Do Review Swaps Influence the Top 1000 Books? And Other Adventures of Ratings on Royal Road

Aug 3, 2025

—

in Royal Road Analytics, Royal Road Tools

After analyzing the Genre Rising Stars lists, comparing books to each other, and creating a bot for Discord that does a lot of useful stuff, I decided to focus on the reviews.

For my first exploration, I took the Top 1000 books on Royal Road. And if you want to hear the answer to the question asked in the title, I can give it to you right away. But I hope you’ll stay here for longer, because there are some other interesting stats there to explore.

So, do review swaps influence the Top 1000 books? In my opinion: not really. They do influence some books among the top-rated. But only a small fraction of them (less than 6%).

Disclaimer

All the data was collected within the same approach I use for collecting data for my Book Comparison Tool and Genre Rising Stars Checker. The scripts have delays between the requests agreed upon with the mods of Royal Road, not to overtax the website’s servers.

No reviewer usernames will be shared in this post, as it is against the Royal Road rules. However, some cases of obvious downvoting and gaming the system are and will be reported to the mods.

Dataset Overview

• Total reviews: 71,956
• Unique reviewers: 32,942
• Average review rating: 4.585/5.000
• Review swaps: 1,944

On average, each reviewer left a bit more than 2 reviews. There are leaders who have reviewed 5-10% of the Top 1000 books. Most of them give ratings all over the board; however, there are outliers. There is a user who left fifty-two 5* reviews and a few more who left thirty or more reviews. On the other side of the spectrum, there are users who leave mostly reviews with ratings below 3*.

Also, there are 28 users whose average rating for the Top 1000 books is at 2.5* or less. And that’s taking into consideration only people who left 5 reviews or more. I’m still not sure what would be the most ethical thing to do about them. At least, I shared my findings with the mods. But I’m almost ready to share usernames with my fellow authors so that they can preventively block them.

Genres and Average Ratings

It seems that there is a tendency for books that have only one or all ten of the Top 10 genres/tags assigned to them to have higher average ratings. The same works for the books that don’t have any of the Top 10 genres. However, this tendency isn’t supported by the general batch of data that includes not only the Top 1000 books.

Top 10 genres:

popularity_rank genre_name tag_count percentage
1 Fantasy 67155 69.42
2 Adventure 58991 60.98
3 Action 57635 59.58
4 Male Lead 43697 45.17
5 Magic 42671 44.11
6 High Fantasy 27201 28.12
7 Supernatural 25322 26.18
8 Strong Lead 21598 22.33
9 Female Lead 21324 22.04
10 Progression 20377 21.07

A similar pattern appears for the Top 20 genres as well:

However, there are no books that are tagged with 17, 18, 19, or 20 out of the Top 20 genres. Also, that may be the result of these books having fewer reviews.

At the same time, I don’t see any major dependencies on the average rating based on the chosen genre:

