lou_fine Posted March 21, 2023 Share Posted March 21, 2023 On 3/16/2023 at 10:09 AM, xriddx said: For instance on one comic I'm looking at the price of a sold 9.8 (during the 2022 mania peak) it shows as $50k price. Then you see 9.0s going for about $5k during that time. This is a price difference of 10 times for some slight variations of issues with the books condition. Especially when we've seen at times when a CGC 9.0 graded copy of a book can turned itself into a CGC 9.8 graded copy of the book. Clearly a sign that you are then buying the label, instead of buying the book since it is EXACTLY the same book, but only with a different label. xriddx 1 Link to comment Share on other sites More sharing options...
Popular Post Brock Posted March 21, 2023 Popular Post Share Posted March 21, 2023 On 3/20/2023 at 3:25 PM, valiantman said: Oh, I recognized that. Ouch. That's pretty defensive... I may regret this, but I'll take the bait. Let me start by saying that I have a lot of admiration for your contributions here and elsewhere. You put a lot of thought and work into your efforts, and while I don't always agree with your analysis or conclusions, I recognize the importance of the effort, and the need to propose hypotheses that can be tested in the hobby. Your PhD in information quality gives you some strong credentials to lead us in this space. However, a blogpost is not an academic publication, and so has not been subjected to peer review. So - in the spirit of "making improvements" and enjoying a collaborative and collegial attempt to build some consensus - let me suggest a few things. I do not mean to question the information or quality here, but to reflect on the analytical approaches and conclusions drawn, which I think could use some further work. I will also do my best to suggest how these could be strengthened. First, though, I should point out that I think your blogpost doesn't quite address what I felt was the main thrust of the thread as I see it. I recognize that the nature of threads is to have several discussions going on at once, so we may have different perspectives on this. Your blogpost suggests that 9.8s are not the best from an ROI perspective, and you present some interesting data in support of that theory. I think what at least some of us were suggesting was that purchasing 9.8s was (to quote myself) "a good way to preserve my initial investment/outlay." This thought was echoed by others, such as @D2 who spoke of "volatility" or @grendel013 who suggested it was critical to "trade or sell your current copy without losing money reselling." Thus, while you are addressing ROI and highest possible returns, we were discussing the preservation of capital, which I think you'll agree is a different thing. I note this largely because it could be relevant to parts of the discussion later on. Having said all that, I think your blogpost is definitely interesting, and does provide evidence in favour of your hypothesis. However, in general, I believe that your analysis and conclusions make too much of your limited data. In the following section I try to do two things: Address what I see as some of the weaknesses in your case Suggest some approaches that might help to overcome those weaknesses +++ The first issue I see - and I suspect it is one you will acknowledge, given your academic background - is that the sample size for your case is very small. With only 12 books discussed, we have only an extremely small sample to support your hypothesis. That does not, of course, mean that your analysis is wrong, but that the limited quantity of evidence makes your case very tenuous, and subject to wild swings in accuracy. This is related to a number of other "sample" issues in your blogpost. All of the books you have selected for your sample are keys, "blue chip" books that are widely traded in the marketplace. In this sense, they are not representative of the market as a whole. Thinly traded books, for example, are likely to be much more volatile in their pricing, and could produce sales results that contradict your findings with "blue chip" books. Without having included those books in your analysis, we can't say whether your hypothesis is more generally valid or not. This is important, as the vast majority of books traded in the comic book marketplace are not "blue chip" keys. Similarly, your sample is all made up of older books - 10 silver, 1 bronze and 1 copper. As a result, we also can't say whether or not your model fits with modern books. Indeed, several people commenting in the thread, including myself, @Stefan_W, @Dr. Balls, @D2, @Aman619, all suggested that modern books behave differently in the market than older books. We have not provided data to support this assertion, but neither does your blogpost provide data to contradict it. Finally - and perhaps most importantly - on the sample size question is the fact that the 12 books you've selected have all been rising in price over the timeframe you're looking at. This is also not representative of the market, as many books fall in price. Indeed, my own hypothesis - based on your work - would be that when prices on a book fall, grades lower than 9.8 fall faster (i.e., at higher percentages). This means that we could conclude the opposite of your blogpost... if prices are falling, then 9.8s could have the highest ROI (even though those ROIs are negative). This is a critical consideration when thinking about ROI, and it is not addressed in your sample or your analysis. In order to strengthen your model, I would suggest a large sample - perhaps 1,000 books to pluck a number