SciFi and Fantasy Book Club discussion

The Compare Books feature was brought up in Virtual Book Club. I thought it'd be interesting to hear how many of you use it regularly, and what you think of the results? Have you found Your Perfect Match? :D

If you don't know what I'm talking about, go to someone's profile and click on the the More dropdown. You'll find Compare Books there. It defaults to the All shelf for both profiles, I usually select the Read shelf for both, as that's more informative. Also note that if you only have very few books in common, the % might be high, but your reading taste could be significantly different.

I don't often look at it, because I don't think it means anything, but it's sometimes fun. I think I once found someone with a high %, 86% I think, but I don't remember who it was. Most people I've looked at have been in the high 70s or low 80s, that's probably because I've only looked at people who mostly read SFF.

I change my ratings a lot, and I sometimes rate based purely on personal enjoyment, and only sometimes try to take into account that someone might look at that rating, so I don't think comparing my ratings to someone else's means anything. Those who have a set rating system will probably get more out of this feature. I do sometimes compare my TBR shelf to someone's Read shelf, to see if there's something I might want to prioritize.

So, have you found high matches? Do you agree with them? Or have you been surprised by unusually low percentages when you thought it would be higher?

I use compare books for friend requests. I don't care about the percentages. I look at the actual books, how they were rated and read any reviews to see if the prospective friend and I have common ground and whether they have something interesting to say in the reviews. Whether the reviews are interesting matters more to me than whether I agree with them.

I use it when friends request me. I don't put much stock in the similarity rating that GR calculates (since I have a ton of reread duplicates and rating changes) but I do use it to see a few things:

1) If the person writes reviews, how often, and what kind of reviews. I don't want book reports, and I don't want "best book evar!" style either. Opinion and reason for it is great, and entertaining at the same time is awesome. :D

2) I sort by "read" books in common and check to see if our ratings for certain books we might have in common are similar, or if not, WHY not.

3) I like to see if their tastes are interesting to me. If I see a plethora of Dean Koontz or James Patterson, I'm gonna dig a lit deeper (if I don't already know the person in groups or otherwise) to see whether I'd get anything out of the friendship.

4) I always check to see if we have friends in common, and WHICH friends. That's very telling. (I know that's not a compare books feature, but I always use it after comparing books, so it counts for me.

Honestly, unless I already know the person through some sort of interaction, I am looking to see what them showing up in my feed is going to do for me. If they aren't reading books I find interesting, or writing reviews I find interesting, but they want to see those things from me, just follow.

I usually check the compare ratings after I read I comment I have enjoyed or disliked greatly just to get an idea of if I am reading it correctly.

Interesting, Anna, that you change ratings. I almost never do, I kind of consider my ratings a fixed moment of time shortly after I read the book. I have changed a couple after re-reading (I almost never do that either) and deciding that my teenage rating should not stand.

My rating is a chaotic mess and should not be trusted! I find that's probably true for the majority of GR users, so I don't feel too bad about it. I do appreciate people who rate and review in a (near) professional manner, so thank you to those of you!

So far all the uses everyone has mentioned seem perfectly logical, and it seems like no one looks at or cares about the %.

I use it to help try to help me better select books for myself by filtering whose reviews will likely align with my own. I'm not sure how great the algorithm is, but my best matches are similarly in the 80's. If the person is that high I can then dig in and look at the actual books to see where we agreed (or didn't). The best is someone who reviews a lot of SFF because then I get a bunch more data, plus I can learn over time where I tend to diverge from that person (say for example they like Star Trek novels more than I do, etc.). Like if I generally agree with someone and they rate something 1-2 stars, I'm pretty likely to skip it (and I might make time for a 5 star read).

Then when there's an opportunity to choose between multiple books (e.g. voting in the monthly polls) I can see if anyone who's on that list reviewed any of the available options. I find this more useful than looking at the book's average rating.

I once did this for films using Metacritic's list of film critic's top 10 lists from a given year. I then sought to watch as many of these films as possible and then run a program to compare my film scores to theirs. In doing so I found two critics who were the best match for me in that particular year and then sought out their advice in following years (unfortunately one of them had the website they work for close down, but the other still is a reviewer for Vanity Fair). I'd probably have to do this over a bunch of years to get a better fit, but it was enough work that I haven't repeated the experiment.

