On this page
Online pokies combine random-number generation, a mathematical game model and a paytable. The selected stake and active game state determine which rules apply; the displayed reels then present the result. Understanding that process explains why matching symbols may not pay, why RTP is not a session refund and why a feature can change the rules without making a losing game “due”.
18+ information and comparisons. Casino links are affiliate links, and we may earn commission. Read our disclosure.
Casino offers and game information
| Casino | Welcome offer and qualifying conditions | Casino link |
|---|---|---|
| Fair Go | Up to $1,000 pokies welcome bonus Code: WELCOME General deposit-bonus terms: $20 minimum; 30× (deposit + bonus); $10 maximum bonus bet, unless the offer states otherwise. Full offer terms — Fair Go | PLAY NOW |
| Ozwin | 400% up to $4,000 Code: WELCOME400 General deposit-bonus terms: $20 minimum; 35× (deposit + bonus); $10 maximum bonus bet, unless the offer states otherwise. Full offer terms — Ozwin | PLAY NOW |
| Ripper Casino | 100% up to $1,000 + 40 spins Code: RC400 Minimum deposit $20; 30× wagering on deposit plus bonus unless specified otherwise; maximum cashout 20× bonus. Bonus-round limit: $5 non-VIP / $10 VIP, including purchases. 40 Fortunate Buddha spins; offer expires 48 hours after redemption. Full offer terms — Ripper Casino | PLAY NOW |
| Uptown Pokies | Deposit $25, play with $125 + 50 spins Code: BANDIT125 The $125 includes your $25 deposit. Offer panel states 40× rollover and no maximum offer cashout. Confirm the wagering base and withdrawal limits in the full terms. Full offer terms — Uptown Pokies | PLAY NOW |
Reviewed 27 October 2026. These are general promotions. Check the actual game, installed return setting, account currency and accepted offer terms. A software explanation does not verify a particular operator’s complete catalogue or turn a bonus into guaranteed cash.
The parts of a paid round
- Stake and mode: the action uses a total price and the selected settings.
- Random inputs: the game obtains the values used to determine its outcome.
- Game rules: those inputs are interpreted under the active mathematical model and feature state.
- Evaluation: the paytable and any qualifying modifiers determine an award.
- Presentation and settlement: the game displays the result and records the completed round.
This describes the main roles, not a claim that every supplier uses an identical technical message sequence. Cascades, choices and feature rounds can add stages. The game-type guides compare those structures without assuming every supplier implements the same round sequence.
The animation helps communicate what happened. It does not replace the rules, and watching a symbol stop near a win does not establish that the next result is more likely to pay.
What the random-number generator contributes
An RNG supplies random inputs. The game’s rules determine how those inputs correspond to symbols, positions, features and awards.
Randomness does not mean every visible symbol or prize has the same probability. A rare prize can be represented by fewer qualifying outcomes than an ordinary low-value combination.
For a simple illustration, imagine a draw from one hundred equally likely numbered tickets. If eighty tickets represent no prize and one represents a large prize, the draw is random while those prize categories remain very differently weighted.
That ticket example is an invented teaching model, not a description of a real supplier’s code. It shows why “random” and “all prizes equally likely” are different statements.
The relevant evidence for a game is its disclosed rules, configured mathematics and testing information, not the speed of its animation or how lucky it looked in a short video.
A hypothetical paytable
Consider this deliberately simplified game costing $1 for each independent draw. Tickets are drawn with replacement, so every draw uses the same probabilities.
- 80 tickets out of 100 pay $0. Their weighted contribution is 80% × $0 = $0 per draw.
- 15 tickets out of 100 pay $1. Their weighted contribution is 15% × $1 = $0.15 per draw.
- 4 tickets out of 100 pay $5. Their weighted contribution is 4% × $5 = $0.20 per draw.
- 1 ticket out of 100 pays $60. Its weighted contribution is 1% × $60 = $0.60 per draw.
The expected return is $0.95 per $1 draw: fifteen cents plus twenty cents plus sixty cents. That gives the hypothetical model 95% RTP.
It does not guarantee $95 returned after exactly one hundred draws. Drawing with replacement allows repeats and does not force every ticket to appear once.
A player could miss the large prize entirely or draw it more than once. The expected value is an average of possibilities, not a fixed schedule delivered to each participant.