from the air. Those 1,000 books should be drawn from all ages (including modern), should include not only keys but a wide range of key/desirable and non-key/less desirable books, and should include books that are both rising and falling in price over time. This might provide us with a more robust data set that allows some more genuine analysis. +++ There are some other concerns which I see with the approach in your blogpost, which are not explicitly related to data. For example, your core argument is that with books you have sample, lesser grades rise in value at a higher rate (%) than 9.8s. This is a well-known feature of mathematics and economics, sometimes referred to as "convergence" or "unconditional convergence" and it has largely fallen out of favour as a tool for analyzing trends in the market or economy. The classic example is its application in the context of National GDP rates. In 2023, for example, the International Monetary Fund anticipates that the US economy will grow by 1%, while the economy of Libya economy will grow by 17.9%. Because of the low starting value of the Libyan economy, any increase will be represented by outsized percentage returns. Basically, this tells us something about math, but not about real returns as you suggest. The alternate view - that these percentages represent a genuine trend - leads to logical fallacies when played out over time. They suggest (in the GDP case) that the Libyan economy will eventually be larger than the American economy, or (in the case of your blogpost) that CGC 4.5s will eventually be worth much more than CGC 9.8s. One way to try to correct this in your work might be to consider other explanations for the rise of lower grade books. Forces like supply and demand could figure here, for example. The number of people who are looking to buy (i.e., demand) a $100,000 key in CGC 9.8 is small, but the number of people looking to buy that same book in a lower grade for $1,000 will be larger. This suggests that price may be a more important factor in driving demand for lower grade books than the assigned grade. This has an implicit suggestion that as lower grade copies see more demand, their prices will rise faster - until they reach a point where higher graded books become more attractive because of their lower price point. This suggests that your research could find ways of exploring this problem by thinking of pricing tiers rather than grading tiers for future analysis. Finally, there is one other key feature of your blogposts' model that raises questions in my mind. If your model is valid, it should have some predictive capacity, but - at least based on the 12 datapoints in your blogpost - I don't see this. In fact, I see the opposite - a lot of noise within the data. If the model is robust, I think I should be able to predict which grades will see higher price rises than others, but instead I see high levels of randomness. In your 12 sample books, for example, I see that 5 rose in value fastest in a grade at CGC 5.0 or above, and 7 that rose in value fastest in a grade below CGC 5.0. On a 10-point grading scale, that's pretty close to 50/50, and this epitomizes a random result in my mind, like flipping a coin 10 times in a row and getting 6 heads and 4 tails. For your model to have UTILITY - that is usefulness - it needs to be better than random, and it needs to help those in the marketplace determine which grade of a book they SHOULD be purchasing. Without some attempt to explain why Amazing Fantasy #15 sees the highest increases in CGC 8.5 while Avengers #1 sees the highest increases in CGC 2.0, your 12 datapoints are interesting, but not particularly useful. It may be that a larger sample size will begin to address the utility challenge, but in my mind, a focus on making your data more useful to recipients would be tremendously helpful. ++++ Again, all of this is offered in the context of collegial and collaborative peer review. If I can help to shape and strengthen your future efforts in this field, I am at willing to help where I can. toro, Dr. Balls, MAY1979 and 2 others 5 Link to comment Share on other sites More sharing options...
Brock Posted March 21, 2023 Share Posted March 21, 2023 TLDR version: I suggest some areas where I think @valiantman's model could use some strengthening, and make a genuine effort to help think through aspects of this. (And pray that it doesn't descend into another Board flamewar...) xriddx and toro 2 Link to comment Share on other sites More sharing options...
valiantman Posted March 21, 2023 Share Posted March 21, 2023 (edited) On 3/21/2023 at 10:25 AM, Brock said: TLDR version: I suggest some areas where I think @valiantman's model could use some strengthening, and make a genuine effort to help think through aspects of this. (And pray that it doesn't descend into another Board flamewar...) You misunderstand my meme. I wasn't asking for how I can spend the rest of MY life working out YOUR (inevitable, easy-to-say, and helpful-but-not-practical) suggestions. Your suggestions of using 1,000 different books and all the sales records for all those books and all the different years and ups and downs of the market are certainly valid. Trying "perhaps 1,000 books" as you suggest - yeah, that's nothing, right? Easy to say, for sure... at maybe 20 to 30 minutes of research per book, that's between 300 to 500 extra hours of work. That's all. No... what I meant (from the very beginning, even before your suggestions were typed out) was... YOU DO IT. Edited March 21, 2023 by valiantman Link to comment Share on other sites More sharing options...
