I SPD-tuned my UNM UK team so that I could use the same team for most affinities. The challenge for Spirit was the 15% SPD debuff. For the first AoE #2, PK isn't protected by ME's block debuffs. However, with a minor boost to PK's SPD, I was able to make it work (on paper)...except that it doesn't work. For some reason, my Vizier kept taking a turn ahead of her even though my calculations said that shouldn't be the case. This eventually snow-balled into the Stun attack going before PK's heal, which messes up the whole team (because CB won't stun a hero with UK buff and <10% HP, which is why I need PK to heal first). 

Since I trust my math more than I trust your coding, I started looking into it. I thought that maybe your programmers stupidly increased the TM by 15%. This would make no sense and wouldn't be the same as a 15% SPD decrease, but then again, the whole TM system is idiotically designed, so I couldn't rule this out. When I used that assumption, PK still went ahead of Vizier.

I played around with some other possibilities and the only way I could get my spreadsheet to replicate what was actually happening was to apply a 30% SPD debuff (instead of the 15% debuff that is shown). When I do that, my spreadsheet, which perfectly matches the turn order for all other affinities, also matched exactly what was happening against Spirit. 

I don't know if this bug is restricted only to UNM, or it's happening for all CBs. I can make it work either way, but you should really make the debuff icon actually match the debuff that's being applied by either:

1. reducing the SPD reduction to just 15%, or
2. changing the icon to the 30% decrease SPD debuff icon