i go with formula
speed Y = speed X * ( X / Y )^0.0691729
for equivalency of performances from 1500 - M ( & i have some trust in it )
comparing various 10k times with 5k :
- clarke 27'38 ( when his 5k wr was 13'16 ) ->13'11
so the 10k was stronger than 5k
- rono 27'22 ( & eventually got wr to 13'06 ) ->13'03
again, 10k wr stronger than 5k
- mamede 27'13wr ( wr was moorcroft's 13'00 & aouita ran 12'58 few years later when mamede still held wr ) ->12'58
moorcroft's wr was slightly weak ( according to numbers only, not circumstances ), but aouita's ended up pretty much equivalent
- then barrios knocked a bit of 10k to 27'08 ->12'55
- then yobes with 26'58 ->12'52 & then
sigei with 26'52 ->12'49 !
whilst 5k was still at a very weak 12'55
- in '95, geb smashed 5k wr with 12'44 ->26'44 & within coupla months, he did run 12'43
the 2 records were "equivalent"
parity continued for a while
- hissou 26'38 ->12'41 ( geb ran 12'41 next year )
- geb 26'31 in '97 ->12'39 ( & komen & geb ran that about that time )
- tergat 26'27 in '97 was beginning of 10k "running away" from 5k ->12'37
- geb 26'22 ->12'34 !
- kennster 26'17 ->12'32 !!
last 13y has seen 10k shoot off into the distance compared to 5k, which is somewhate surprising seeing as infrequency of the run
explanation i favour is that the 5k wr has traditionally been run by 5k/10k types ( geb somewhat of an exception with 3'31i speed - he shouda run much quicker for 5k ) & their 5k wrs are more based on 10k endurance - there's a limit to how fast 10k endurance will get you over 5k & it's been reached
i can see 10k still improving, but i really can't see the 5k ever really getting close to the 12'30 it shoud be at, in the forseeable future