My friends and I started playing Warmachine back during the Apotheosis book release cycle (before Hordes was even out!) Back then, the game was much more limited in scope, so the goal of list building was pretty straightforward: build the "best" list you can for a given 'caster, with "best" often being qualified as "the list you can use against just about anything and do well."
A number of years and an edition change later, I feel like that definition has shifted. The game is much, much bigger than it was when we all started playing, and that added size and depth has made it commensurately more difficult to design lists to fit that old standard. A "best" list for a 'caster may be something that isn't particularly well rounded, but it asks a very direct question that the opponent has to answer effectively or else they will lose (a number of theme lists emphasize this style of list building, which I think helped open the mental floodgates to that type of theorycrafting.)
One of the reasons I feel the definition of "best list" has changed is the sheer number of possible challenges your list will have to deal with. At it's most extreme: there are around 141 warcasters and warlocks in the game (including yet-to-be-released Vengeance warcasters, but they're inevitable and mostly known at this point,) and each of those 'casters/'locks has a theme list associated with them that changes the type of list they can bring (some have more than one, thanks to NQ.) At base, that's around 280+ different challenges you may face off against, which is a helluva lot to plan for.
The saving grace is that the practical meta concerns are lower than that. It's way more practical to try to build your list to handle expected archetypes (infantry spam, heavy spam, anti-infantry, anti-armor, heavy ranged, heavy melee, etc,) and tweak your list based on any super specific things you think will monkey wrench you. However, even that can be a daunting task, as there are a lot of archetypes to plan for and most require their own considerations/inclusions.
One of the biggest factors to consider when trying to build a well rounded list is the capabilities of your chosen warcaster or warlock. This is (in my experience) where the concept of "power level" comes from: the more situations that a 'caster or 'lock can handle on their own - or the more problems they can successfully enable their army to overcome - the easier it is to build an "all comers" list with that 'caster/'lock.
As a result, when you're working with a 'caster/'lock that has a limited set of tools, you need to focus on what those tools are, how to best exploit those tools, and how to cover up any limitations that 'caster/'lock may have in their game. If your chosen figurehead has a particularly egregious set of weaknesses you may need to focus most of your list on overcoming those weaknesses, which is why some 'casters/'locks see little play outside of their occasional affectionate champion.
Looking at the Khador stable of warcasters, its pretty easy to see why players consider the "worst" to be as such. The most commonly provided names are some combination of: Karchev, Harkevich, Strakov, and Zerkova. All of these warcasters have serious flexibility issues that are evident right away when you read their cards, so a great deal of the challenge when using these models is trying to come up with a list that plays to their strengths while also overcoming their weaknesses.
That is also why it can be so addicting to theorycraft for models like them: for "stronger"/more flexible warcasters, either the list composition kind of writes itself (Sorscha2 is a good example) or its so open ended that you can do almost anything you want (Vlad1.) Trying to make the best of the worst imposes it's own interesting set of limitations ("how in the hell am I going to make a good list with this chucklehead at the helm?") while also providing you with tons of possibilities (you never know what combination may yield staggeringly good results.)
Across my many years of playing Khador, I've bought and played every single in-faction model available to the army, which includes all of the wacasters of course. Having played all of them, I can say that none of them are "un-winnably bad"; if you build a good enough list and keep your head in the game, you have a shot with any of them. However, the 'casters most often cited as the "worst" definitely have a harder time of it than the rest of the bench, and frequently run into uglier match ups as a result (running Karchev nowadays is often like playing Russian Roulette with 5 chambers filled.)
Out of the four Khador warcasters I listed before, I think Zerkova and Harkevich have the most potential for redemption. Karchev has serious design issues that seem to get compounded every release cycle, instead of getting any easier, and Strakov can be used well, but he's really unforgiving and unfortunately suffers from what I call "Early Mk. 2 Jank Syndrome." Harkevich and Zerkova have enough of a functional kit to put together workable lists without too much struggling, and I think both of them are worth messing around with because they each bring something interesting to the Khador bench (Karchev and Strakov are both fun to play, but are also somewhat redundant in terms of what they offer.)
So, I very often come back to Harkevich whenever I want to tinker around with lists. I've played him enough to feel like if I could just find that right combination of models, I'd have a list with him that I would really love. I don't think it'd ever be something I'd feel comfortable dropping in every situation (Harkevich's core set of a abilities is too limited for that) but I do feel like it's possible to come up with a list for him that could make him a good pairing in Steamroller, or a good third list in Masters.
To clarify: I don't think Harkevich is a stronger choice than the commonly accepted "best" Khador warcasters (push comes to shove I'd take any Vlad over him in a heartbeat.) More than I think he has unexplored potential that may make him a reasonable (if sub-optimal) choice competitively. I don't think he'll ever be the scourge of the tournament circuit, but I do think he has more going for him than most give him credit for.
I mentioned in a previous post some of the list theory I've had with Harkevich. I still really like his theme list, and I think I'll always be flip-flopping between the benefits of running the theme vs. the flexibility of a normal list. To shake things up, I used a warjack heavy list in the last game I played using Harkevich, and it performed pretty well. Having that many heavies let me trade pretty cleanly with my opponent, and I was able to handle his one infantry unit and support pretty well.
However, I can't shake the feeling that a good part of that was luck. I had at least one very crucial Critical Devastation with Conquest that helped clear out the Temple Flameguard, and I killed one of the Vassals due to a lucky spiked blast damage roll. Both of those were a huge help in whittling down my opponent's army, which ultimately set up the exchange and positioning that lead to me being able to win via assassination.
Remove that crit and spiked blast roll from the equation, and I'm not sure the game goes so well. Likewise, I think my opponent could have made my life a lot harder by being more conservative and careful with positioning his support, and by having Feora2 abandon the flag she was Dominating (in that scenario it was causing me to lose VP, but standing by that flag ultimately let the Drakun easily get a bead on her.)
The warjack heavy configuration ran into the problem that most "heavy heavy" lists run into: you typically have little in the way of redundancy, so everything needs to go mostly according to plan lest you end up screwed. Lists that run multiple heavies well usually have some way around that, and Harkevich has a little help with Fortune and Broadsides, but it's not enough to counter-act some of the issues he runs into (specifically: trying to deal with infantry using Khador warjack shooting.)
My initial thought was to treat the Harkevich list as a skew list, and go in with the mindset of only dropping it in favorable match ups. But I don't think that's very practical with how limited the list feels: even one unit of jamming/chaff infantry is enough to divert all my resources in that list, which feels like it tremendously lessens the ranged presence I have. I think that its unreasonable to expect to not have to deal with at least one squad of infantry in any set of opposing lists, and I think it's pretty reasonable to expect to see two units (even if one is more of a "workhorse" unit than the other.)
For that reason, I think I'm going to go back to my "old faithful" version of the theme list, or some variant thereof (which I'll touch on in the next post.) As much fun as it is to put 4-5 heavies on the table, it's really difficult for me to do that and feel like my list is flexible enough to deal with a significant enough % of the lists I'm likely to face. That's probably my more conservative side talking, but I feel like it'll be better in the long run, and I can always revisit the 4-5 heavy list if anything changes (i.e. Khador gets a sick anti-infantry warjack, though that will happen at the same time as Khador gets a unicorn.)
In typing all of this up my thoughts spiraled out into a "theme vs. non-theme" debate for Harkevich's lists, but I'll save that for a (hopefully soon forthcoming) follow up post. Keep an eye out!