from this one (price and quality), but damn tasty!

May I ask where your 23 points rating is coming from?

Nailed it.