In theory I feel like it would be cool to datamine GR and find users with the highest similarity rating out there, but no such search exists to my knowledge.

Christopher wrote: "In theory I feel like it would be cool to datamine GR and find users with the highest similarity rating out there, but no such search exists to my knowledge."

People have asked for this, and I agree it would be really cool to have a top10 or top100 list of users whose "taste" is most similar to mine. But! Say it worked anywhere near as well as their recommendations feature? XD

I used it a handful of times when I first joined GR just to test it out. Honestly I don’t put much work into finding books to add to my tbr because I never seem to be at a loss for what to read. So not sure what else I would use it for.

I just did it for the people in this thread and my highest compatibility rating was with Anna at 85% for 300 common books read.

Bill wrote: "I just did it for the people in this thread and my highest compatibility rating was with Anna at 85% for 300 common books read."

OK now this is getting really interesting, because I get a 70% with you!

Does this thing even do anything except generate random numbers? :D

I use it pretty often when I visit somebody’s profile for the first time (or the second or third or fourth if I can’t remember what I saw the last time I was there). I’m more focused on # of books in common and whether or not they’re writing reviews. I especially enjoy looking for the books we rated very differently and reading any reviews they wrote for those, as well as any of the books high on the list when sorted by inverse popularity.

I don’t necessarily pay too much attention to the similarity percentage. I look at it, but don’t use it to make any decisions. I don’t really care if somebody has similar tastes to me and enjoy reading reviews from people with really different opinions. I’m really more focused on wanting to read reviews about books I’ve already read rather than reading them to get ideas for what to read next – I have no problems finding books to read next as it is! I have some friends though with similar tastes and I pay more attention when I see them rate something highly that I haven’t read before.

One other problem with the similarity % also is that some people might have many more years’ worth of data on their account and their tastes may have changed so it might not be an accurate reflection of how similar your tastes are now.

I tend to shy away from people who have large #’s of friends as I feel like I would get lost in the shuffle, especially since GR is apparently a bit buggy with showing all of your friends’ reviews on your feed, especially if you have lots of friends. Shying away goes quadruple if I look at their profile and see that they liked several lengthy reviews in the space of 0 minutes. I don’t want to get a like just for the awesomeness of showing up on somebody’s feed. I’m not offended if somebody isn’t interested in a particular review I’ve posted and therefore skips it and doesn’t click the “like” -- ‘cause how can you like it if you didn’t read it?!

As far as changing ratings, I never do either. Similar to what Hank said, I see it as “this is how I felt at the time I finished it” and leave it as a snapshot of that moment in time. I think I would drive myself bonkers if I rerated because my ratings are super subjective, based mostly on how I felt. For example, if I read an influential classic and hated it, I’ll give it a low rating even if I can objectively see that it had a lot of merit. As time passes and my memory of a great book fades more quickly than I expected, or a book I didn't care for sticks with me far longer than I expected, I sometimes feel like I should have rated them differently based on their sticktoitiveness, but I leave the ratings alone.

Anna wrote: "OK now this is getting really interesting, because I get a 70% with you!"

Could it be that you’ve also read a lot of comics in common? Because Bill has an exclusive shelf called “comics”, so if he’s looking at all shelves and you’re looking at only the “Read” shelf, maybe yours is excluding some of the comics you have in common?

YouKneeK wrote: "Anna wrote: "OK now this is getting really interesting, because I get a 70% with you!"

Could it be that you’ve also read a lot of comics in common? Because Bill has an exclusive shelf called “comi..."

That’s interesting. I limited it to just the read shelf. Time to do some testing.

It’s now killing me that I can’t compare two OTHER people’s books with each other because I want to troubleshoot this. But I don't want it badly enough to copy your shelves into Excel and do vlookups. ;)

For research purposes, here's everyone who has posted in this thread. Christopher's profile is private, so I can't compare with him, but everyone else is listed below. Common shelves: Read.

Shomeret 81% (96 books)
Becky 78% (175 books)
Hank 76% (258 books)
Bill 70% (300 books)
YouKneeK 82% (142 books)

YouKneeK wrote: "It’s now killing me that I can’t compare two OTHER people’s books with each other because I want to troubleshoot this. But I don't want it badly enough to copy your shelves into Excel and do vlooku..."

lol, I know how you feel, I also want to know! :D

Anna wrote: "YouKneeK 82% (142 books)"

Another discrepancy! I show 86% with you, using only the Read shelf. 142 books in common, but only 78 being used for the 86% accuracy since they're the only ones we both rated. I get the same # if I change it to "All" also.