Why a hit is not necessarily a profit
In that teaching model, twenty tickets return something, giving a 20% frequency of non-zero returns. Only five tickets return more than the $1 cost.
The fifteen $1 outcomes simply return the stake. They do not produce a $1 net profit on top of it.
For another hypothetical paid round costing fifty cents and returning ten cents, the net result is a forty-cent loss even if the game highlights a winning symbol combination.
When reading statistics, establish what counts as a hit. A non-zero return, a profitable round, a feature trigger and a maximum win are different events.
The same RTP can hide a different prize pattern
Imagine a second invented $1 game returning $1 on ninety-five of its one hundred equally likely tickets and zero on the other five.
Its expected return is also $0.95, so its RTP is also 95%. Yet its pattern is very different from the first model’s rare $60 award and frequent zero returns.
This shows why RTP alone does not describe the whole distribution. Maximum potential and frequency also need context.
The second model never returns more than its $1 cost, despite returning something on 95% of draws. A frequent return and a profitable result therefore remain different events. Neither teaching model is a recommended game or an available casino product. The examples isolate the mathematics so that similar-looking statistics can be compared accurately.
Our volatility guide explains how prize distribution differs from average return.
Why matching symbols may not pay
A payline game evaluates specified paths. Matching symbols need to occupy eligible positions and satisfy the required count and direction.
A ways game uses its own adjacent-reel rules. A cluster game looks for qualifying groups, while a paying-anywhere game can count symbols under a different rule.
For a concrete contrast, the supplier describes original Starburst as using ten fixed lines paying in both directions. Original Reactoonz instead describes clusters of five or more matching aliens on a seven-by-seven grid.
The NetEnt comparison and Play’n GO comparison explain those named examples separately. Seeing five matching symbols is therefore insufficient information on its own. Read where they must be and how the title evaluates them.
Likewise, an inactive line, a missing starting reel or disconnected symbols can explain why an attractive-looking arrangement does not qualify. The actual paytable settles the question.
Wilds, scatters and special symbols
A wild usually substitutes under defined conditions. A scatter can have its own award or trigger role, but those details need reading for the particular title.
Do not assume a wild replaces every special symbol. It might help an ordinary combination while being unable to substitute for a scatter or collector.
Expanded, sticky, walking and multiplier wilds add different effects. One changes coverage, another persists, another moves and another modifies an eligible award.
A symbol with a printed cash value can also require a collector before that value is awarded. Its appearance alone should not be counted as money already credited. The hold-and-win guide explains locked symbols and collector-style feature conditions in more detail.
Game state can legitimately change the rules
A free-spin feature may retain wilds, expand the grid or apply a different multiplier. Those are disclosed states within the game rather than evidence that its ordinary model owes the player a win.
Read what persists and for how long. “Sticky” can apply to a defined sequence without surviving a new round, device change or session.
Some choices also change the feature entered. The number of attempts, starting modifiers and price can differ between options.
This is why it is too broad to say every screen always has identical conditions. The important distinction is between defined feature states and an unsupported claim that losses force repayment.
Cascades belong to the complete round
When winning symbols disappear and replacements fall, a sequence can contain several evaluations after one starting action. Read the title’s rules on cost and continuation. Our video pokies guide compares these evaluations with ordinary paylines and other symbol systems.
For a hypothetical twenty-cent round with five-cent and ten-cent components, the total return is fifteen cents. The net result is a five-cent loss.
Do not add an intermediate subtotal again if the final display already includes it. Nor should each animation automatically be counted as another ordinary stake.
Use the completed-round record when comparing charges and returns. Feature totals and symbol values need their defined context.
The casino selects the available RTP version
Where a supplier offers several approved return configurations, the casino chooses which version to run. The highest percentage on a reference page should not automatically be assigned to every operator.
A concrete supplier example is Nolimit City’s San Quentin sheet, which lists 96.03%, 94.11% and 90.11% builds. These are published references, not verified settings in the featured NZ accounts. The Nolimit City guide keeps the supplied game sheets and their named builds together.
Read the actual title, build and active mode beside its return disclosure. A paid feature mode can also have distinct conditions, so a percentage should remain attached to what it describes.
A short run cannot identify the installed model. Keep the configuration record separate from how a particular sequence happened to perform.
