Are paid invite-only TradingView indicators worth buying?

Updated August 27, 2026

What you’re actually buying

An invite-only script is a Pine script published on TradingView with its code hidden and its access list controlled by the author. Only Premium and Ultimate accounts can publish one. The author adds your TradingView username through a Manage access dialog, and from then on the script shows up in your Invite-only scripts folder.

TradingView isn’t in the middle of that transaction. For classic invite-only publications they don’t take the payment and don’t vet the seller. They moderate content, they can hide a publication or ban an account, and that’s it. You pay a stranger through whatever payment page they run, and they type your username into a box.

The newer Paid Spaces program is a different arrangement. There TradingView handles the subscription, you cancel it from your own account settings, and authors were selected after a review by their team. Not a quality guarantee about any script, but being able to cancel without emailing anyone is a real structural difference.

The one thing you cannot check

Closed source isn’t a minor inconvenience. It removes exactly the check that decides whether the pretty chart means anything.

You can’t see whether it calls request.security() with barmerge.lookahead_on and no offset, the pattern that hands a script tomorrow’s data on historical bars. You can’t see whether it branches on barstate.isrealtime and runs different logic live than on history. You can’t see whether the arrow and the condition behind the alert are even the same calculation. You can’t count the inputs, which is how many knobs were available to twist until the backtest looked good.

All of that is readable in ten minutes on an open source script. On an invite-only one it’s gone, and the rest of this is about working around that hole.

Patterns that keep repeating

None of these prove anything on their own. Two or three together and you know what you’re looking at.

• A single chart screenshot with the good signals circled. The interesting screenshot would be the same script during a chop day, and it’s never the one you get.
• A win rate with no drawdown, no trade count and no date range next to it. 82% is meaningless without knowing whether the losers were four times the size of the winners.
• “Non-repainting” printed like a certification. It’s a claim, not a property, and it takes about five minutes to test yourself. That’s covered in how to tell if an indicator repaints.
• A strategy tester screenshot with a vertical equity curve. Almost always percent-of-equity compounding with commission left at zero. The strategy tester page walks through why that curve shape is basically automatic.
• Defaults that happen to be perfect on the chart you’re being shown. Settings chosen after seeing the data are invisible from outside and are the most common reason a tool that dazzled in the demo does nothing on your screen.
• Urgency. Seat counts, closing soon, price going up Friday. Trading tools don’t sell out.
• The indicator being a doorway to a Discord where the real product is somebody calling trades.

TradingView’s own vendor rules ban “unrealistic or unsubstantiated claims or guarantees about your script’s performance”, and restrict pricing talk to the Author’s instructions field rather than the description. So a publication whose description reads like a sales page is already outside the rules, which tells you something about how the rest of it was assembled.

What you can check before paying

More than people assume, actually.

Look at the author’s profile. How long has the account existed, what else have they published, is any of it open source. An author with readable public scripts has handed you a work sample, and code quality is visible even when this particular script isn’t.

Read the comment thread on the publication, and pay attention to whether it’s a thread at all. Deleted criticism leaves gaps and one-sided praise.

Ask whether there’s a strategy version. If the seller will publish or grant an invite-only strategy() version, you can run the tester yourself on your own symbol with your own commission and slippage, and TradingView’s strategy publishing rules already require 100 trades minimum with commissions enabled. Refusing to produce one is not proof of anything, but it’s an answer.

Then get a trial with your own username and run it on your own chart for at least two weeks. Not a screen recording, a trial. In that window you can do the reload test, step it through bar replay, watch the live bar to see how much the signal moves before the close, and check whether alerts fire on close or intrabar. That covers most of what the missing source would have told you.

What no trial will tell you

Whether the parameters were fitted to the past. That’s the big one and it’s genuinely unknowable from outside. A tool tuned on 2019 to 2024 data can look coherent for a month purely by chance, and a two week trial has no power to tell that apart from anything else.

Also unknowable: whether the demo screenshots came from this version of the script, whether the seller trades it themselves, and whether it’ll still be there in a year. Access is a list on someone else’s account.

When paying is reasonable

There’s a real category here and it gets lost in the noise.

Buying a display or a utility rather than a signal. Session and volume mapping, a multi-symbol dashboard, an alert engine that does something fiddly you’d otherwise wire up by hand. The common thread is you can look at it and see whether it’s correct. It draws something instead of predicting something, so closed source costs you much less.

Buying time from someone whose open source work you’ve already read, monthly, cancellable, for a thing you could have built but won’t. That’s a normal software purchase.

What doesn’t hold up is paying a few hundred dollars for arrows on a chart from a seller you can’t identify, based on a backtest you can’t reproduce, with a source you can’t read.

Questions worth asking

Send these before you pay. The answers, and the speed of them, tell you most of what you need.

• Does the script use request.security() with lookahead, and will you confirm that in writing.
• Which symbol, timeframe and date range produced the results in your marketing, and what was the max drawdown and the trade count.
• Were the default settings chosen before or after that test period.
• Is there a strategy version I can run in my own tester with my own commission.
• Can I have two weeks on my own username before paying.
• Do the alerts fire on bar close or during the bar.
• Is it monthly, and can I cancel it myself without contacting you.

A seller who answers the drawdown question with an actual number is a different kind of seller than one who answers it with a screenshot. And if you get told the question is too technical, that’s an answer too.