Fantasy (# 1): 60662 reviews | Avg: 4.570 ± 0.835
Adventure (# 2): 58138 reviews | Avg: 4.573 ± 0.831
Action (# 3): 47152 reviews | Avg: 4.552 ± 0.847
Male Lead (# 4): 37909 reviews | Avg: 4.578 ± 0.830
Magic (# 5): 48423 reviews | Avg: 4.560 ± 0.845
High Fantasy (# 6): 29567 reviews | Avg: 4.571 ± 0.827
Supernatural (# 7): 16258 reviews | Avg: 4.620 ± 0.782
Strong Lead (# 8): 27886 reviews | Avg: 4.570 ± 0.831
Female Lead (# 9): 21851 reviews | Avg: 4.619 ± 0.773
Progression (#10): 35607 reviews | Avg: 4.584 ± 0.815
Romance (#11): 7988 reviews | Avg: 4.610 ± 0.815
Multiple Lead Characters (#12): 9640 reviews | Avg: 4.704 ± 0.670
Sci-fi (#13): 12592 reviews | Avg: 4.633 ± 0.768
Anti-Hero Lead (#14): 13510 reviews | Avg: 4.614 ± 0.799
LitRPG (#15): 36078 reviews | Avg: 4.532 ± 0.870
Drama (#16): 12293 reviews | Avg: 4.610 ± 0.811
Comedy (#17): 18820 reviews | Avg: 4.599 ± 0.816
Portal Fantasy / Isekai (#18): 22655 reviews | Avg: 4.535 ± 0.872
Reincarnation (#19): 16296 reviews | Avg: 4.526 ± 0.874
Martial Arts (#20): 8773 reviews | Avg: 4.565 ± 0.842
Slice of Life (#21): 18319 reviews | Avg: 4.557 ± 0.845
Psychological (#22): 8273 reviews | Avg: 4.632 ± 0.771
Mystery (#23): 6305 reviews | Avg: 4.685 ± 0.691
Non-Human Lead (#24): 13816 reviews | Avg: 4.611 ± 0.773
Strategy (#25): 11411 reviews | Avg: 4.644 ± 0.731
Low Fantasy (#26): 5337 reviews | Avg: 4.701 ± 0.692
Tragedy (#27): 1645 reviews | Avg: 4.753 ± 0.611
War and Military (#28): 5347 reviews | Avg: 4.696 ± 0.680
Grimdark (#29): 5699 reviews | Avg: 4.658 ± 0.724
Secret Identity (#30): 7722 reviews | Avg: 4.614 ± 0.776
Urban Fantasy (#31): 7152 reviews | Avg: 4.644 ± 0.763
Short Story (#32): 489 reviews | Avg: 4.834 ± 0.430
Mythos (#33): 6175 reviews | Avg: 4.675 ± 0.708
School Life (#34): 7608 reviews | Avg: 4.585 ± 0.825
GameLit (#35): 18314 reviews | Avg: 4.561 ± 0.836
Horror (#36): 4583 reviews | Avg: 4.714 ± 0.670
Dungeon (#37): 9663 reviews | Avg: 4.589 ± 0.791
Attractive Lead (#38): 3854 reviews | Avg: 4.629 ± 0.746
Soft Sci-fi (#39): 6737 reviews | Avg: 4.613 ± 0.804
Post Apocalyptic (#40): 5086 reviews | Avg: 4.638 ± 0.759
Ruling Class (#41): 6842 reviews | Avg: 4.611 ± 0.777
Harem (#42): 2402 reviews | Avg: 4.460 ± 0.977
Dystopia (#43): 4710 reviews | Avg: 4.658 ± 0.747
Artificial Intelligence (#44): 3646 reviews | Avg: 4.692 ± 0.687
Contemporary (#45): 3559 reviews | Avg: 4.595 ± 0.808
Virtual Reality (#46): 3213 reviews | Avg: 4.659 ± 0.714
Villainous Lead (#47): 4342 reviews | Avg: 4.693 ± 0.673
Xianxia (#48): 9522 reviews | Avg: 4.515 ± 0.887
Historical (#49): 2230 reviews | Avg: 4.785 ± 0.561
Super Heroes (#50): 5208 reviews | Avg: 4.675 ± 0.737
Time Travel (#51): 2829 reviews | Avg: 4.679 ± 0.731
Genetically Engineered (#52): 3023 reviews | Avg: 4.672 ± 0.705
Technologically Engineered (#53): 2978 reviews | Avg: 4.670 ± 0.716
Cyberpunk (#54): 3413 reviews | Avg: 4.648 ± 0.743
First Contact (#55): 3317 reviews | Avg: 4.622 ± 0.812
Gender Bender (#56): 1505 reviews | Avg: 4.624 ± 0.758
Space Opera (#57): 2508 reviews | Avg: 4.694 ± 0.677
Hard Sci-fi (#58): 693 reviews | Avg: 4.723 ± 0.602
Wuxia (#59): 2102 reviews | Avg: 4.640 ± 0.727
Satire (#60): 3293 reviews | Avg: 4.652 ± 0.771
Reader Interactive (#61): 416 reviews | Avg: 4.570 ± 0.736
Time Loop (#62): 3869 reviews | Avg: 4.677 ± 0.721
Steampunk (#63): 1454 reviews | Avg: 4.707 ± 0.666
Sports (#64): 394 reviews | Avg: 4.832 ± 0.488

Average Rating Distribution

Among the Top 1000 books, there are 94 books with ratings over 4.8, 886 books with ratings over 4.5 but below 4.8, and only 20 books with ratings between 4.35 and 4.5. Interestingly, the latter group on average has more followers:

4.8-5.0: 2,756 avg followers | 619 median
4.5-4.8: 3,759 avg followers | 2,636 median
4.35-4.5: 7,612 avg followers | 5,946 median

In any case, if your rating is higher than 4.5, you are doing better than 2% of the Top 1000 books! Yay! 🙂

Average review length: 884 characters
Average rating variance within elite books: 0.496
Highest variance book: 2.125 (that seems to be a very controversial book!)
Lowest variance book: 0.000

Yes, there is at least one book that has all reviews with 5*-ratings. No, I’m not telling you what the book is just in case someone decides to go a rating-bomb it.

