Survivorship Bias

Up until the super series I was having a lot of success in skewing my lists towards assassination, in fact I had hit 15+ assassinations in a row (out of attempted assassinations not out of games won and lost) and the only games I had lost were games I had tried to play attrition in. I was starting to feel like assassinating was the way to go and that I was more likely to win that way than playing attrition. What I was overlooking was that I was playing smart and only assassinating when I had a very high percentage chance of the assassination succeeding. Vayl2 in particular encourages that type of smart play. Skewing my lists, in particular my Lylyth lists more towards assassination and going for assassinations in situations where attrition play was probably a greater percentage chance for success for me hurt me a great deal in the super series. That's what surivorship bias is by the way, it's looking only at those things that "survived" some process or thing. In this case I was judging the strength of assassination play by looking at how good the assassinations were that had "survived" my judgement process that the assassination was a good idea.

So I'm trying to learn from that and in that vein I think I will change my Kreoss list. The one I put up yesterday was:

Protectorate of Menoth
Standard Army
PCs: 35 / 35
26 models

[00] High Exemplar Kreoss (0 / 5 WJ remaining)
[08]  >> Reckoner
[06]  >> Redeemer
[06]  >> Redeemer
[02] Vassal of Menoth
[02] Vassal of Menoth
[01] The Wrack (3)
[02] Choir of Menoth (3 grunts)
[08] Exemplar Errants (9 grunts)
[02]  >> Exemplar Errant Officer & Standard
[03] Eiryss, Mage Hunter of Ios

The problem with this list is that the only thing that kills heavies in it is the Reckoner. The Redeemers need the feat in order to shoot most heavies or need the Reckoner to tag it first and then still miss 1-2 of their 4 boosted damage shots. That said it kills infantry dead and has a really strong assassination threat. Almost nothing can survive purification, knock down, Boosted POW 15, unboosted 15, 7 x POW 14s. (Or a boosted 15, two boosted 14s and unboosted 15 and three unboosted 14s).

Protectorate of Menoth
Standard Army
PCs: 35 / 35
27 models

[00] High Exemplar Kreoss (0 / 5 WJ remaining)
[08]  >> Reckoner
[08]  >> Reckoner
[02] Exemplar Errant Seneschal
[02] Vassal of Menoth
[02] Vassal of Menoth
[01] The Wrack (3)
[02] Choir of Menoth (3 grunts)
[08] Exemplar Errants (9 grunts)
[02]  >> Exemplar Errant Officer & Standard
[02] Gorman Di Wulfe, Rogue Alchemist
[03] Eiryss, Mage Hunter of Ios

This list has a much less vicious assassination threat. At best there are two boosted POW 15s and two unboosted POW 15s (as well as errant shots if needed). That's still a decent assassination threat. It also relies a lot less on the feat.

That or there is my old Kreoss list from before I bought errants:

Protectorate of Menoth
Standard Army
PCs: 35 / 35
26 models

[00] High Exemplar Kreoss (0 / 5 WJ remaining)
[08]  >> Reckoner
[08]  >> Reckoner
[06]  >> Redeemer
[02] Vassal of Menoth
[02] Vassal of Menoth
[01] The Wrack (3)
[02] Choir of Menoth (3 grunts)
[06] Temple Flameguard (9 grunts)
[02]  >> Temple Flameguard Officer & Standard
[03] Eiryss, Mage Hunter of Ios

This list relies a lot less on the feat turn than the above lists. Temple flameguard with defenders ward are ridiculous. Particularly if they are sitting just behind a Reckoner. That's quite possibly the most ludicrous thing ever. (DEF 13, +2 for defenders ward, +2 for set defence, +2 for Ashen Veil, it's like playing Kayazy except that your ARM is also ridiculously good and the unit costs less points and has reach). 

I don't know. I have to make a decision before Sunday. The last list is fully painted so that may win me over. That will also leave me with two lists with very little pathfinder (as in none really).