Armor stat weirdness

Hey,

There’s something odd going on with the armor stat when I simulate/BiB my blood DK. Even when the blender is set to ‘all toughness’, it for example suggests using a 915 str/versatility/leech stat stick over the 900 darkmoon card. But, when simulated, switching out the darkmoon card increases my chance to die by 10-15% in some cases. Likewise, the armor neck enchant seems to have a very low value at times. Furthermore, checking the stat weights after simulations (not using them for gearing, but it’s a rough guideline I suppose), the value of armor has dropped so low that it’s very barely above my second best stat in value, while it used to be worth 3-4 times as much as other stats.

Armor seems to simulate correctly, since it recognizes that unequipping the darkmoon card results in a massive death chance increase, but when it comes to the gearing suggestions in BiB and the item ranking lists, something seems to be really off regarding it?

The latest simulation where the problem occurs when I follow up with BiB:
http://www.askmrrobot.com/wow/simulator/report/3e019baeeae541c4b1b0c22b4ec70d17?weights=true&strategyId=b98c8d6c435140819b73e39ea7aa931c

It’s a custom mythic+ script, but the only change is a lower healer ilvl to better fit the people I play with. Everything else is untouched.

Thanks!

If you have the string of data that you copy from the addon to the website to do BiB handy, I can try to reproduce it and see what’s up.

Some comparisons -

Conservative suggestion - 398k NPS, 42.3% death chance, 811k DPS.
Conservative + manually adding darkmoon + neck armor enchant - 446k NPS, 21.6% death chance, 748k DPS.

A bit less damage, certainly, but a massive 20% lower death chance and 50k extra NPS. Not a good tradeoff for “conservative”.

All toughness suggestion - 435k NPS, 26.3% death chance, 746k DPS (still lower survivability than conservative + manual)
All toughness + manually adding the same as above - 459k NPS, 16.3% death chance, 748k DPS. (best survivability)

This mainly seems related to armor being evaluated incorrectly by BiB/item rankings. The darkmoon card is the worst offender by far, but armor overall seems to be rated too low as well.

Addon string here (the equipped gear is what “conservative” suggested):