Review Count by Rating:
0.5 stars: 396 reviews ( 0.6%)
1.0 stars: 307 reviews ( 0.4%)
1.5 stars: 416 reviews ( 0.6%)
2.0 stars: 1,210 reviews ( 1.7%)
2.5 stars: 1,340 reviews ( 1.9%)
3.0 stars: 2,658 reviews ( 3.7%)
3.5 stars: 2,386 reviews ( 3.3%)
4.0 stars: 5,113 reviews ( 7.1%)
4.5 stars: 8,787 reviews ( 12.2%)
5.0 stars: 49,343 reviews ( 68.6%)

As you can see, the most popular rating for the Top-1000 books is 5*. And even the median rating is 5*. The average rating is 4.585.

Reviewers

There are a few hundred reviewers who have never left a 4*, 4.5*, or 5* rating for a Top-1000 book. There are more than a hundred of them who have left three or more ratings.

There are seventy-three people who have only rated the Top 1000 books with 2.5* or less. Together, they left 2% of all 3,669 low-rating reviews.

Most common words in sub-3.0 review titles (out of 3669):
‘good’: 355 times
‘story’: 287 times
‘great’: 175 times
‘start’: 125 times
‘this’: 121 times
‘with’: 102 times
‘just’: 87 times
‘read’: 87 times
‘boring’: 86 times
‘much’: 84 times

Most common words in sub-4.0 review titles (8713):
‘good’: 1170 times
‘story’: 839 times
‘great’: 478 times
‘with’: 319 times
‘start’: 299 times
‘read’: 265 times
‘this’: 250 times
‘interesting’: 230 times
‘much’: 179 times
‘like’: 177 times

Luckily for the authors, among the most active reviewers, there are only four people who would almost never leave a 5*-rating 🙂

Most common words in 5* review titles:
‘story’: 6721 times
‘good’: 4372 times
‘read’: 4308 times
‘great’: 4286 times
‘this’: 3060 times
‘best’: 2268 times
‘amazing’: 1770 times
‘very’: 1552 times
‘love’: 1510 times
‘with’: 1420 times

However, as I mentioned earlier, there are people who left such low ratings for the Top 1000 books they rated that their average is below 2.5* and even below 2*. There is one user with an average review rating of 1.227. This is just mean (and not in a mathematical way).

Review Swaps

1,944 reviews out of the total of 71,956 are marked as review swaps. I use the word ‘marked’ here, because Royal Road attributes a review as a swap if you post a review to someone’s book, even if you reviewed it without swapping with the author, while they also read and reviewed your book. I had this at least once on one of my books and I know of the other authors who had a similar situation (even while they categorically do not do any review swaps).

Books with review swaps: 232 (23.2%)

Low (0-10%): 131 books (13.1%)
Medium (10-30%): 45 books (4.5%)
Very High (50%+): 32 books (3.2%)
High (30-50%): 24 books (2.4%)

Category	Books	Avg Rating	Std Dev
No Swaps	695	4.616	0.215
Low (0-10%)	131	4.616	0.210
Medium (10-30%)	40	4.706	0.166
High (30-50%)	23	4.834	0.130
Very High (50%+)	32	4.893	0.060

Well, no surprises here, books with a higher amount of review swaps on average have higher ratings.

However, there are only 34 books among the Top 1000 that have more than 50% of their reviews done as swaps.

The fifty-five most active review-swapping books contribute approximately 73% of all review swaps.

Of all books with a high share of review swaps, only a couple were able to reach the Top 100. Most of them sit well below #500. Which is still a significant achievement. But it seems that the closer you get to the Top 10, the more chances there are that organic reviews will correct the real position of the book.