Here’s what I’m getting listed, in inverse popularity order. Putting it in spoilers because it's long! (view spoiler)[
Regarding Ducks and Universes by Neve Maslakovic
The Philosopher Kings by Jo Walton
This Alien Shore by C.S. Friedman
City of Bones by Martha Wells
Remnant Population by Elizabeth Moon
Memory of Water by Emmi Itäranta
The Willful Princess and the Piebald Prince by Robin Hobb
A Night in the Lonesome October by Roger Zelazny
The Shadowed Sun by N.K. Jemisin
The Just City by Jo Walton
Babel-17 by Samuel R. Delany
All Clear by Connie Willis
The Anubis Gates by Tim Powers
The Kingdom of Gods by N.K. Jemisin
The Other Wind by Ursula K. Le Guin
Tales from Earthsea by Ursula K. Le Guin
Blackout by Connie Willis
The Broken Kingdoms by N.K. Jemisin
The Killing Moon by N.K. Jemisin
Among Others by Jo Walton
Cloud Atlas by David Mitchell
The Last Policeman by Ben H. Winters
Lord of Light by Roger Zelazny
Tehanu by Ursula K. Le Guin
The Moon Is a Harsh Mistress by Robert A. Heinlein
Deathless by Catherynne M. Valente
Watchmen by Alan Moore
To Say Nothing of the Dog by Connie Willis
A Natural History of Dragons by Marie Brennan
Harry Potter and the Chamber of Secrets by J.K. Rowling
Bunnicula by Deborah Howe
Tigana by Guy Gavriel Kay
Doomsday Book by Connie Willis
The City & the City by China Miéville
Three Men in a Boat by Jerome K. Jerome
The Tombs of Atuan by Ursula K. Le Guin
The Farthest Shore by Ursula K. Le Guin
Redshirts by John Scalzi
The Hundred Thousand Kingdoms by N.K. Jemisin
Assassin's Quest by Robin Hobb
Dirk Gently's Holistic Detective Agency by Douglas Adams
Royal Assassin by Robin Hobb
Kindred by Octavia E. Butler
Watership Down by Richard Adams
The Neverending Story by Michael Ende
Annihilation by Jeff VanderMeer
The Handmaid's Tale by Margaret Atwood
Anansi Boys by Neil Gaiman
The Martian Chronicles by Ray Bradbury
The Fifth Season by N.K. Jemisin
Jonathan Strange & Mr Norrell by Susanna Clarke
A Wizard of Earthsea by Ursula K. Le Guin
Assassin's Apprentice by Robin Hobb
Neuromancer by William Gibson
Do Androids Dream of Electric Sheep? by Philip K. Dick
Harry Potter and the Deathly Hallows by J.K. Rowling
The Graveyard Book by Neil Gaiman
The Lord of the Rings by J.R.R. Tolkien
The Ocean at the End of the Lane by Neil Gaiman
Good Omens by Terry Pratchett
The Road by Cormac McCarthy
Ready Player One by Ernest Cline
Dune by Frank Herbert
The Martian by Andy Weir
The Night Circus by Erin Morgenstern
American Gods by Neil Gaiman
The Princess Bride by William Goldman
Jane Eyre by Charlotte Brontë
Fahrenheit 451 by Ray Bradbury
Mockingjay by Suzanne Collins
Catching Fire by Suzanne Collins
Harry Potter and the Half-Blood Prince by J.K. Rowling
Harry Potter and the Order of the Phoenix by J.K. Rowling
Harry Potter and the Goblet of Fire by J.K. Rowling
Harry Potter and the Prisoner of Azkaban by J.K. Rowling
Pride and Prejudice by Jane Austen
The Hunger Games by Suzanne Collins
Harry Potter and the Sorcerer's Stone by J.K. Rowling (hide spoiler)]

I get 86% with Bill when I only compare my read graphic novels to his Comics shelf :)

I get 83% for "All" with you, YouKneeK. My inverse popularity list for "Read" is also different from yours XD

Anna, we are 80% (242 books)

I wonder if they didn’t account for re-reads or multiple editions consistently in their comparison query? Like maybe if you read a book twice and I only read it once, it counts our rating comparison on that book twice when you run it and only once when I do?