THE_BEYONDER Posted March 21, 2023 Share Posted March 21, 2023 Dr. Balls and RockMyAmadeus 1 1 Link to comment Share on other sites More sharing options...
valiantman Posted March 21, 2023 Share Posted March 21, 2023 (edited) On 3/21/2023 at 10:38 AM, THE_BEYONDER said: All you've gotta do is come up with a $2,000 car that gets 200 miles to the gallon. That's it. Just do that one extra thing. Oh, and when you're done, remember it was my idea... see, that's how collaboration works... -AND- You're welcome. Edited March 21, 2023 by valiantman THE_BEYONDER and xriddx 2 Link to comment Share on other sites More sharing options...
Brock Posted March 21, 2023 Share Posted March 21, 2023 On 3/21/2023 at 11:32 AM, valiantman said: You misunderstand my meme. This seems likely... the message you intended to send ("YOU DO IT") was different than the message I received ("you like to complain about other's work but have zero skills for making improvements"). On 3/21/2023 at 11:32 AM, valiantman said: No... what I meant... was... YOU DO IT. So, if I'm understanding you correctly now, we were all having a conversation about the relative price stability of CGC 9.8s. You suggested we were wrong, because blogpost. I explained that I (and others) didn't find the blogpost particularly valuable in this discussion, because reasons (see my world's longest thread post). But now you want us to work to make your blogpost, which we don't agree with, better? On 3/21/2023 at 11:32 AM, valiantman said: YOUR (inevitable, easy-to-say, and helpful-but-not-practical) suggestions. This suggests that - even before you put your blogpost in the thread - you were aware of the idea that your blogpost was not statistically strong (see: sample size), theoretically unsound and logically fallacious (see: unconditional convergence) and not particularly useful (see: utility). Perhaps you can help me understand why it was important to link to your blogpost then? I'm genuinely not trying to provoke conflict, but to understand what your key message is here... I'd love to have your help, expertise and input into some research projects I'm working on, but am not sure what to take away from this discussion. xriddx 1 Link to comment Share on other sites More sharing options...
valiantman Posted March 21, 2023 Share Posted March 21, 2023 (edited) On 3/21/2023 at 12:49 PM, Brock said: Perhaps you can help me understand why it was important to link to your blogpost then? Because the topic is "9.8 vs. other grades" and the point of the blogpost was "9.8 vs. other grades". It never fails that no matter what I post, how much data I use, or how long I spend researching the topic, SOMEONE (not always you) will come along with five minutes of suggestions for how to improve something that would take hundreds of hours to complete. I'm talking about 20 years of history here, not "Brock is my enemy", but "people with suggestions who do no work are not as helpful as they think they are". Five minutes of "helpful direction" is obviously worthless in practical terms, because those five minutes are always presented like "did you ever try this" or "did you think about that", and the answer is always "yes, of course, I thought about it. It would take months do that, at no pay, and inevitably, someone else with five minutes of thoughts would come along with months of additional work (that they won't do themselves) that still wouldn't be acceptable to the people who can always picture and describe perfection but can do nothing to achieve perfection." Talk-the-talk, never walk-the-walk. So, a meme, where someone is surrounded by jewels, which everyone can easily see, but they cross their arms and refuse to actually do any mining themselves. "Here's how to make things better" but "I don't have time to do this myself, so please, waste all your time on something that was obvious from the start, and make sure I get credit for my suggestion because it was so helpful". "Did you think about making the car cheaper with better gas mileage?" - basically the same thing. Yeah. 20 years later, it gets old. Edited March 21, 2023 by valiantman grendel013 and xriddx 2 Link to comment Share on other sites More sharing options...
lou_fine Posted March 21, 2023 Share Posted March 21, 2023 On 3/15/2023 at 9:23 AM, NP_Gresham said: I just don’t get why there are reams of copper/modern drek in 9.6 It makes much more sense to send those on a 9.8 screen. Isn't there like a minimum quantity or number of books that you have to meet in order to submit for a 9.8 screening? xriddx 1 Link to comment Share on other sites More sharing options...
