Most of the literature I've read on BRM has pointed to a BRM of 250-300 average buyins, all the way up to 1000abi+ for professionals and high stakes players. Based on this I've generally worked within the 300-400abi range, with shots taken at bigger games. 

One thing I have not seen mentioned much though is how this should be altered based on field size. For example, on a ~$8000 bankroll an ABI of around $20 seems reasonable, however typically on Stars the tourneys of this buyin and lower often attract much larger fields than those slightly higher, especially if playing outside the 'peak' hours in European evenings. 

So my question is this: based on this $8k bankroll, would variance/risk of ruin be increased, reduced, or remain about the same if we exceeded these BRM limits and played the small field $27/$33/$55 type freezeouts that attract less than 500 runners (or even $33 1R1A with sub-200 fields) or stuck strictly within the guidelines and play $27 and smaller fields which potentially attract many many more entrants? I know that in practice a $20 ABI allows scope to play a range of buyins including all of these, this is more of a hypothetical question as to whether it is actually correct BRM to stick rigidly to an ABI ,or if variance can actually be reduced (and therefore bankroll better protected) by playing higher if the field sizes are reduced considerably as a result. 

My gut feeling is that short-term swings will increase due to bigger per-session outlay, but over the middle-long term variance would be reduced a little due to more frequent cashes and final tables. Much like we see with 180 SnG grinders vs MTT grinders. I don't have any maths to back this up however, and the articles I've seen tend to lump all 181+ field MTTs together under 'large field tourneys'. 

Thoughts?