$52;EU;Argent Dawn;Chiyoe;Potato Paranoia;2;1;110;13:800,7:800,15:800,12:800,14:800,5:810;1;.s1;1;3112132;272,1,1,1,1,1,1,1,1,1,1,1,1,1,1,3,1042,28,122,1,1,1,143;4,4,4,4,4,4,4,4,4,1,1,1,1,1,1,1,1,1,4,1,1,16,1;137363b1607b3337b3418,137399b1582b3337b3536,137543b1572b3337b3535;1,1739,35,-4,-1491,-2,-1,1,0,0,0,0,0/2,1463,35,3,-1497,-2,-6,1,1,0,0,0,0/3,1467,40,-9,-1492,-3,-1,1,0,0,0,0,0;.s2;2;2213221;108,1,1,1,2,1,1,2,2,1,2,1,1,966,1,1,240,28,125,1,1,1,140;4,4,4,4,4,4,4,4,1,1,1,1,1,4,1,1,1,1,4,1,1,1,1;137308b1532b3415b3528,133684b1562b3528b3536,142185b1507b3418b3528;1,1739,42,2,-1674,8,-9,1,0,0,0,0,0/2,1631,42,-6,-1658,-9,5,1,0,0,0,0,0/3,1626,35,10,-1676,7,975,1,0,0,0,0,0;.s3;3;0000000;149,1,1,1,1,1,2,1,1,104,1,1,1,1,1,852,214,28;1,1,1,1,1,1,3,3,3,3,1,3,3,3,3,3,1,1;143701b1547b1824b3337,144464b1522b1727b3336,137302b1522b1727b3336;;.q1;128402s16b716x137363y36z144e3368;3963s9b2743b111;1770s5b-3529b1536b1841b110;48s7b-1931b1740b81;85s10b-1841b1759b200;244s1b-1954b1755b199;15s11b-1954b1755b199e2060;2s2b-1964b1764b82e10;2241s15b-1836b1755b199e-4;753s6b-1964b236b1528b200x14040;6s8b-1964b1764b81;1573s13b-3376b566b970b1759b237;2380s14b-2101;2799s3b1987b70b41e497;200s12b-1998b236b1528b200x0e-503;.q2;128292s16b717x137308y-3624z8501e3370;1s17e-2;4150s8b1094b1759;1199s13b-1988b1755b199;541s7b-1939b1740b81;3s9b-1846b1956b8;82s10b-1959b1759b200;92s3b-1949b1750b199e2515;140s5b-1954b1755b199;25s12b-1959b231b1529b199x-11939e-456;17s11b-1954b1755b199e0;2228s15b-1954b1755b199e7;675s1b-1954b1755b199;1657s14b-2931b962b1770b236;3072s2b-2061b1998b18e4;4493s6b-69b111x-27y0z0;.q3;128403s16b718x143701y763z-7162;4050s7b1093b1644b3;1318s15b-1906b1785b199e5434;408s5b-1974b246b1529b199x-7081;90s6b-1994b1794b173;89s1b-1947b1775b199;181s12b-1989b1790b173e-4;2436s13b-1978b276b5b1722x-2;442s8b-2003b276b5b1722x0;115s11b-1978b1780b81e-1;3361s9b-1936b1855b183;14s10b-2028b1845b177;575s14b-2042;692s2b0b336b1644x27e10;2107s3b-1641b1718b1e444;.r;_;.inv;3419;0;0;4;1;3524;10256;1358;1;77;2175;383;8105;436;1071;1282;746;25;957;514;3495;34;4917;227;2031;953;734;149;2841;193;318;5;20692;5138;70;9;1;1;592;9409;404;852;0;1713;0;0;0;0;0;0;2765;801;3;1;1;58;2603;3371;0;2814;720u493x83141e4421;1u0x-6468e1;8310;1260;1861;0;140;1;1;2331;114;106;41;0;0;0;1;1;1;1;1;1;1;1;1;5;968;2860;1867;4;210;1;1;1;1577;1;0;0;0;188;0;0;0;1;0;0;0;1;0;1;1;0;0;1;0;0;0;5;0;0;0;0;0;0;1;0;0;1;1;0;0;0;0;0;1;3;0;0;0;0;0;0;0;0;0;0;0;0;1;312;1;0;1;107u36b42;88b593b14e904;3b-8b9e-1;32;0;0;3086;79;5;10;0;1;442b67x60635y-3624z8501e-1955;12;49;50b1x1516y763z-7162;100;4;3;201b-117b78b10;389;1072;1;1;1;1;1;1;1;1;1;1;1;720;2;812;27;623b2770b70b41;9b-111b111e2064;67b-1759b1759;1b-111b70b41x-7081e455;4b-111b111;4b-111b111x-2e-462;1b-111b111;5b-111b111;53;3;1051;2;1;1;4;6;28;23b-2013b1780b81;2b-1876b1876b110;0b-1986b1876b110;10b-1946b1755b199;5b-1974b1966b45;33;0;0;85b-2031b266b1610b110;6b-1966b246b1720b8;203b-1725b1759x27e1;205b-1988b1755b198;7b-1963b1956b8;4b-2929b965b1956b45;0b-2968b967b1956b45;0b-2970b929b1805b137;0b-3434b564b978b1754b237;38b-1996b1759b200;40b-1974b1774b82;54b-2814b968b1956b45;38b-3530b1559b1735b81e503;2b-1866b256b1610b110;43b-1966b1775b81;1b-1861b1971b7;2b-3494b1516b1971b7;2b-1973b1775b81e-497;80b-3377b1511b1976b6;10b-1952b1755b199;1b-3494b1530b1764b82;2b-1861b1971b7;5b-1993b266b1528b80;1b-3374b1495b271b5b1723;2b-1959b1759b82;7b-1886b1883b113;2b-1981b1963b18;2b-1986b1876b110;3b-1951b231b1529b199x0e-7;1b-1999b1880b111;4b-1986b266b1610b110;1b-1941b1750b81;1b-1831b1750b81e1;1b-1871b1789b82;8b-3378b1517b1780b81;1b-1836b1755b199e-1;1800;8;22;400b-1979b1971b7;203b-2003b276b5b1722x0;3b-1958b1760b236;337b-2011b246b1720b45;23b-2011b1966b45;6b-2006b1769b200;3b-1974b1966b8;3b-1969b1851b110;9b-1971b1780b81;2b-1866b1784b82;2b-1866b1784b82;6b-1926b235b81b5x-25;1b-241b1765b197;30b-1972b1966b45;19b-2061b296b1528b79;21b-3373b1495b271b1528b237;5b-1991b1755b199;37b-1979b1971b7;1b-1963b236b1528b82x21362e7;4b-1891b281b1529b80x-21362e455;12b-1860b1780b81;25b-1861b251b1529b81;1b-1876b1876b110;2b-1966b1966b8;5b-1979b1780b81e-459;1b-1841b1759b82;2b-1856b1966b8;1b-1974b1966b8;106;469;666;279b-1964b1956b8;39b-1979b251b1702b18;7b-2923b977b1754b237;693;7;7;2;1;2;62;1;1;127;187;379;2;9;1;184;130b-2081b1845b178;102;1;535b-3473b1480b2006b45;70;47;465;7b-2046b2001b8;3b-2014b2006b8;1b-2014b286b1528b199;33b-2008b2001b45;5b-2096b2051b45;2b-2031b1795b236;1b-2076b1919b112;1b-2006b1815b81;1b-1886b1805b81e0;2b-1896b1815b81e0;1b-1906b1998b18e8;33b-2006b2006b8e451;8b-2019b2011b7;89b-2013b1815b81;32;0;70;102b-2815b924b281b1720b45;1741b-114b111;219b-2013b1780b81;8b-1876b266b1610b110;2b-1961b1769b199;4b-1973b1966b45;15b-2021b1784b82;3b-1866b1784b82;2b-3375b1514b1780b81;244b-3378b1497b271b1529b236;488b-2021b1784b237;844b-3532b566b935b1986b45;589b-114b111x-2y0z0;318b-2083b1849b225;40b-2064b1839b237;40b-2081b1844b225;129b-3519b1465b1829b226e-459;84b-2959b2566b167b237;173;103;3;15;135;43b-2091b326b1720b44;4b-2090b2046b44;183b-2030b134b1661b242;1781b-2007b1956b45;1292b-399b162b60;0b-222b223b131;1b-349b157b61;2b-223b223b131;178;121b-1991b190b1801;255;1;11;5;56b-1717b1759x21366e0;80;75b-111b111;1b-111b111;642;343b-2098b2056b101;1b-2152b331b1528b293;153;64;22;2;48;119$

