How We Review RNG Casino Games

A casino game review methodology is the documented set of data sources, metrics, and weights we use to consistently score RNG table games, so every ranking on our site maps back to evidence rather than opinion.

We review dozens of RNG casino games each year at OnlineCasinoRank, and we noticed a pattern across the wider industry: most review sites publish rankings and star ratings without ever showing the math behind them. A game earns "9.2 out of 10," and readers are expected to trust the number. We think that asks too much.

This article walks you through our review model. We cover the four data streams that feed into our popularity scores, the seven weighted metrics that drive our composite ratings, and two worked examples that show exactly how the numbers come together for real games on the market today. We also cover what the model does not do well, because a methodology that hides its limitations is not really transparent at all.

How We Measure Game Popularity

Game popularity is a score we build from four things: what people search for online, how they read our reviews, how players behave across the wider market, and which games show up in casino lobbies.

Popularity is the biggest part of our final score, so we are careful. We do not want a single metric to decide which games look popular. If we only looked at search numbers, a game that went viral for a week would look huge. If we only looked at casino lobbies, we would just be picking the games that operators paid to show off. So we use four sources at once and compare them before assigning a game its popularity score.

Our Four Core Data Sources

We check four sources every month. Each one tells us something different about how a game is really doing.

• Search data. We watch Google results all the time to see which games are rising, which new games are appearing, and which older games are falling behind. This helps us choose what to review next. When a new Pragmatic Play or Hacksaw Gaming slot starts showing up in search results soon after it launches, we know we need to review it quickly, before players look for a review that isn't there yet.
• Organic search performance. We look at how readers use the game reviews we already have on our site. Which ones do they spend time on? Which ones do they scroll all the way through? Which ones do they come back to? When a review gets lots of attention, we make it better. We add more detail on the bonus rounds, include fresh screenshots, and answer the questions readers seem to be asking. We want the games people care about most to get the best coverage.
• Player engagement on the wider market. We use outside sources to see how players behave, not just our own site. Other companies collect data from many casinos and compile it. That data gives us a clearer picture of how long people play, how many spins they do, and whether they come back. Then we share what we learn with our readers so they can see how a game really plays out in the real world.
• Casino lobby coverage. We check how many of the 120 or so casinos we follow have the game in their lobby, and where they put it. A game in the "Popular" tab or at the top of the page is a bigger deal than a game buried deep in the list. That is because casinos put their best games up front, based on their own data.

Popularity Across Software and Game-Type Taxonomies

One big list of the most popular games is not very helpful. A huge Pragmatic Play slot and a small Nolimit City horror slot are not competing for the same players. If we put them on one list, the smaller game would look like it was losing, even if it was a star in its own group.

So we make three types of lists instead of one. The first type sorts games by provider, meaning the company that made the game, like Pragmatic Play, NetEnt, Play'n GO, Hacksaw Gaming, Nolimit City, and Relax Gaming. That way, if you want the best NetEnt slot, you get a NetEnt list, not a mix. The second list sorts games by type, such as roulette, blackjack, and keno. The third list type is the full market view, for readers who just want to see the biggest games for a specific period. You can explore some examples of such lists that we conducted in the 2025 Q1 Report and 2025 Six-Month Report of the most popular game titles.

On each list, we score games against the other games in their own group, not against every game on the site. That way, a top-10 roulette game and a top-10 blackjack game are both winners in their own group, and neither one gets punished for not playing the same kind of game as the other.

Pragmatic Play

NetEnt

Play'n GO

Roulette

Blackjack

Baccarat

Keno

Our Scoring Model: Metrics and Weights

Our casino game ranking model assigns each game a score from 0 to 10 across seven categories. We then multiply each score by how much that thing matters and add them up to get a final score out of 100.

Think of it like a school report card with seven subjects, but some subjects count more than others. A game can do great in one area and poorly in another, and the final score shows how it balances out overall. The scoring is not fully automatic. Some of the scores come from our team's judgment, formed after reviewing the data. That is why we call it "semi-quantitative." The numbers matter, but so does human review.

The Seven Metrics We Score

Here are the seven things we look at, how much each one counts, and how we turn them into a score from 0 to 10.

Not every metric counts the same. Some things matter more than others, and the weights show why. Country availability and market popularity are the biggest pieces, at 20% each. The reason is simple. A great game nobody in your country can play is not much use to you. And a game nobody wants to play, no matter how fancy it is, is not a game we can honestly put at the top of a list. These two metrics ensure our rankings reflect what real players can actually do.

RTP and feature complexity come next, at 15% each. RTP is the math behind the game, and features are the fun of playing it. Both matter, so we give them the same weight. Volatility, bonus frequency, and provider reputation each count for 10%. They are important, but they are not as important as the other four. Our casino game review analytics run every game through this same setup, which is how we make sure two reviewers scoring the same game land on a similar number.

Worked Examples: Scoring Two Popular RNG Table Games

To show how the model works in practice, we ran two well-known RNG table games through the same seven-metric setup. The raw scores come from our team's review of live data. The weights stay the same for every game, every time.

Example 1: Premium Blackjack by Playtech

Premium Blackjack is a single-hand RNG blackjack game from Playtech. It uses six decks, pays 3:2 on a natural blackjack, and lets the player double down on any two cards and split up to three times. It is one of the most played RNG blackjack games in regulated European markets.

Premium Blackjack earns 8.3 out of 10. The strong RTP, clean design, and wide availability carry most of the score, while the lack of side bets keeps the feature total from climbing higher.

Example 2: Jacks or Better by Play'n GO

Jacks or Better is a classic RNG video poker game from Play'n GO. Players are dealt five cards, choose which ones to keep, and draw new ones to make the best poker hand. A pair of jacks or higher pays, and the top prize comes from a royal flush. It is one of the most played video poker titles in regulated markets.

Jacks or Better earns 7.8 out of 10. The excellent RTP and solid hit rate do much of the work, but the simple, classic design and the smaller audience for video poker keep the final score a step below that of top-rated table games.

Limitations and How We Update Scores

Our casino game ranking model is careful but not perfect. We want you to know what it does well and where it falls short, because a review method that hides its weak spots is not really open at all.

The model is semi-quantitative, meaning real numbers drive much of the scoring, but some parts still come down to our team's judgment. Volatility, feature complexity, and provider reputation all take some reading between the lines. Two reviewers looking at the same game will sometimes land a point apart on one of these, and that is normal. We talk it through as a team and settle on the score that best fits the game, but we will not pretend the number came out of a machine when it did not.

Market popularity is checked every month, because search trends and lobby placements shift fast. RTP and country availability are checked every three months, since providers quietly release new RTP versions and markets open or close to certain games more often than players realize. Provider reputation is updated whenever something significant happens, such as a license action, a major player dispute, or a new round of test results.

There are also times we update a score outside the normal schedule. If a provider releases a new RTP version of a game we have already reviewed, we reopen the score within 14 days. If a regulator pulls a game from a market, or a new market gets access, we adjust the country availability score right away.

One last thing worth saying clearly: we do not let business deals push scores around. The model runs the exact same way for games from casinos we work with and games from casinos we do not. If a game scores 6.5 out of 10, it scores 6.5 whether or not the operator behind it is a partner. That is the whole point of writing the method down in the open.