Possible coordinated and policy-violating reviews on Desert Warrior

Hi Maya,

Thank you for the response, really appreciate you taking a look.

I just wanted to add a bit more specific context that may help with the review:

• A number of reviews appear to be duplicated almost word-for-word (including repeated paragraphs and identical phrasing across different accounts)

• Several accounts posted reviews immediately after creation within the same short time window

• Some reviews explicitly indicate the users have not seen the film, yet are rating it 1/10

• There are also instances of the same review text being posted multiple times by the same user

Additionally, some reviews include language targeting casting choices based on ethnicity (e.g. rejecting actors based on origin or skin tone), which may fall outside standard review guidelines.

Given these more specific patterns, it may be worth a closer manual review of:

• duplicate reviews

• newly created account activity clusters

• and reviews that do not reflect actual viewing

Happy to share specific examples if helpful.

Thanks again

Hi Maya,

Thank you for the action taken on this title in March. The review cleanup was appreciated, and I understand the effort that went into it.

However, I'm following up because the resolution appears incomplete, and the title's rating integrity has not been restored.

What happened:

After my report, the title was locked for approximately 81 days (mid-March through April 24). During that time, 179 user reviews were removed. When the title was reopened on release day, the written reviews were gone, but all the star ratings from the same period were restored unchanged, including those from accounts whose reviews were deleted for policy violations.

The title reopened on April 24 with a 1.3/10 rating based on approximately 4,700 pre-release votes. It now sits at 1.9/10 with over 5,200 votes. The underlying data tells a very clear story.

The rating histogram is not consistent with organic behavior:

- 73.1% of all votes are 1/10- 13.8% are 2/10- Combined 3 through 9 stars account for only 4.1% of votes- Then 10/10 jumps to 4.9%

No film, regardless of quality, produces a distribution where 87% of votes are 1 or 2 stars and only 4% fall between 3 and 9. This is a textbook manipulation pattern. The near-total absence of middle ratings is the clearest indicator.

The geographic concentration is extreme:

74% of all votes originate from a single country, averaging exactly 1.2/10, while the film had limited or no theatrical release in that market during the voting period. This is the same pre-release voting pattern I flagged originally.

External ratings confirm the anomaly:

I'm not claiming the film is great. Every other platform reflects a mixed-to-below-average film:

- Rotten Tomatoes: 33% critics / 63% audience- Metacritic: 41/100- Letterboxd: 2.6/5

IMDb's own Metacritic page (https://www.imdb.com/title/tt13570066/criticreviews/) shows critic scores ranging from 20 to 70. Not a single professional critic rated it anywhere near 1/10. There are also 12 external reviews listed on IMDb's own external reviews page (https://www.imdb.com/title/tt13570066/externalreviews/). The consensus across all platforms and professional reviewers places this film in the 3-5/10 range.

A 1.9/10 on IMDb against a 63% audience score on Rotten Tomatoes and 2.6/5 on Letterboxd is a gap too large to explain by platform differences. It can only be explained by the pre-release vote manipulation that was already acknowledged.

Post-release data from actual viewers supports this:

Since the title was reopened, new votes from markets where the film is actually playing in theaters are averaging around 6/10. These are people who bought tickets and watched the movie. Their votes are being mathematically overwhelmed by the 4,700 pre-release manipulation votes that were restored.

The inconsistency:

If the written reviews from these accounts were deemed manipulative enough to warrant deletion, the star ratings submitted alongside those same reviews should not have been restored. The reviews and ratings came from the same accounts, during the same coordinated campaign, as part of the same behavior. Removing one but keeping the other doesn't address the actual problem. The rating is the part that affects the title page, search rankings, and audience perception.

What I'm asking:

Could the team please review the star rating data with the same scrutiny that was applied to the written reviews? Specifically:

- Ratings from accounts whose reviews were previously removed- The statistical anomaly in the histogram distribution- The geographic concentration from a single market during the pre-release window

The current weighted average does not appear to be correcting for this level of coordinated activity. The film has real problems, and a fair rating would likely land in the 4-5/10 range. But 1.9/10 is not a reflection of audience opinion. It's a reflection of the campaign that was already identified and partially addressed.

Thank you for your time.

Hi Maya,Did you have a chance to look into this? Your help will be much appreciated.Thanks

Hi @ahmetyavuz​,

Our team have looked into this and I can confirm that our weighted average rating is displaying as intended on this title. As previously mentioned, we have safeguards in place to detect and address attempts to artificially influence ratings, and the weighted average already accounts for this.

For more details on how our ratings work, feel free to check out our Ratings FAQ.

Thank you for your understanding, and for taking the time to raise this.

Hi Ozzy,

Thank you for the reply, but I have to respectfully push back here. Saying the weighted average is "displaying as intended" doesn't address anything I raised. I provided specific, verifiable data points. I'd appreciate a response to those points rather than a general reassurance.

Let me put the question more directly:

Is this histogram consistent with organic audience behavior?

Out of 5,741 votes:•⁠  ⁠71.8% are 1/10•⁠  ⁠12.9% are 2/10•⁠  ⁠10.2% are spread across 3 through 9•⁠  ⁠5.1% are 10/10

That means 84.7% of all votes are either 1 or 2, while only 10.2% fall across the entire middle range. There is no film in IMDb's database that produces this distribution organically. Not one. I'd genuinely welcome being shown a counterexample.

For comparison, this film has a 63% audience score on Rotten Tomatoes (verified ticket holders) and 2.6/5 on Letterboxd. These platforms suggest a below-average but watchable film, somewhere in the 4-6 range. IMDb shows 2.1/10. That's not a platform difference. That's a data integrity issue.

The weighted average is not correcting for this. If the safeguards were working, the weighted average would diverge meaningfully from the arithmetic mean. Right now, the unweighted mean and weighted average are nearly identical. That tells me the weighting system is treating these votes as legitimate.

A specific question I'd like answered:

In March, your team removed 179 written reviews from this title after I reported coordinated, policy-violating activity. Many of those reviews came from newly created accounts, contained duplicate text, and admitted to not having seen the film. The accounts that posted those reviews also submitted star ratings. When the title was reopened in April, the reviews stayed deleted, but the star ratings from those same accounts were restored.

Why were the ratings treated differently from the reviews? If the accounts were problematic enough to warrant review deletion, why are their star ratings considered legitimate?

This isn't a rhetorical question. I'm trying to understand the logic so I can accept the outcome if there's a sound reason behind it.

One more data point:

Since the film opened in US theaters on April 24, new ratings from markets with active theatrical releases are averaging around 5-6/10. These are people who actually watched the film. Nearly 70% of total votes on this title come from a single country where the film had no wide theatrical release during the primary voting window. Those votes average 1.2/10.

The weighted average is supposed to account for exactly this kind of pattern. If it's "displaying as intended" at 2.1/10 while actual ticket-buying audiences rate it 5-6/10, then the system is not working as intended, regardless of what the internal check says.

I appreciate your time and I understand these things involve nuance. But the data here is not ambiguous, and a copy-paste response about safeguards doesn't match the specificity of what I've reported. I'm asking for a specific answer to a specific inconsistency.

Thank you.

Even the most active IMDb reviewers are complaining about this:

https://www.imdb.com/title/tt13570066/reviews/?item=rw11430512&ref_=ext_shr_lnkTake a look at this please.Thanks