I did take a look at this… does seem to be something a bit further off than desired in this case, though a 10k NPS difference isn’t that big, given the margin of error being used.

Our adaptive strategies for tanks that I am working on this week and next should help out a bit – I’ll revisit this after those updates if the issue still persists.

I played around some more with this. As far as I can tell, it was happening when simulating reasonably difficult m+ levels with the raging affix selected. That obviously has the potential to be extremely unpredictable and dangerous, and seems to be what caused the odd results.

All the suggestions and results make a lot more sense if I sim with fortified+tyrannical+nothing selected, as that’s a relatively small and more stable damage increase compared to raging.

Definitely looking forward to those adaptive strategies though!

Alright, I had this happen again. Seems like raging wasn’t the problem, just a coincidence.

I did a new simulation, and ended up with an extremely low rating for armor again, compared to normal, if looking at the stat weights. When I then ran BiB with the blender on “conservative”, it again wanted me to sacrifice 70k NPS (increasing my death chance from 10% to 40%) to gain around 60k DPS. Again, because it removes a ton of armor with the trinket + neck enchant.

In the simulations where the armor value ends up being correct (3-4 times higher than in these weird ones), BiB + conservative correctly improves my DPS a little bit while keeping my NPS/death chance roughly in the same spot. In these ones however, it sacrifices a ton of survivability by dropping all that armor.

I’m not really sure what’s going on there. In some simulations, armor gets a high value and there’s nothing weird going on, and in others, it’s really low and BiB tries to kill me. I can’t think of anything significant that would have changed between the different simulations.

This is my most recent one where the problem occurs:

http://www.askmrrobot.com/wow/simulator/report/71ddbf86aa914ac38e17f956b96206f8?weights=true&strategyId=5e0b0781dc5c4779a0c97f09300b8dc5

That is strange… I can take a look. If you look at the stat weights in that report you linked, armor is the highest stat… which indicates it must be giving it a decent amount of value… not sure why it would want to drop a lot of armor when optimizing.

