You can't be serious posting that with no context; of course you aren't going to have 99% rtp or anything close when you end up doing 46 unit blackjack hands with a lose point of barely 200. Were you doing increase on loss, or just tilt betting? The RTP could be >100% and betting like that will still often lead you to a very bad result.

I know this whole topic is kind of a joke, so I wanted to contribute with some code. That's the only way, I find, to avoid user error when reporting the statistics of these games.

Note also that RTP (1 - house edge) is not magic or a directive; it's just the behavior of any casino wager being an 'unfair bet'. It can be found for any period by taking 1 - (pnl / wager). This is all-inclusive as far as bet sizes, changing bets, changing multipliers, etc. But if you are doing a very risky strategy, it will reflect back in that value, by deviating very high or very low. That is introducing volatility (a good/bad thing).

As for missing bet archives, or stakexposed or whatever- I do find missing bet archives concerning. None of that is relevant to whether og blackjack is listed house edge or not, because the implementation of the original games are fully open source and verifiers can be made independent of any of these websites. When trying to argue that RTP (house edge, which would imply the actual results are being tampered with to be less favorable like an adjustment in the middle attack, for games like blackjack, limbo, etc) is being changed or malleable, you have to first explain how they are able to predict and make actionable arbitrary HMAC inputs, which you can find more information about here: https://datatracker.ietf.org/doc/html/rfc2104