Brock Posted March 21, 2023 Share Posted March 21, 2023 On 3/21/2023 at 2:56 PM, valiantman said: It never fails that no matter what I post, how much data I use, or how long I spend researching the topic, SOMEONE (not always you) will come along with five minutes of suggestions for how to improve something that would take hundreds of hours to complete. I'm talking about 20 years of history here, not "Brock is my enemy", but "people with suggestions who do no work are not as helpful as they think they are". That's good, because in my 10 years here I don't think I've ever interacted with you about anything other than our shared love of early Valiant. xriddx 1 Link to comment Share on other sites More sharing options...
lou_fine Posted March 22, 2023 Share Posted March 22, 2023 (edited) On 3/15/2023 at 10:42 AM, Dr. Balls said: When I was into 9.8s, the theory for me was that was the best way to retain it's future value - then again, I was collecting those way before the stimmy money comic book run-up, so the vast multiples from 9.6 to 9.8 wasn't what it is now. Nowadays, I'll probably never bother with another 9.8 again unless it's such a good deal I can't pass it up. Playing a bit of devil's advocate here, but wouldn't taking the exact opposite approach actually be better at not only preserving capital, but also give you a much better chance at acheiving a positive ROI going forward. In other words, if you pay top dollar to buy CGC 9.8 graded copies of MA books and yet these books don't have any value in say anything below CGC 9.0, I don't think these books will be gaining much value going forward longer term. Especially when it would appear that the underlying book itself has no real value per se as it's really all in that big big number at the top left hand corner of the slab. Even more so from the sense that the majority of these books are simply nothing more than flavors of the day and speculators will be moving onto the next flavor as the speculative buying cycle on these type of books are relatively short. Now, if you take that same money and buy into say a low grade and possibly even Restored RAW copy of a much older HTF GA book for example that is selling at single or even double digit multiple to condition guide, wouldn't that book actually have a greater potential of not only preserving your capital, but also possibly giving you a better chance of acheiving a positive ROI going forward longer term. Especially when it would appear that all of the value is derived from the underlying book itself with none at all coming from the wishy washy CGC label itself. Edited March 22, 2023 by lou_fine Link to comment Share on other sites More sharing options...
THE_BEYONDER Posted March 22, 2023 Share Posted March 22, 2023 On 3/21/2023 at 4:21 PM, lou_fine said: Isn't there like a minimum quantity or number of books that you have to meet in order to submit for a 9.8 screening? 25 ADAMANTIUM 1 Link to comment Share on other sites More sharing options...
Stefan_W Posted March 22, 2023 Share Posted March 22, 2023 On 3/22/2023 at 1:32 AM, lou_fine said: Now, if you take that same money and buy into say a low grade and possibly even Restored RAW copy of a much older HTF GA book for example that is selling at single or even double digit multiple to condition guide, wouldn't that book actually have a greater potential of not only preserving your capital, but also possibly giving you a better chance of acheiving a positive ROI going forward longer term. Especially when it would appear that all of the value is derived from the underlying book itself with none at all coming from the wishy washy CGC label itself. If you look at it in terms of ROIs and basic marketplace principles I see GA books in general and SA keys as being like blue chip stocks that continue to modestly appreciate over time. Moderns are typically more volatile so they are essentially higher risk/higher reward books. So if you go back to 2019, if you put $400 into a GA book it would be worth $500-$600 today. If you put that same money into a 9.8 copy of Ultimate Fallout 4 you would have a book around the $1800-ish range in value. If you put that money into 9.8 copies of other modern keys you may be up a bit depending on which one you picked. I think the basic idea is that you can do well with 9.8 moderns if you pick well, but maybe not so much if you catch it too high on the type train when pricing is inflated. GA are a bit easier in the sense that it is harder to go wrong. I am leaving out buying in 2021 since that was an anomaly - almost no 2021 purchases look particularly great right now after the large price correction. Link to comment Share on other sites More sharing options...