I’m going to start the Adaptive blood strategies sometime today… which should complete sometime tomorrow, we’ll see how those look in comparison.

Yeah. In the simulations where the problem occurs, the value of armor is only slightly higher than the second best stat. Compare to this other recent simulation I did, where everything is working perfectly. Here, armor is significantly higher than the other stats:

http://www.askmrrobot.com/wow/simulator/report/7044094ecf6c432ea40b395c4cf6e4b5?weights=true&strategyId=a01d47263ac64b1f9ba8f4f27e601d37

In the simulations where armor is high, BiB + conservative keeps my NPS around 475k and death chance around 10%, while slowly trying to improve DPS. I assume this is what’s intended? In the simulations where armor is low, it sacrifices an unacceptable amount of survivability for a somewhat larger DPS gain.

When I look at the stat weights from one of the problem sims, I can see a pattern. It seems like when the stat weights were generated, it thinks I have a lot more survivability than I really do. The value of armor and stamina is low, and the value of leech/avoidance/crit is high, compared to the accurate sims. This might explain why BiB in those cases thinks it’s okay to get rid of so much survivability?

I don’t know why it happens though, there are no significant changes in gear/talents/settings between the sims as far as I can tell. Looking forward to see how the adaptive strategies turn out though!

I’m still experiencing the same issue, both with any custom simulations I run, and with the new adaptive strategies (only tried m+ 14-17 adaptive so far, since that’s what I mainly run).

Armor is valued very low, and BiB doesn’t like to equip the Darkmoon trinket and/or the neck enchant, even with ‘All Toughness’ selected in the blender. Manually equipping them gives a large boost for survival when simulated.

The value of all the other stats makes perfect sense to me, based on the difficulty of the content and the amount of focus on DPS, but something’s up with the armor.

Do you have a recent addon export where this is happening that I could try out?

Here you go, this is my latest setup. Out of my 10 latest custom m+ simulations or so, only one of them led to BiB suggesting that I should keep the armor boosts equipped. I don’t know what was different that time.

