The Sweepstakes Casino Complaint Index: 232 sweepstakes casinos tracked, 118 currently meet our publication threshold

The numbers at a glance

Corpus: 118 of the 232 operators we track with a Trustpilot snapshot clear the 10-review floor, drawn from 232 operators in our review catalog (every operator with a published review page, including those that have left the market) · snapshots July 3, 2026 to August 16, 2026 · static page rebuilt August 22, 2026

How to read these numbers

• "1 and 2 star share" is the fraction of an operator's full public Trustpilot review history sitting at 1 or 2 stars. It is cumulative sentiment, not a monthly window.
• "Review base" is the review count the share is computed over. We publish no share under 10 reviews, because two angry reviews out of three is noise, not a measurement.
• For a minority of operators the base is an archived sample of roughly 60 reviews rather than the all-time total, which is why we say review base and never "reviews this period".
• "Penalty" means points already subtracted from the operator's published trust score. It is never a further deduction waiting to happen.
• A high share is what reviewers say. What happened when we put our own money in is a separate evidence stream: the payout study.

Every operator with a publishable review-sentiment metric

Sorted by review base, largest first, so the most-reviewed brands lead and no small-base operator can headline. Each row links to the operator's full trust-score page, where the score's evidence lives.

Table basis 118 operators clearing the 10-review floor, out of 232 with a Trustpilot snapshot, rebuilt August 22, 2026. Row drill-down: each operator name links to its own trust-score page, and the same rows are published as a dated CSV in the open-data repository linked under "Cite this index".

Where reviewers are most negative

The gate first: only operators with at least 100 public reviews appear in this table, so no small-base operator can be named here on the strength of a handful of angry posts. Within that cohort, these are the operators whose public review distributions carry the heaviest 1 and 2 star share. A place on this list is a measured statement about reviewer sentiment over a large base: it says a lot of the people who reviewed the operator publicly came away unhappy, and it says nothing by itself about why. The themes column in the main table above and the operator's own trust-score page carry the detail.

Review counts and the 1 and 2 star share come from each operator's public Trustpilot review distribution, refreshed on a rolling schedule.

Where reviewers are least negative

Same gate, opposite end: the lowest 1 and 2 star shares among operators with at least 100 public reviews. This is explicitly not an endorsement and not a payout guarantee; it means the operator's public review base skews positive, which is one input among many. Whether an operator actually paid us is a different question with different evidence, answered in the payout study. We publish this side of the distribution because an index that only ever ranks villains reads like a hit list, and this is a measurement.

Operators with no publishable review-sentiment metric

These operators are scored. They have a published Trust Score built on our other evidence streams; what they do not have is a review-sentiment share we are willing to publish. Their latest snapshot gives us nothing usable here: either the review base sits under 10 reviews, too small a base to publish a share, or the latest fetch recorded no ratio at all: 1uckyduck, Ace, Acebet, Amazon Jackpots, Bare Knuckle Sweeps, Betr Social Casino, BigCashSweeps, Blitzmania, BoomWins, BucksJet, CapySpin, CardCrush, Cashoomo, Cheddr, Coin Wizard Games, CoinsBack, CoinsBucks, CoinsMania, CosmoSlots, Courtside, CrashMania, CrazyCoins, Dara, Deal or No Deal Win, Diam.bet, DoggHouse, DollarMills, Dorados, Dragon Phoenix, EpicSweep, FortunaRush, Fortune Purple, GameDayZone, Gleaming Slots, GoldenHearts, GoldieCity, GoldSlips, GoodVibes, Grand Fortune Club, Grand Vault Casino, HappyEnding, Havana Fortuna, Jackpot Daily, Jackpot Golden, JackpotGo, JackRoyals, JefeBet, KingCoins, Lady Lucka, LavishLuck, LegacyCasino, Legendz, LionsDenGames, LuckLake, LuckShack, Lucky Buddha, LuckyBitsVegas, LuckyBunny, LuckyCanoe, LuckyDinero, LuckyOne, LuckyRush, MetaWin, Midas Jackpots, MyBets.us, MySweepz, Neon Rush, NewLuck, NightWager, NovaSpin, Ocean King Casino, RadPoker, RegalCoins, RollRoyale, RoxyMoxy, Runestake, Scrooge, Sidepot, Sixty6, Slotomo, SlotPlay, SlotSpot, Smiles, Soprize, Spin4America, Spindoo, Spinsly, Spinszy, Spintime, Sweepolis, SweepsBlast, SweepsLasVegas, SweepsNight, SweepsRoyal, SweepstakesCasino.com, TheMoneyFactory, ThrillCoins, TurboStakes, United Slots, UptownSweeps, VegasBonanza, VegasX, Victory Rush, Victory Sweep, VikingRiches, WildWorld, Winera, WinWinSweeps, WiseSpin, Yayz Casino, Zonko, Zoot, Zumo, Zunado. Where a small-base ratio does exist in our database we withhold it deliberately; publishing a percentage computed over a handful of reviews would dress noise up as data. Where a fetch failed, no ratio exists and we impute nothing. Either way the absence is ours, not a finding about the operator, and its full trust-score page is unaffected.