lou_fine Posted March 23, 2023 Share Posted March 23, 2023 On 3/15/2023 at 10:42 AM, Dr. Balls said: When I was into 9.8s, the theory for me was that was the best way to retain it's future value - then again, I was collecting those way before the stimmy money comic book run-up, so the vast multiples from 9.6 to 9.8 wasn't what it is now. On 3/15/2023 at 12:00 PM, Brock said: This is it for me... I buy 9.8s of the specific things I collect, especially with Copper and Modern books. Although the intent of collecting these is not resale, the fact that the 9.8 market commands such premiums generally makes them a good way to preserve my initial investment/outlay. On 3/17/2023 at 8:36 AM, THE_BEYONDER said: I don’t know. In dark times, that 9.8 is going to take a much larger hit than the exponentially cheaper grades below it. Rather have my 9.6 drop from $1000-$100, than my 9.8 drop from $10,000-$1000 After reading that thread, "What Books Had Tanked the Hardest", I would tend to lean much more towards Beyonder's point of view here: Especially when you see most of these former red hot movie related hyped books dropping by 75% up to a speculative Nasdaq like drop of 90% from its recent highs. Personally, if I was looking to preserve my initial investment, I would tend to avoid books where their values are based upon either their grades, movie/TV or other media related hype, and blatant auction house overhype like what took place with the Promise Collection. My theory has always been the best way to preserve your initial outlay and to hopefully obtain some future ROI gain going forward is to acquire books whereby the percentage of the value derived from the underlying book itself is maximized to as high of the total value as possible, and the percentage of the value derived from the subjective grade assigned to the book is minimized to as low of the total value as possible. Dr. Balls 1 Link to comment Share on other sites More sharing options...
lou_fine Posted March 23, 2023 Share Posted March 23, 2023 (edited) On 3/16/2023 at 1:54 PM, Aman619 said: if you buy slabbed comics with the awareness of someday trying to sell them, higher grades generally hold or increase in value. For Moderns, there is no greater assurance of a decent chance of getting your money back someday than buying a 9.8. They arent ridiculously expensive and you can be sure you made the best decision you could to ensure a positive return. Except: if the book WAS hot when you bought it and the movie tanked, (etc). or not that many people ever really caught onto it and just dont want to buy it The problem that I see with this theory here is that when it comes to the overwhelming vast majority of the Modern market, unless the book becomes super hot going forward longer term (e.g. UF 4), they generally tend to become forgotten drek over time. With the MA market, it's all about the flavor of the day, and unfortunately, they tend to spoil rather quickly because their Best Before Expiry dates are usually quite short, especially since new hot items are constantly going onto the shelves all the time. Edited March 23, 2023 by lou_fine MAY1979 1 Link to comment Share on other sites More sharing options...
Dr. Balls Posted March 23, 2023 Share Posted March 23, 2023 On 3/23/2023 at 3:47 PM, lou_fine said: After reading that thread, "What Books Had Tanked the Hardest", I would tend to lean much more towards Beyonder's point of view here: Especially when you see most of these former red hot movie related hyped books dropping by 75% up to a speculative Nasdaq like drop of 90% from its recent highs. Personally, if I was looking to preserve my initial investment, I would tend to avoid books where their values are based upon either their grades, movie/TV or other media related hype, and blatant auction house overhype like what took place with the Promise Collection. My theory has always been the best way to preserve your initial outlay and to hopefully obtain some future ROI gain going forward is to acquire books whereby the percentage of the value derived from the underlying book itself is maximized to as high of the total value as possible, and the percentage of the value derived from the subjective grade assigned to the book is minimized to as low of the total value as possible. I would agree with you and BEYONDER on that one. The 9.8 game now is so very different than 8 years ago, the examples you guys point out are frightening. Link to comment Share on other sites More sharing options...
Aman619 Posted March 23, 2023 Share Posted March 23, 2023 On 3/23/2023 at 5:53 PM, lou_fine said: The problem that I see with this theory here is that when it comes to the overwhelming vast majority of the Modern market, unless the book becomes super hot going forward longer term (e.g. UF 4), they generally tend to become forgotten drek over time. With the MA market, it's all about the flavor of the day, and unfortunately, they tend to spoil rather quickly because their Best Before Expiry dates are usually quite short, especially since new hot items are constantly going onto the shelves all the time. times have evolved for MA 9.8s sure. But while we see that you can lose a lot f money with ill-timed purchases, (like buying on the downslope of lesser demand after prices jumped up wildly and after the movie opens and/or bombs, , or when buying during a once in a lifetime (we hope) global change of lifestyle Event that is fueling demand irrationally. You still have 2 choices when collecting Moderns: 9.8s (you bought the best, now sit back and let the market go up or down) or raw -- HG or any grade you like -- but as cheap as you can get them. You can slab the HG ones it later if advantageous if you want; or sit cozy knowing "you HAVE a copy too!" if it takes off. lou_fine 1 Link to comment Share on other sites More sharing options...