$54;EU;Argent Dawn;Chiyoe;Potato Paranoia;2;1;110;13:800,7:800,15:800,12:800,14:800,5:810;1;.s1;1;3112133;272,1,1,1,1,1,1,1,1,1,1,1,1,1,1,3,1042,28,122,1,1,1,143;4,4,4,4,4,4,4,4,4,1,1,1,1,1,1,1,1,1,4,1,1,21,1;137363b1607b3337b3418,137347b1562b3528b3536,137543b1572b3337b3535;1,1739,35,-4,-1491,-2,-1,1,0,1,0,0,1/2,1463,41,2,-1503,1202,-1201,1,0,1,0,1,0/3,1459,40,-9,-1492,-3,-1,1,1,0,0,0,0/137544b1577b3336b3418,1465,40,5,-1507,3,-2,0,0,0,0,0,0/144464b1562b3417b3528,1461,40,-4,-1502,5,-6,0,0,0,0,0,0/151011b3179b3336b3397,1467,32,4,-1497,-3,2,0,0,0,0,0,0/151013b3184b3336b3397,1462,32,4,-1503,6,-4,0,0,0,0,0,0/151011b3179b3336b3397,1465,35,10,-1511,-1,6,0,0,0,0,0,0/137350b1567b3418b3528,1461,40,3,-1502,-1,-5,0,0,0,0,0,0/137363b1552b3336b3418,1465,35,1,-1501,-1,3,0,0,0,0,0,0/151292b1567b3418b3528,1463,39,5,-1506,-3,1207,0,0,0,0,0,0/133763b1567b3418b3528,258,41,-10,-1490,-2,-3,0,0,0,0,0,0/151011b3179b3336b3397,1464,32,4,-1496,-7,1,0,0,0,0,0,0;.s2;2;2213221;108,1,1,1,2,1,1,2,2,1,2,1,1,966,1,1,240,28,125,1,1,1,140;4,4,4,4,4,4,4,4,1,1,1,1,1,4,1,1,1,1,4,1,1,1,1;137308b1532b3415b3528,133684b1562b3528b3536,142185b1507b3418b3528;1,1739,42,2,-1674,8,-9,1,0,0,0,0,0/2,1631,42,-6,-1658,-9,5,1,0,0,0,0,0/3,1626,35,10,-1676,7,975,1,0,0,0,0,0/137340b1567b3418b3528,649,35,10,-1667,-6,979,0,0,0,0,0,0/151010b3174b3336b3396,649,42,3,-1669,-1,-6,0,0,0,0,0,0/137380b1572b3336b3418,1631,40,3,-1669,-2,-1,0,0,0,0,0,0/137370b1572b3337b3534,1629,41,4,-1674,-2,1,0,0,0,0,0,0/151010b3174b3397b3528,1630,35,10,-1676,3,4,0,0,0,0,0,0;.s3;3;0000000;149,1,1,1,1,1,2,1,1,104,1,1,1,1,1,852,214,28;1,1,1,1,1,1,3,3,3,3,1,3,3,3,3,3,1,1;143701b1547b1824b3337,144464b1522b1727b3336,137302b1522b1727b3336;;.q1;128402s16b716x137363y-16z196e3368;309s13b-115b78b10;3654s9b2770b111;1269s11b-1963b1730b81e2062;2s2b-1841b1759b82e459;854s12b-1821b1740b81e-459;15s8b-1846b1764b82;7s1b-1836b1755b199;2s7b-1934b1735b81;3009s6b-1846b236b1528b200x14042;8s15b-1949b1750b81e4;3951s14b-1946;2799s3b1987b70b41e497;221s10b-1978b1745b81;7876s5b-1936b1854b278;.q2;128292s16b717x137308y-3624z8501e3370;1s17e-2;4150s8b1094b1759;1747s14b-3530b564b978b1754b237;170s3b-3530b1559b1735b81e2563;130s12b-1821b1740b81e-501;24s7b-1816b1735b81;18s11b-1831b1750b81e0;2805s9b-3378b1527b1851b110;108s1b-1946b1755b199;86s15b-1949b1750b81e4;4643s2b-1906b1998b18e4;2328s10b-1936b1745b81;2165s6b41b111x-11966y0z0;611s13b-2967b2566b167b237;5100s5b-2091b1854b278;.q3;128403s16b718x143701y763z-7162;4050s7b1093b1644b3;1318s15b-1906b1785b199e5434;408s5b-1974b246b1529b199x-7081;90s6b-1994b1794b173;89s1b-1947b1775b199;181s12b-1989b1790b173e-4;2436s13b-1978b276b5b1722x-2;442s8b-2003b276b5b1722x0;115s11b-1978b1780b81e-1;3361s9b-1936b1855b183;14s10b-2028b1845b177;575s14b-2042;692s2b0b336b1644x27e10;2107s3b-1641b1718b1e444;.r;_;.inv;3419;0;0;4;1;3524;10256;1358;1;77;2175;383;8105;2789;746;25;363;594;514;3529;4917;227;2031;953;734;149;2841;193;318;5;20692;5138;70;9;1;1;592;9409;404;852;0;1713;0;0;0;0;0;0;2765;804;1;1;2661;3371;0;2814;720u493x83141e4421;1u0x-6468e1;8310;1260;1861;0;140;1;1;2331;220;41;0;0;1;1;1;1;1;1;1;1;1;5;968;2860;1867;4;210;1;1;1;1577;1;0;0;0;188;0;0;0;1;0;0;0;1;0;1;1;0;0;1;0;0;0;5;0;0;0;0;0;0;1;0;0;1;1;0;0;0;0;0;1;3;0;0;0;0;0;