Optimising F2L

[Rob2109]

Hi guys, I’ve been working on optimising my intuitive F2L. I’ve written out all my algs with mirrors, checked against difficult cases on various sites and have what I hope is least amount of moves or finger tricky algs for each case.
I figured that if I can perform the hardest/longest algs in a second or so with practice I should suck a little less than I do (Lookahead notwithstanding of course)
The mean average move count of 41 cases * 4 came to 6.969... moves, so before spending time learning them can I optimise my F2L more? I’d appreciate any feedback, thanks.

 

[Zarxrax]

Don't focus too much on move count. The cases with fewest moves aren't always fastest.
Also, make sure you can do the cases from several angles, as this helps speed a good bit.

 

[frogmanson]

Luckily, the optimal cases for F2L are mostly the fastest cases (with a few exceptions). So don't worry about that.

 

[FatBoyXPC]

Seeings how your math went to just under 7 moves for each case, the average move count for each pair is 7, so you're right on the mark. It looks like you are right where you need to be if you want to be sub20. I got this math from Jessica Fridrich's website, I believe.

 

[Akuma]

You are really exaggurating when it comes to move count, just don't bother with them.

You shouldn't optimize them based on move count - you should optimize them based on simplicity and ability to improve the lookahead for the next pair. I would rather do an algorithm which requires 9-10 turns but gives me superior lookahead rather than a 7 turn algorithm with bad lookahead.
I usually stay away from algorithms that have the cross pieces floating on the top layer and algorithms that makes it difficult for me to see what the
heck is going on and where the edge/corner pieces are.
Having superior knowledge where all the corner/edge-pairs are while sovling the cube beats optimizing F2L. You can have your F2L algorithms as long as you want as long as you know where the rest of the corner/edge pairs are and where they're supposed to go *WHILE* solving F2L.


If you're obsessed with algorithms and execution time the only thing I would recommend you to practice on is OLL and PLL in that case.

 

[Rob2109]

Thanks for the advice guys, just hoped I was kind of on the right path. I have in some cases chosen a longer alg because it felt more comfortable and finger tricky, and I've also kept away from double slice moves as it messes up tracking pieces for me. Also, I've got all my cases for all four angles sorted so just comes down to practice now.
I probably am obsessing a little but I don't want to have to come back and relearn everything lol.