Themes reviewers raise

Beyond the star math, our pipeline labels each operator's fetched review text against a fixed vocabulary of recurring themes. Labels are only ever applied to review content we actually fetched; no theme is ever invented. Across the mined corpus, 65 operators' latest snapshots carry at least one theme. Here is what each negative theme means for you as a player, and what to check before depositing:

Redemption delays 55 operators

Reviewers describe cashouts that sat in processing far longer than the operator promised. Before you deposit, check the measured redemption speed on the operator's trust-score page and start with a small first cashout.

KYC problems 49 operators

Reviewers describe identity checks that rejected routine documents or looped without resolution. Verify your identity before you have a balance waiting, not after a win, so a document dispute never holds your money hostage.

Unresponsive support 53 operators

Reviewers describe tickets that went unanswered or bots that could not escalate to a person. Send support a simple question before depositing; how they handle an easy ask predicts how they will handle a stuck cashout.

Account closures 37 operators

Reviewers describe accounts closed or suspended with balances still inside. Read the operator's terms on dormancy and closure, and never let a large balance sit unredeemed.

Game disputes 50 operators

Reviewers describe wins that were voided or rounds that misfired without a clear resolution. Screenshot anything unusual as it happens; a dispute with evidence goes very differently from a dispute without it.

Terms-change complaints 18 operators

Reviewers describe terms that changed after they had already bought coins, usually around redemption rules. Re-read the redemption terms each time you cash out rather than trusting what they said last month.

The vocabulary is not all negative, and we show the other side openly rather than filing it off: "Payouts confirmed" appears on 4 operators' latest snapshots, and "Good experience overall" appears on 7 operators' latest snapshots. That is why this section is titled what reviewers raise, not "complaints": the same labeling pass that finds redemption delays also finds confirmed payouts, and hiding the second half would turn a measurement into an argument.

Complaint penalties we currently apply

A complaint penalty is a capped deduction with two components. Documented legal-event penalties come from events recorded on our state legality tracker whose allegations or consequences include tangible player harm, and each one eases automatically on a fixed schedule as the event ages. A filing that only disputes the sweepstakes model's legality in some state stays on the tracker and the state adjustments; it does not reduce the national score. The editor-applied component is set only after a human confirms gated evidence and never eases on a timer; it stays until a human clears it. Both components are already baked into every published score you see on this site: when we show an operator's trust score, the penalty has already been subtracted. Mined review signals alone never move a score. Penalties under editorial review are staged privately and are not shown anywhere until applied. The formula and caps live in the methodology's adjustments section.

As of this build, 1 operator carries an active complaint penalty, withholding a combined 8 trust-score points. The deepest single deduction sits on Moonspin, at 8 points.

• Moonspin: carrying an active complaint penalty (score reduced by 8 points), already reflected in the score above.

Recent movement: 21 penalty changes across 9 operators in the last 90 days, most recently August 13, 2026. Every change is logged to our public score history.

What this index cannot see

We would rather you hear the limits from us than discover them yourself. First, this is Trustpilot only. We also mine Reddit chatter for internal leads, but Reddit sentiment is a language-model estimate over small snippet samples, so we never publish it as a measured ratio; it points our human review somewhere and stops there. Second, shares are withheld below the review floor because tiny bases produce noise, which means genuinely new operators are invisible here for a while. Third, selection bias is real: people with problems review more readily than people who got paid and moved on, so a high share is a signal to investigate, not a verdict. Fourth, review bases mix all-time totals with archival samples, which is why we say "review base" and never "reviews this period". Finally, the limit that matters most: a share is reader sentiment. It is not evidence that an operator refuses redemptions; our first-hand evidence on that lives in the payout study.

