So, I spent some time observing game situations on dozens (to be precise, 80) of real-time online roulette tables to check what you wrote. I observed for about 5-6 hours straight, and my eyes got very tired looking at these numbers. Nevertheless, it was interesting to discover this new strategy for me.
First I can say it's a very good strategy; it plays as you wrote, and bets win in the majority of cases (I would say about 98-99% of the time).
Unfortunately, it doesn't win always: situations arise, as I assumed in my previous messages, when the number of misses exceeds the limit set by your strategy.
Initially, I decided to look at all corners that don't appear 26 or more times, find tables where there are more than 2 of such positions, and among them, there are no number intersections. It was very informative, but unfortunately, it was my mistake because there are many such tables, and it was challenging to track the relevant ones.
Nevertheless, I almost immediately (within an hour of observation) found the issue (I provide pictures below):
As you can see here, on the highlighted table (EZugi, Oracle 360 Roulette), there are two of the rarest corners — "19,20,22,23" — 56 misses in a row and "29,30,32,33" — 50 misses. I tracked this table among many where the misses reached 26. In the second picture, I give you the numbers that came up on this roulette so you can count for yourself if you don't trust me. I placed a blue circle above the number when the bet would have won (23 came up), so you can use this number as a reference and look at the numbers to the right and down in order.
By the way, this table is in a real casino where people place bets with chips. It just has the capability to accept bets online. I mentioned this to show that in this case, according to your strategy, I would have had to wait for less than 26 misses, which is even worse. Nevertheless, I waited for 26.
In about 5 or 6 hours of observation, I counted about 12 such situations where the minimum number of misses from two non-appearing corners exceeded 50, i.e., when your bet progression would have exceeded the declared betting limits. The maximum number of misses in my observations reached 61 times, this happened on the table Roulette Green, provider Pragmatic Play Live, corner "26,27,29,30" did not appear 80 times, and corner "8,9,11,12" — 61 times. Unfortunately, due to fatigue from observation, I forgot to take a screenshot.
Much later, I figured out to increase the observation limits for tables so that only those miss counts that exceed 50 are highlighted when tracking. This significantly simplified observation, but I did it quite late. Just another example for you in the screenshots below:
Table ``'Speed Roulette 1', provider Pragmatic Play Live, corner "29,30,32,33" — 55 misses, and "2,3,5,6" — 51 misses. Again, I give you the last 1000 numbers that came up on that table so you can check it yourself. Blue circle above a number displays that here your bets would win. Unfortunately, it would take too much money to bet after the displayed amount of misses.
I suppose the author of this strategy simply didn't have the opportunity to observe a large number of tables simultaneously to see the results I provided above. You can play for quite a while on one table and not lose once with this strategy, or you can lose right away — it's a matter of luck, and the author just got lucky.
In conclusion, I want to reiterate that the strategy is really good, and the only thing needed for its improvement is to wait not for 26 misses but at least 40 misses on the lower limit of non-appearing corners to succeed.
You can believe me or not — it's your right. I was genuinely interested in this approach, and I tried very hard to find out how successful it is, and now I can draw conclusions based on real data, not assumptions. Perhaps this information will help you (or someone else interested in this strategy) avoid losing money.