Not sure that really makes sense though! I don’t have any books with either multiple read dates or multiple editions on my shelves.

OK (sorry everyone else!), YouKneeK, what do you get for "fantasy" and "scifi" ? They're under a 100 books in common, so the calculation should at least use the same subset?

fantasy: 75% (56 books)
scifi: 77% (47 books)

Christopher wrote: "Anna, we are 80% (242 books)"

I can't check on my end, so I don't know if it's different :(

Anna wrote: "OK (sorry everyone else!), YouKneeK, what do you get for "fantasy" and "scifi" ? They're under a 100 books in common, so the calculation should at least use the same subset?"

Fantasy matches!
Scifi matches # of books (47), but I get 76% instead of 77%.

Conclusion: GR is nuts and they use fake computers. :)

At least it's closer! I don't have rereads on those shelves (or I shouldn't have, I'll check scifi to make sure), so that could be part of it. Or when there are more than 100 books in common they cap it randomly :D

The 100 limit probably makes the most sense. I could see them setting limits to keep processing time under control.

But 70% vs 85% is still wild! I once got a friend request from some random person I didn't know from anywhere. They said "I've never seen such a high compare books % so I had to send you a friend request." I went to check and it was something like 63% and I remember thinking the poor thing must have had abysmal matches! :D But this kind of might explain why.

does it work if my what about those who don't rate many books or have their profile set to friends only?

LOL, yeah, I will now be paying even less attention to the % than I did before!

There has to be some logic, as opposed to being completely random, to which books it chooses for the comparison because I get the same % every time I compare with you. (And I’ve now done that so many times that GR may ban me for stalking!)

Maybe the “consistent” thing it focuses on for choosing the books is something that would be different for each of us – like more recently-read books for the person running the comparison, or a certain # of books from different star ratings.

But I have now run out of energy for analyzing this strangeness. ;)

You can't match against private profiles unless you're friends. And of course if you don't rate, then there's nothing to compare. Although GR probably thinks no rating for both parties means 100% match!

I also thought about what the logic might be. It could be date added to shelves, date rated, it could be anything! I doubt GR would know if we asked.

What I've learned today is that doing matches against certain shelves makes more sense if you want to look at the percentage. That's what I usually do anyway if I use compare books, since I'd mostly be interested in how they rated SFF books, not literary fiction or mysteries, etc.

Bill wrote: "YouKneeK wrote: "Anna wrote: "OK now this is getting really interesting, because I get a 70% with you!"

Could it be that you’ve also read a lot of comics in common? Because Bill has an exclusive s..."


As I recall, the comparison is run on a limited number of books, maybe 100, and they are selected randomly. So when you run the comparison with X the comparison will be run with a different set of books than when X runs a comparison with you. The numbers are unlikely to match exactly. As other people have said, I don't really care about the number, I care about whether we read similar books, whether they write interesting reviews, etc.

Faith, yes, it makes sense that the more books you have in common, the less the % matches, if the subset is random.

I use this all the time and love it. Especially when I get friend requests. It doesn’t usually determine if I’ll accept them or not....

But I secretly judge people on how they rate certain books. 😂

I use this frequently as well, both when I get friend requests and when someone new shows up in one of my groups and I want to figure out whether we have similar tastes. Like most everyone else, it sounds like, I find the percent agreement interesting but not super insightful. I think it's a lot more useful to see what books we have in common, which books we've rated very differently, and whether/how the person reviews books.

Sarah wrote: "I secretly judge people on how they rate certain books"

Don't we all? :D

I use this when people ask for recommendations, just to see if we might like the same things.

And when people ask to friend me, I have a look at their books - firstly to see that they're actually a genuine Goodreads person, and secondly to see if we have books in common.

Becky wrote: "I use it when friends request me."

But I don't use it to deny a friend request. I just check it out of curiosity. Well … maybe if I saw the person 5-starred a bunch of books I hate maybe I'd ignore the request (hasn't happened yet).

Highest match I've seen I think was in the low 80%. Usually it's much lower.

Micah wrote: "Becky wrote: "I use it when friends request me."

But I don't use it to deny a friend request. I just check it out of curiosity. Well … maybe if I saw the person 5-starred a bunch of books I hate m..."

