Burn 100 before calculating rating peak?

Has anyone suggested the idea of ignorning the first 50-150 games when calculating a player's personal highest and lowest ratings? I've been making progress in my live blitz rating over the course of the past year and a half, but my highest and lowest rating marks were both obtained in my first few games when there was too low of a sample size to accurately assess my skill level (and I suspect it will be some time yet before I top 1447). Ignoring the games before a player's rating stabilises might offer a more accurate assessment of skill and a more satisfying sense of improvement over time.

Try shaving, and ditching the Leon Trotsky hat.  Your rating should rise soon after.

Nice idea. 50-150 might be too many though, I'd stick with 10-20 games for the player to find his rating.

I had no idea Trotsky wore an English driving cap, nor that my pattern recognition skills were hindered by facial hair.

I know this was in good humour, so I can't help but mention in jest that I topped myself soon after deciding to grow a full mustache.