ngram-keychords
Deciding appropriate letter-pairs we can use for keychording seems a daunting task, but by analyzing Google's N-Gram datasets, we can find the least common key-pairings in the English language, and avoid conflicts.
This isn't infallible; the pair "dk" seems to be infrequent enough to include in the list of preferable chord-pairs, but if you type "mkdir", "weekday" (!), or "bodkin" enough, your prose will eventually collide with a command.
I've included a list of keypairings I've bound to commands, along with some commentary about their utility, but if you want to do your own analysis, check out the datasets themselves:
Google N-Gram Datasets https://storage.googleapis.com/books/ngrams/books/datasetsv2.html
My preferred Emacs chording package is keychord.el, as demonstrated in
Emacs Rocks! Episode 7 http://emacsrocks.com/e07.html
The way you bind the command to a keychord using this package is:
(key-chord-define-global "fj" 'delete-local-region)
Here is the list of keychords I currently use:
;; Home row to number row keybindings are too stretchy
;; Numrow to numrow conflicts with *actual* number entry.
;; ;; Left-hand numrow chords
;;Natural toprow and numrow
"e1"
"e2"
"r2"
"r3"
"t3"
"t4"
;; Inverted toprow and numrow (only q is acceptable, ime)
"q3"
"q4"
"w4"
"w5"
"e5"
"e6"
;; Slightly awkward close pairings
"q1"
"w2"
"e3"
"r4"
"t5"
;; ;; Right-hand numrow chords
;; Natural toprow and numrow
"u9"
"u0"
"y8"
"y9"
"i0"
"i-"
"o-"
"o="
;; Inverted toprow and numrow (only [ is acceptable, ime)
"[-"
"[0"
"p9"
"p8"
"o8"
"o7"
;; Slightly awkward close pairings
"y7"
"u8"
"i9"
"o0"
"p-"
;; Alpha-only
"jk"
"kk"
"jl"
"jj"
"hk"
"hj"
"fg"
"fb"
"hb"
"hh"
"zj"
"zn"
"gj"
"gk" ; maybe conflict? had this commented out
"dj"
"fj" ; I use this for 'delete-local-region
"vv" ; I use this to kill-line and leave a new line
"vj"
"vh"
"vk"
"vm"
"vb"
"qw"
"qr"
"qp"
"qq"
"xq" ; awkward
"xh"
"zh"
"xx"
"zg"
"vf"
"zx" ; awkward
"xj"
"xk"
"bg" ; awkward
"xg"
"xb"
"qk"
"jt"
"yy"
"uu"
"cv"
"xv"
"l["
"l]"
",;"
",'"