0;0;0;0;1;312;109u36b42;88b593b14e904;3b-8b9e-1;3118;79;5;1;9;0;1;442b67x60635y-3624z8501e-1955;61;50b1x1516y763z-7162;100;4;3;590;1072;1;1;1;1;1;1;1;1;1;1;1;721;1;812;27;623b2741b70b41;9b-111b111e2064;67b-1759b1759;1b-111b70b41x-7081e455;4b-111b111;4b-111b111x-2e-462;1b-111b111;5b-111b111;1b-1759b1759;52;3;1051;2;1;1;4;0;6;28;25b-2003b1851b110;5b-1941b1750b81;5b-1836b1755b199;5b-1959b1760b236;32b-3531b1535b1759b82;1;0;0;83b-1851b1851b110;4b-1961b1851b110;4b-1966b246b1720b8;203b-1725b1759x27e1;161b-3529b1536b1841b110;6b-1951b1841b110;5b-3487b566b965b1956b45;33b-1991b1755b198;4b-1938b1740b81;3b-3376b1545b221b1528b82;0b-1846b1956b8;4b-3496b564b978b1754b237;38b-1996b1841b110;0b-1951b1759b200;2b-1959b1841b110;38b-1966b1774b82;0b-1841b1759b200;10b-2929b975b1754b237;78b-1991b1754b82;4b-3375b1559b1735b81e503;0b-1841b1841b110;2b-1976b256b1610b110;4b-2923b967b236b1720b45;39b-2011b1775b81;1b-1861b1971b7;2b-3494b1516b1971b7;2b-1973b1775b81e-497;80b-3377b1511b1976b6;10b-1952b1755b199;1b-3494b1530b1764b82;2b-1836b226b1529b81;5b-1876b266b1528b80;3b-1839b1759b82;9b-1871b1963b18;2b-1986b1876b110;3b-1951b231b1529b199x0e-7;2b-1954b1755b199e1;2b-1964b1764b82;0b-1846b1764b82e461;1b-1876b266b1610b110;1b-1941b1750b81;1b-1831b1750b81e-459;1b-1871b1789b82;7b-1851b1851b110;1b-3488b1517b1780b81;1b-1836b1755b199e-3;1800;8;420b-1954b1755b199e7;2b-1979b1971b7;206b-3495b1522b1966b45;0b-1996b1760b236;245b-114b111;92b-2008b246b1720b45;22b-3533b1527b1851b110;3b-1961b1851b110;4b-1956b236b1528b237;6b-2006b1851b110;5b-1961b1851b110;6b-1976b1784b82;1b-1851b1769b237;1b-2021b1784b82;6b-1926b235b81b5x-25;1b-241b1765b197;10b-1962b1764b82;20b-1856b1966b45;18b-1986b1750b81;1b-3378b1527b1851b110;11b-1966b1966b45;10b-3531b1495b271b1528b237;5b-1991b1755b199;37b-1979b1971b7;1b-1963b236b1528b82x21362e0;16b-1861b1780b81;30b-1846b1764b81;3b-1860b1780b81e-4;1b-1831b1750b81;2b-1856b1966b8;1b-1974b1966b8;2b-1974b246b1720b45;6b-1996b1759b82;98;469;666;279b-1846b1956b8;39b-1979b251b1702b18;7b-2923b972b231b1528b237;0b-2968b977b1754b237;0b-3532b566b970b1759b237;0b-2970b974b1759b237;138b-1991b226b1529b81;555;14;2;1;2;62;1;1;127;187;379;2;9;1;184;232;1;535b-3376b1480b2006b45;70;47;465;7b-3531b1490b1886b110;0b-2001b2001b8;3b-2014b2006b8;1b-2014b286b1528b199;2b-2008b1891b110;38b-1986b1795b236;1b-3533b1482b2006b45;1b-2051b1815b81;1b-1886b1805b81e0;2b-1896b1815b81e0;1b-1906b1998b18e8;33b-1991b1799b82;301b-2815b924b281b1720b45;1741b-114b111;10b-111b111;205b-2008b1855b111;14b-1961b1769b199;3b-1963b236b1528b200x2e-8;1b-1974b246b1720b45;15b-2021b1784b82;4b-1851b1851b110;1b-3485b1514b1780b81;244b-3378b1497b271b1529b236;488b-2021b1784b237;1433b-114b111x-21366y0z0;318b-2083b1849b225;38b-2059b1834b226;2b-2065b1839b237;40b-2081b1844b225;129b-3519b1465b1829b226e0;84b-2959b2566b167b237;173;103;3;15;135;230b-2031b134b1661b242;184b-2032b1789b237;2889b-399b223b131;0b-344b153b59;0b-222b162b60;1b-217b157b61;0b-218b157b61;0b-218b157b61;2b-213b152b61;178;101b-1830b1851b110;17b-1961b241b1610b110;24b-3486b1530b1764b82;234;1;0;0;11;5;56b-1607b1759x21366e0;78;1;1;1;1;73b-111b111;1b-111b111;407;176b-2098b2056b86;59;7;337b-3573b1461b306b1720b45;1b-3532b1471b1824b237;152;64;0;22;2;48;101b-2071b306b1720b45;18;0;36$