I do! If I don't already know/interact with the person and it's a random request out of the blue, I don't have any other way to know what kind of reader they are. If people simply rate books, that's fine, but I want to know more than that, and so I generally friend people who write reviews consistently. Even with people that I've been friends with for YEARS on this site, if they just leave a rating, I find myself wondering why, and why they rated the way they did, despite being pretty familiar with their tastes. With strangers who only rate, there's literally nothing for me to tell from a rating only.

I am not a friend collector, and I don't want to have my feed filled with stuff I'm not interested in. Or nothing worthwhile to me.

OH! I also check their rating stats on their profile so I can see whether they are a "every book is someone's 5 star baby, or else I just won't rate it" reader or if they have a fair mix of ratings. I'm not interested in someone who magically only reads 5 star books OR someone who is hypercritical and rates everything super low because nothing is ever perfect enough.

We don't have to agree on every book (I can vary wildly between absurdly nitpicky and ridiculously lax, depending on the book) but if they don't write reviews or have ratings that are consistently opposite to mine for my "qualification books" then I'm not getting anything out of this "friendship", and they should just follow.

I only check it if I think it seems like this person and I agree a lot. If we do, and we have a high compare, then I tend to give their reviews a bit more weight in terms of what I might prioritize on my own TBR. But this is such a rare occurrence that I think I've done it with 5 or 6 people since I joined.

I kinda like the idea of it, but it hadn't seemed to give great results, so I've kinda not used it in quite awhile.

That said, feel free to compare with me for the purpose of testing, and/or to ask me to compare against any of you from my direction!

Also, if anyone does get a great match with me, please let me know!!

Cheryl wrote: "Also, if anyone does get a great match with me, please let me know!!"

I was pretty sure the result was going to be good, but I checked my scifi shelf against your read shelf and got 85%. I think I'm going to go through your read shelf to see if I can find some fun scifi I haven't read yet :)

(This doesn't help you because I don't review, sorry!)

I rarely use this feature but I would welcome anyone who would like to add me as a friend and compare books.
Warning: I read just for entertaining purposes so it's chaos, LOL...

Here from the group there was a similarity of taste of 75% with Gabi. And 77% with Eva.

I got very good matches with everyone on this list that I checked so far (anna, Allison, Cheryl, Kaa): between 77-86% on their default and usually about the same or a little lower when set it for our “read” shelves. I was especially interested in the comparison with Cheryl, since we both seem to read a lot of SFF and children’s books. I think the 86% was with Allison on default. But we have read an astonishing number of the same books.

Sarah wrote: "I use this all the time and love it. Especially when I get friend requests. It doesn’t usually determine if I’ll accept them or not....

But I secretly judge people on how they rate certain books. 😂"

Totally! And we should probably start a thread (if it doesn't exist already) on trigger books.

One of my most obvious is Jonathan Strange & Mr Norrell. Many of my friends loved that book (Alison, how could you!) but the only thing that book is good for is being used as a weapon both literally and emotionally. It even made my kindle heavier. Totally off topic, too much wine, and yes I am being silly.

Haha! Idk man. Regency, evil fae, riffing on fairy tales, bookish main characters...it'd be hard to upset me with that combo. Also I was cooking aaaall day while listening to it and I've found that books I listen to while I'm making food I like more than I might if I read them just before bed. Cook and book are just too delicious together.

Hank wrote: "Totally! And we should probably start a thread (if it doesn't exist already) on trigger books.

One of my most obvious is Jonathan Strange & Mr Norrell. Many of my friends loved that book (Alison, how could you!) but the only thing that book is good for is being used as a weapon both literally and emotionally. It even made my kindle heavier."

LOL! Unpopular opinion time! I feel the same about The Night Circus, only I would use it as psychological attack instead of emotional (because it's lacking the latter completely). Listening to Jim Dale "perform" the audio made me want to claw my own ears off my head. If one was feeling particularly cruel, one could just play the 2nd person circus description sections on repeat until their "audience" is nothing but a quivering puddle of goo muttering about clocks and pen scratches and tents...

I'm all in on your new thread idea, Hank!

Lol, Hank! I didn't like JS&MN, but I can forgive people for enjoying it. I have a harder time with Wheel of Time, which is equally weapon-like and which I have resented since high school.

Yes please start a thread about books you would weaponize 😄