Methodology

• Source. Each operator's public Trustpilot star distribution, fetched from Trustpilot's own widget data. Where the widget serves the all-time distribution, the 1 and 2 star share is arithmetic over that full distribution; there is no model anywhere in the ratio.
• Fallback. Where no live widget exists, we use an archived snapshot of roughly one review page. Snapshots older than 45 days do not count as fresh evidence.
• Floors. No share is published under 10 reviews, and the named most and least negative cohorts require at least 100 reviews.
• Latest snapshot per operator. Only each operator's newest Trustpilot snapshot feeds the index; we do not blend runs with shifting review-count semantics.
• Failed fetches record zero. A fetch that fails records zero volume and no ratio. Nothing is imputed, ever.
• Themes. Labeled from fetched review text against a fixed vocabulary that includes positive labels. Sample links must appear in fetched content; no theme is invented.
• Reddit is mined but excluded. Reddit rows feed internal leads only and never a published ratio.
• Penalties are guarded. Complaint penalties are applied through a guarded scoring function with every change logged to score history, including the scheduled easing of legal-event penalties. The mining job cannot write a penalty; one starts only from a documented player-impact legal event or a human editorial decision.
• Bound at build. Every figure on this page binds to our database at build time and re-derives on every deploy, last read from our database on August 22, 2026.
• Affiliate-funded, not affiliate-weighted. BonusBandit earns from disclosed affiliate links. Scores and this index are computed mechanically; an operator cannot buy its way off this page or into a better number.

Static build stamp: 2026-08-22T00:07:53.406Z. See which date is which below.

Which date is which

A single "updated" stamp would be the easiest thing to publish and the easiest thing to misread. Several different events can each make a page like this look fresh, they mean completely different things, and this page holds only some of them. Here is the whole list, including the ones we cannot answer from this dataset.

Static page rebuilt

August 22, 2026 When this page was last generated. A rebuild re-reads the database; it is not evidence that anything underneath changed.

Source recrawled

July 3, 2026 to August 16, 2026 When we last fetched the public review distributions in the table. Each row carries its own date in the source recrawled column.

Database row refreshed

Not recorded on this page The index reads each operator's newest stored snapshot. We do not keep a separate row-update stamp, so we do not publish one.

Human evidence reviewed

Not recorded on this page Editorial review of an operator's evidence is dated on that operator's trust-score page, not in this dataset.

Operator retested first-hand

Not recorded on this page Deposits and redemption requests on our own accounts are a separate evidence stream, dated in the payout study.

Verdict revised

Not recorded on this page Written verdicts live on the operator pages and carry their own revision dates. Nothing on this page revises a verdict.

Complaint penalty last changed

August 13, 2026 Most recent complaint-penalty change logged to our public score history in the last 90 days.

An unrecorded date stays unrecorded. We never borrow the build stamp to fill one in.

Questions readers ask

Does a high complaint share mean an operator won't pay?

No. The share measures reader sentiment: the fraction of an operator's public Trustpilot reviews at 1 or 2 stars. It is a signal to investigate, not proof of anything about payouts. Our first-hand payout evidence, real deposits and real redemption requests on our own accounts, lives in the payout study.

Why is my operator not listed in the table?

Missing from this table means no publishable review-sentiment metric, not unscored. Either its review base is under our publication floor, so any share we computed would be noise, or our mining pipeline has not reached it yet. Coverage expands on a rolling schedule, operators below the floor are named on this page without a ratio, and every one of them still carries a full Trust Score built on our other evidence.

Where do the numbers come from?

Each operator's share is arithmetic over its public Trustpilot star distribution, and the penalty figures come from our own scoring database. Everything is recomputed from the database when the page is built, so the figures cannot drift from the underlying records.

Can an operator pay to be removed?

No. BonusBandit is funded by affiliate links and we disclose it, but scores and this index are computed mechanically and are not affiliate-weighted. An operator cannot buy its way off this page or into a better number.