Mr P, maybe I don't understand the task but the most "biased" number of your 300 spins is #13 which hit 14 times. This is a stdev of a little over +2 which is well within the norms of randomness. And I don't see how you can come to any conclusion regarding bias in as few as 300 spins without some corroborating visual evidence?
Should I be thinking more in terms of sectors in my analysis?
You are free to structure your bet as you wish( sectors, numbers, mix.. ) It's just a simple exercise.
Wheel is super biased. I selected 300 numbers in best conditions, so no need filtering from your part.
Next 100 numbers , wich l post later are from same conditions as previous 300.
Task is : find best numbers wich has higher chance to win in next 100 spins.
I give you a hint. My choice would be from 6 to 12 numbers. I would pay atention not only to best numbers ( wich hit the most), but these wich hit more consistently thru out the sample.
If you can, set up Kelly bet ( specify fraction).
Task has limits. Minimum bet- 1 pound , bankroll - 100 units.