lou_fine Posted March 23, 2023 Share Posted March 23, 2023 On 3/22/2023 at 8:57 AM, Stefan_W said: If you look at it in terms of ROIs and basic marketplace principles I see GA books in general and SA keys as being like blue chip stocks that continue to modestly appreciate over time. Moderns are typically more volatile so they are essentially higher risk/higher reward books. So if you go back to 2019, if you put $400 into a GA book it would be worth $500-$600 today. If you put that same money into a 9.8 copy of Ultimate Fallout 4 you would have a book around the $1800-ish range in value. If you put that money into 9.8 copies of other modern keys you may be up a bit depending on which one you picked. I think the basic idea is that you can do well with 9.8 moderns if you pick well, but maybe not so much if you catch it too high on the type train when pricing is inflated. GA are a bit easier in the sense that it is harder to go wrong. Like you correctly stated, it really depends upon what books you pick, but in your example here, you are using what is probably the winningest book out of tens of thousands of books that came out in the Modern Age. Likewise, and probably just as misleadingly representative as if I had chosen this "fire sale dump" of 18 copies of this water stained book that still somehow managed to fetched $418.25 back in 2009 for my GA example here: https://comics.ha.com/itm/golden-age-1938-1955-/chamber-of-chills-19-multiple-file-copy-group-harvey-1953-condition-average-gd-vg-total-18-comic-books-/a/19062-12153.s?ic4=ListView-ShortDescription-071515 Especially now that we've even seen CGC 2.5 graded copies of this book fetching 5-figures which mmeans the lucky winner of this lot here should have no problem at all pocketing well into 6-figures for this lot of 18 copies if they had decided to hold onto all of them. Of course, no doubt that's just an one-off and also requires a lot of luck in terms of snagging a red hot book become it gets recognized by the collecting base. A more ususal case is the willingness of the buyer to step up and open their wallet for say a book like 'Tec 31 with its classic Batman cover in CGC 2.0 back about 5 years ago when it was sitting in the $20K's price range and hold onto it long enough to where CGC 2.0 graded copies are now approaching 6-figures, if not there already. Of course, not all of your GA books would have moved up by 25% or 50% since 2019 as per your example here. Like always and as you have already alluded to, it's about picking the right books to buy, but I will say in watching the GA books that I was interested in during last week's CC Event Auction, they all seem to be continuing on their upward trajectory with many selling for record prices in their grade. The one thing that these auctions do confirm for me though so far to date, is that capital preservation and positive ROI return seems to stand a much stronger chance when your purchasing decision is based more upon the underlying book itself, as opposed to that big big number on the top left hand corner of the slab. Link to comment Share on other sites More sharing options...
lou_fine Posted March 27, 2023 Share Posted March 27, 2023 On 3/22/2023 at 6:07 AM, THE_BEYONDER said: On 3/21/2023 at 1:21 PM, lou_fine said: Isn't there like a minimum quantity or number of books that you have to meet in order to submit for a 9.8 screening? 25 Oh wow...........that's a pretty large quantity you have to send in for a screening and a pretty easy money maker for CGC for not much work. Do they all have to be the same book or can it be a combination of different books? Even then, I can't think of 25 different books that I would want to do a grading screen on that would be worth all of the time, cost, and hassle. Link to comment Share on other sites More sharing options...
Sigur Ros Posted March 27, 2023 Share Posted March 27, 2023 (edited) On 3/26/2023 at 10:33 PM, lou_fine said: Oh wow...........that's a pretty large quantity you have to send in for a screening and a pretty easy money maker for CGC for not much work. Do they all have to be the same book or can it be a combination of different books? Even then, I can't think of 25 different books that I would want to do a grading screen on that would be worth all of the time, cost, and hassle. They don't have to be the same book. They will probably change that maximum to 1, as they continue to adopt features of the competition. Edited March 27, 2023 by Sigur Ros lou_fine 1 Link to comment Share on other sites More sharing options...