Theoretical return and a measured result
RTP is theoretical expectation under a stated model. A measured return instead divides recorded prizes by recorded turnover for a particular period.
If a hypothetical session stakes $40 in total and returns $30 in prizes, its measured return is 75%. Its net gambling result is a $10 loss.
That measurement does not establish that the game’s theoretical RTP is 75%. A small sample can differ substantially from the designed average.
Turnover includes replayed credits. It should not be replaced with the deposit when calculating a session’s measured return.
Our RTP guide and calculator explanation separate these uses of the numbers.
Where jackpots and casino ceilings fit
A fixed maximum expressed as a stake multiple differs from a progressive jackpot pool. Read which system the complete game supplies.
For jackpot systems offering operator-configurable upper limits, the casino can select a ceiling within those settings. A centrally governed network pool can use different rules.
Keep the seed, live meter and upper limit separate. A must-drop condition applies under its defined rules and is not a personal promise that every participant receives a prize.
A bonus cashout cap and account withdrawal limit are additional conditions. Neither automatically replaces the amount the game can award. Our withdrawal-limits guide explains the account-payment side independently of the game’s mathematical cap.
See the jackpot guide for these distinctions.
The controls do not create a winning system
Changing animation speed or choosing when to press a button does not establish a method for predicting random outcomes. An attractive near miss is not evidence of a countdown.
A larger stake can change monetary exposure and, where the rules specify it, particular eligibility conditions. It should not be assumed to repair the effect of earlier losses.
Feature purchases need their own full cost read. A 100× purchase at twenty cents costs $20, which is much more than one ordinary twenty-cent action. The bonus-buy explanation covers the complete charge and chosen starting mode. If a promotion is active, the bonus-terms guide explains why game availability does not establish purchase eligibility.
A genuine demo can teach the interface, but virtual credits cannot be withdrawn. A short demo sequence does not forecast paid results or prove the setting installed in another account. The free pokies guide separates demo credits from bonus funds and cash. Our mobile guide covers finding the same help and total-price controls on a phone.
Checking an unexpected result
Retain the complete title, time, stake, currency and round reference. Note the mode, return disclosure and relevant symbol arrangement.
Compare it with the exact help information. An explanation from a different edition may use the wrong line system, trigger or modifier.
If a round freezes, inspect history before repeating a paid action. Official support can investigate the recorded settlement and configuration. A completed game award and an outgoing account payment are different records; the withdrawal guide explains that later payment process.
The guides directory and FAQ guide provide routes to related explanations. Keep spending and time limits independent of a feature appearing. Continue with how to play, the glossary or New Zealand support information according to what you need next.
Sources and teaching-model scope
The technical distinctions use the UK Gambling Commission’s explanations of random inputs and game rules and theoretical and measured RTP. These explain concepts, not New Zealand licensing or an operator audit. Named examples use the official Starburst, Reactoonz and San Quentin descriptions. The numbered-ticket models are original hypothetical examples, not real games or simulated paid tests.
How pokies work: questions
Does random mean every prize is equally likely?
No. Random inputs can map to differently weighted outcomes under a game’s mathematical model. The hypothetical ticket example demonstrates this without describing a real supplier’s code.
Does 95% RTP guarantee $95 after one hundred $1 rounds?
No. It is an average of possible returns. Independent draws can repeat outcomes and do not force every possible prize to appear once.
Is every non-zero return a profitable round?
No. Compare the return with the complete stake. Returning ten cents from a fifty-cent round still gives a forty-cent net loss.
Why do matching symbols sometimes pay nothing?
Their positions, count, direction or connection may not meet the actual game’s evaluation rule. Read its paytable rather than applying another title’s system.
Can a bonus feature use different rules from the base game?
Yes. Defined states can change the grid, modifiers or retained symbols. That is different from claiming a game must repay a person after losses.
Who selects an available RTP configuration?
The casino chooses which approved supplier version it runs. Check the actual complete title, build, mode and disclosed return.
Does a losing session prove the theoretical RTP changed?
No. A measured short-session result is different from the installed mathematical model. Retain the actual disclosure and round records if investigating a concern.
Can pressing at a particular time guarantee a win?
No such guarantee follows from random play. Timing patterns, near misses and past losses do not establish control of the next random outcome.