Category	Count	Avg Rating	Std Dev	Organic Avg
No Swaps	768	4.634	0.221	4.634
1-2 Swaps	119	4.634	0.218	4.626
3-10 Swaps	61	4.700	0.193	4.678
10+ Swaps	52	4.835	0.128	4.802

There are 119 books with only one or two review swaps. My guess is that they are false positives, as their average ratings fit within the pattern of the books without review swaps.

However, there is an interesting pattern among the books with high numbers of review swaps; the organic reviews seem to have higher ratings there as well. Is it peer pressure, hidden review swaps, or just the consequence of the fact that there are not enough organic reviews there, I can’t tell.

Trends by Year

There doesn’t seem to be any significant change in the average rating, neither by the year of book publication, nor by the year of review. If we exclude 2014-2016, when there were just not enough books for any reliable analysis, the averages fluctuate around the overall mean 4.585.

Review Year	Reviews	Books Reviewed	Avg Rating	Reviewers	Swap %
2014	246	5	4.791	208	0.0%
2015	758	14	4.728	624	0.0%
2016	726	25	4.710	621	0.1%
2017	884	46	4.599	717	0.2%
2018	1458	85	4.615	1143	0.5%
2019	2510	126	4.537	1683	0.4%
2020	5306	212	4.518	3169	3.1%
2021	8072	329	4.523	5048	1.1%
2022	9447	448	4.550	5911	2.9%
2023	13487	602	4.617	8517	2.6%
2024	16174	737	4.599	10404	3.4%
2025	12888	751	4.613	8923	3.9%

The number of book reviews and reviewers continues to grow every year. For 2025, we already have almost the same number of reviews and a higher number of reviewers than Royal Road had in 2023. There should probably be a correction for the deleted accounts and reviews, but there’s no way for me to gather that data.

Interestingly, of the weeks of the year, the lowest is the 8th week (February) with 996 reviews, which sees 31.3% fewer reviews compared to the highest week 45 (November) with 1,308 reviews. The yearly average is 1,133. I excluded 2025 from this analysis. Generally, the early weeks of the year tend to have fewer reviews compared to the later weeks. It would be interesting to learn if that corresponds to the reading patterns on Royal Road. However, my guess is that the winter reading slump, resolution fatigue, and other aspects characteristic of late winter will kick in. Summer also shows some high results as well.

Top Books by Review Count

Rank	Reviews	Avg Review	Title	Author
1	1157	4.826	Mother of Learning	nobody103
2	1094	4.647	Beware Of Chicken	Casualfarmer
3	1006	4.531	The Primal Hunter	Zogarth
4	973	4.640	Super Supportive	Sleyca
5	882	4.308	He Who Fights With Monsters	Shirtaloon
6	680	4.438	Azarinth Healer	Rhaegar
7	596	4.367	Beneath the Dragoneye Moons	Selkie
8	589	4.775	The Perfect Run	Maxime J. Durand
9	523	3.969	Delve	SenescentSoul
10	488	4.761	Runeblade: A Delving & Skill Merging LitRPG	Bacon Macleod

Total reviews for the Top 10 books: 7,988
Percentage of all reviews: 11.1%
Average review rating of the Top 10 books: 4.526

What’s Next?

I’m preparing an analysis of the whole dataset of all reviews. If you think I should add something to it, look at some specific data, or if you have any other recommendations, I’d be happy to see them in the comments.

Comments From the Community I’d Like to Mention

Milc: It’s interesting to note that review swap farming doesn’t affect much of the top rated stories but I would still recommend new writers should do a few just to shield against drive by low ratings.

Comments

4 responses to “Do Review Swaps Influence the Top 1000 Books? And Other Adventures of Ratings on Royal Road”

deLongchamp

August 3, 2025

“There is one user with an average review rating of 1.227. This is just mean (and not in a mathematical way).” This was funny. Very helpful data to have, interesting trends!

Reply
1. stepanchizhovc
  
  August 3, 2025
  
  Perks of being an author, I guess. You start to look for puns everywhere 🙂
  
  Reply
Dave Pallant

August 4, 2025

Interesting that the average rating you give for the top ten books does not necessarily match the Overall Score in the public statistics. I wondered how a book with 3.969 could be high up in the tables but in the public statistics for the book the Overall Score is 4.49 which feels more like a top book. Do you know where the difference between the two lies?

Reply
1. stepanchizhovc
  
  August 4, 2025
  
  You are right to point that out. It’s the average review rating, not the average rating in general for those books. I corrected the label.
  
  Reply