I think you're right that the 50k+ sell at 3:56:42/43 was not...

I think you're right that the 50k+ sell at 3:56:42/43 was not BG, however I believe there is a good chance that the 7000 shares sold in the few minutes afterwards until close could have been BG's shares.

Up until that big sale, there had only been 52048 shares sold for the day. Based on the assumption that historically BG's shares make up 12%-15% of daily volume, that means he would have sold somewhere between 6200 and 7800 shares for the day up until that point. However, after the same of 53429 shares at 3:56:42/43 (almost doubling the daily volume traded so far), BG's share of volume would have dropped to approximately 6-7% of the daily volume.

What would need to be sold to get back to the 12%-15% range? As shown below, 12% of 53429 is 6411, 15% of 54329 is 8014, meaning that somewhere between those numbers would bring BG's share of volume back up to the historical average.

As it turns out, there were 7000 shares sold between that big sale and close - one parcel of 3000 at 3:57:28 and one parcel of 4000 at close. Possibly coincidence, but perhaps more likely was that this was Shaws selling BG's shares.

I think this also demonstrates the impact that those sales can have on the RAC share price, as to guarantee the volume they must be constantly sold into the bid stack and they have to take whatever that price is. And the second thing is that anyone else selling begets more selling from Shaws to keep the volume ratio. Lastly, we now have an approximately 0.5% (or slightly less) of SOI holding continuously selling 12%-15% of daily volume, so until BG's holding is reduced to zero, the relative daily impact on shares available to sell gets higher and higher.