What you are owed
A duty that obliges an operator to watch play carries a matching set of things the player is owed: notice that a decision was taken, a reason that names the rule rather than a feeling, a route to appeal that is not the same desk that decided, and access to the record the decision rests on. Without those, a protection duty is indistinguishable from an unexplained restriction.
- things you are owed
- 4
- contacts with notice
- 2,400
- appeals
- 480
- appeals upheld
- 176
- appeal uphold rate
- 36.7%
- records available on request
- 4,000
Read the signals
Five comparisons the account already supports, run continuously rather than after a complaint. 6,000 flags in the sample, 96.0% of them relative to the account own history.
Choose a response
Nothing, a message, a limit or a break, or a restriction - one of four, graded by how much it takes from the player. 960 of 4,000 reviews in the sample changed access.
Write it down
Signals, crossing dates, rule version, path, response and outcome - one record per review, including the 1,600 that ended with nothing done.
Take the appeal
A route outside the desk that decided. 480 appeals in the sample, 176 of them upheld, which is 36.7% - high enough that the route does real work.
A protection duty owes the player four things: notice that a decision was taken, a reason that names the rule and the period behind it, an appeal to somebody outside the decision, and access to the record the decision rests on. In the sampled desk 480 of 2,400 contacts became appeals and 176 were upheld - a rate high enough that the appeal is worth making.
Four things, and why the fourth is the useful one
Notice is the cheapest obligation and it is where most policies stop. A reason is harder, because it requires the operator to say which rule was met, which is exactly the information that makes a decision testable. An appeal needs a route outside the deciding desk, since an appeal to the person who decided is a review rather than a remedy. And access to the record is the one that makes all three enforceable: a player who can read the signals, the dates and the rule version can check the decision rather than argue with it.
Sample E is the desk in numbers: of 2,400 contacts, all 2,400 carried notice; 480 became appeals; 176 of those 480 were upheld (36.7%), which is a high enough rate to say the appeal mechanism does real work rather than providing cover. A desk that upheld none of its appeals would be telling the player the route existed rather than that it worked.
What each thing looks like in practice
| Owed | What counts | What does not |
|---|---|---|
| Notice | a dated message naming the rule and the period | a silent limit with no message at all |
| A reason | the signal, the threshold and the crossing date | the phrase unusual activity on your account |
| An appeal | a reviewer who did not take the original decision | a reply from the same team, restating it |
| The record | the signals, dates, rule version, path and outcome | a summary that omits the threshold |
Each row in the right-hand column is the same failure in a different costume: the player is told that a decision exists without being given anything that would let them test it. When the record is available, all four rows become checkable at once, which is why this desk treats record access as the load-bearing right rather than a nice-to-have.
Making an appeal that can be upheld
- Ask for the record first, in writing, and ask specifically for the rule that was met and the dates of the signals behind it.
- Read the rule against your own facts. A ratio against a declared band, for instance, changes the moment the band is corrected.
- Answer the specific signal rather than the decision as a whole - 1,200 of 1,680 answers in the sample confirmed the action, and the 480 that changed it addressed a signal rather than the outcome.
- Send the appeal to the route the contact named, not to the address the message came from.
- If the appeal is refused, ask for the reason in the same form: the rule, the threshold and the dates.
- Keep the reply. The record of the appeal is part of the record of the decision.
Worked example / sample E
- contacts: 2,400
- answered within seven days: 1,680
- actions changed on an answer: 480
- 480 / 1,680 = 28.6% of answers changed something
- appeals escalated to a second reviewer: 480 of the 2,400 contacts
- upheld on appeal: 176
- 176 / 480 = 36.7% of appeals succeeded
The limits of these rights, stated plainly
These rights are policy, not physics, and nothing on this desk is legal advice. What a player is actually owed depends on the licence that covers their account and the terms they accepted, and those vary enormously by market - some regimes prescribe the contact contents down to the wording, and some require only that the operator have a policy. The figures here describe one sampled desk and should be read as a worked model of how the mechanism fits together, not as a description of any operator or regime.
- A declared band is the one input the account holder writes, so keeping it current is the cheapest way to influence the rule.
- An unanswered contact is not a pending decision; the action is already in place.
- A break with no stated end is the hardest decision to challenge, so ask for the end date first.
- An appeal is worth making: one in three succeeded in the sampled desk.
- Support routes are separate from appeal routes, and they exist whether or not an appeal succeeds.