When Goat flipped over to level 52 I dutifully gave him another rank each of FA and Survival and set about to establish a baseline again, and then test the effect of kneeling, and of skinning in offensive stance. This time around I focused on just lesser and greater faeroths; vor'taz raw data is included for completeness but is insufficient to draw any conclusions.

lesser faeroths
criteria	standing, guarded	kneeling, guarded	standing, offensive
skin success	98%	97%	98%
value	1033	1023	937
magnificient	34%	61%	44%
exceptional+	84%	91%	89%

greater faeroths
criteria	standing, guarded	kneeling, guarded	standing, offensive
skin success	95%	92%	92%
value	1010	1028	946
magnificient	14%	16%	18%
exceptional+	49%	62%	53%

n=~100 for trials that fed the above tables

Conclusions
The data suggest that kneeling improves skin quality. The increase in skin quality without an increase in value suggests to me that the kneeling trials suffered some downward loot pressure from the earlier non-kneeling trials.

The data suggest (weakly) the offensive stance improves quality. The offensive stance trials came last, and so were subject to the strongest downward loot pressure, which may again explain in increase in skin quality but a decrease in skin value.

Bonus findings
With 64 ranks FA, 59 ranks survival, and 11 ranks trading, I appraised these skins successfully (getting a quality and value estimate rather than 'decent shape') in 272 of 383 attempts (71%).

It appears that estimated value assumes sale at a shop with no racial bias against the player, and also ignores trading skill. Goat has enough trading skill to pick up 6% higher prices. Accounting for that makes the average faeroth estimates accurate within a few percent (in aggregate).

I suspect that skinning is not an open roll - I have tracked 150+ vor'taz attempts and have never gotten a magnificient horn. I suppose it's also possible that magnificant horns become shiny vor'taz horns instead (imbeddibles that do not appraise as skins).

Future Research
I'm looking forward to setting this aside to go back to normal hunting, rather than playing like a robot. Still, I plan to eventually run trials on something with skins from the hand, and look at the effects of limb disrupting both hands before killing.

In-depth data follow

Baseline
level 52: don't kneel, and skin from guarded stance. skin with knife in left hand. Sometimes wounded for a few skins, in locations that are not thought to affect skinning (e.g. back minor). For one or two attempts, had a minor that is thought to affect skinning (eye, hand, etc). Unfortunate, but should not impact things too badly.

Lesser Faeroth
killed	100
skinned	98
botched	2
cumul_value	101315
avg skin value	1033
avg kill value	1013

Lesser Faeroth
quality	count	estimated value
crude	2	375
poor	1	200
fair	4	418
fine	10	495
exceptional	13	830
outstanding	16	1034
superb	20	1035
magnificent	33	1145
avg est value for all skin qualities: 943

Greater Faeroth
killed	100
skinned	98
botched	5
cumul_value	98992
avg skin value	1010
avg kill value	989

Greater Faeroth
quality	count	estimated value
crude	2	825
poor	5	770
fair	22	747
fine	21	835
exceptional	8	762
outstanding	12	954
superb	14	1092
magnificent	14	1325
avg est value for all skin qualities: 927

Vor'taz
killed	17
skinned	14
botched	3
cumul_value	5798
skinned, shiny	2
avg skin value	414
avg kill value	341

Vor'taz
quality	count	estimated value
crude	2	350
poor	5	285
fair	2	287
fine	1	750
exceptional	1	750
outstanding	1	1150
superb	0
magnificent	0
avg est value for all skin qualities: 445

Like baseline, but kneeling

Lesser Faeroth
killed	100
skinned	97
botched	3
cumul_value	99252
avg skin value	1023
avg kill value	992

Lesser Faeroth
quality	count	estimated value
crude	1	400
poor	0
fair	3	216
fine	6	412
exceptional	11	827
outstanding	6	800
superb	12	1050
magnificent	59	1060
avg est value for all skin qualities: 944

Greater Faeroth
killed	98
skinned	93
botched	8
cumul_value	95692
avg skin value	1028
avg kill value	976

Greater Faeroth
quality	count	estimated value
crude	1	300
poor	1	1350
fair	22	700
fine	9	722
exceptional	14	805
outstanding	11	1113
superb	18	1247
magnificent	15	1206
avg est value for all skin qualities: 962

Vor'taz
killed	11
skinned	10
botched	1
cumul_value	4984
skinned, shiny	1
avg skin value	498
avg kill value	453

Vor'taz
quality	count	estimated value
crude	3	375
poor	0
fair	3	358
fine	0
exceptional	0
outstanding	2	875
superb	1	550
magnificent	0
avg est value for all skin qualities: 500

Like baseline, but in offensive

Lesser Faeroth
killed	100
skinned	98
botched	2
cumul_value	91860
avg skin value	937
avg kill value	918

Lesser Faeroth
quality	count	estimated value
crude	1	350
poor	2	237
fair	2	412
fine	6	616
exceptional	14	716
outstanding	15	766
superb	14	1046
magnificent	44	945
avg est value for all skin qualities: 848

Greater Faeroth
killed	101
skinned	92
botched	8
cumul_value	87108
avg skin value	946
avg kill value	862

Greater Faeroth
quality	count	estimated value
crude	1	650
poor	4	631
fair	24	702
fine	15	790
exceptional	11	763
outstanding	9	1038
superb	11	950
magnificent	18	1119
avg est value for all skin qualities: 862

Vor'taz
killed	14
skinned	11
botched	2
cumul_value	5928
skinned, shiny	1
avg skin value	538
avg kill value	423

Vor'taz
quality	count	estimated value
crude	3	500
poor	3	358
fair	3	450
fine	2	1000
exceptional	0
outstanding	0
superb	0
magnificent	0
avg est value for all skin qualities: 538

I set a new baseline once I added the extra skin rank and extra surv rank, so that does not mess with these three trials.

Individual appraisals were somewhat inaccurate (based on spot tests of appraising the same skin multiple times). However, they were very accurate in aggregate. As such, imperfect appraisals are additional noise rather than bias. Note that I didn't draw any real conclusions off of value appraisals - the stuff in the charts up top are actual sales prices.