To build out list analysis, I’m attempting to create a general set of rules that categorize armies into archetypes. I previously used k-means clustering, but that approach isn’t very intuitive—and it’s hard to explain why a list ends up in a certain group. So I’m switching to a rules-based system instead.
Here’s the idea. For each tournament, I scale every list to a 2300-point baseline so they’re easier to compare. Then I look at a few key stats: expected melee damage, ranged attacks, average speed, total units, unit strength, and a few others. Based on how those numbers stack up—especially compared to percentile cutoffs—I assign one of several archetypes:
Current Archetype Rules
Alpha Strike: Fast and dangerous. Either above the 75th percentile both for speed and expected damage, or above the 90th percentile of speed.
Gun Line: Above 50 ranged attacks
Trash: Swarms of cheap units—either high unit count (16) and US (27) , or just extreme on one of them (17, 28)
Grind: Low offense, but takes a ton of shots to remove (defensive lists) (shots to six nerve above 395)
Mixed Arms (Shambling-heavy): Moderate shooting (at least 19 shots) and at least two Shambling units
Mixed Arms: Moderate shooting (at least 19 shots) but more flexible overall.
Balanced: Anything that doesn’t fit neatly above.
Behind the Scenes
Stats are scaled to 2300 points.
Thresholds are based on percentiles from a large dataset (e.g., top 25% for speed = Alpha).
All this is handled in a script (generate_dataset_for_tourney_comparison.py), which adds the archetype to each list.
I’ve hardcoded the current thresholds based on past events, but the plan is to update them over time as more data rolls in.
What changes would you make to capture the list archetypes better in a rules-based system?
Big same! I think it shows that alpha strike has faded pretty substantially in the face of NS and Dwarf hammer-and-anvil lists built for scenario and responsiveness.
We’ve seen some RFO lists relying heavily on tooled up fight wagon legions and helstriker hordes in the last couple years, however the competitive set seem to have given up on that concept (and/or those players didn’t qualify or want to travel to Reno this year). See above on alpha strike re: helstrikers being put aside.
Certainly the alpha GL list and the shooty/alpha SK one does struggle against dwarfs/NS trash type.
Possibly the GL/SK thing from my pov is slightly tilted, since a couple of really good players use them and do very well - punching more than the lists themselves are good?
The individual region breakdown is interesting - obviously given the geographical size of the US, players will get used to running across generally a similar group, so meta/styles develop. Then you either follow, look to run counter meta or just do your own thing?
Thing the army type breakdown is pretty sensible.
The shot count (does that include magic?) at the US masters was really gradual until you hit about the top 10. Is that something quite common with other events?
Is anything like effective shot range doable - since there is a lot of difference between AD with decimators, flame throwers and fireballs; dwarfs with xbows, sharpshooters and cannons; a mobile shooting list like NA with frostclaws, bolt throwers and 18" steady aim
Yep, for simplicity takes the highest number of offensive magic shots (so lb4 and fireball 8 becomes 8 shots). It’s possible to do effective shot range, but it gets super messy because of edge cases (artillery with multiple shot options). I always try to strike a balance between presenting a lot of data and yet making it accessible. I worry that’s so much detail even the data diehard will zone out
That’s fascinating!
It’s good to see that the NA and TR updates helped.
Mantic having a good selection of models now might have helped too.
Nightstalkers and Abyssals also got updated though.
I think Nightstalkers already had thier interest and were toned down.
Abyssals have data over 2 entries. Comparing Abyssals in 2023 to “Abyssals 2024” in 2025 gives a delta of +2.2.
Which is in line with TK, so new minis + rules appear to help popularity.
One quibble I have is with the results for Noble Undead.
I don’t think there are a statistically significant number of games.
0.3% of 2000 is 6 games? Sounds like 2 one days, possibly one player.
Which means the data could be more reflective of one player’s style, skill and/or a list they tried for a bit than the army as a whole.
I don’t see why the results should be significantly different from Undead proper. Which has more average kills and wins, for a smaller gap.
The other point about Noble Undead is that it isn’t actually a different faction, just set up separately on the Companion to assist with army building/allies [one list is Good and only LL listed is Jarvis, the other Evil and excludes him]. The UK Masters ranking roll them in with undead.
As mentioned above FoA and FoA24 are simply the old and new versions of the same army. Trident Realm were certainly played in the 23/24 tournament year, but when the new rules dropped there were two versions on the Companion (TR/TR24) and, unlike FoA, the old one was removed and the new version renamed.
With all the talk on fanatics, I was kind of surprised to see so clearly laid out.
I’ll try to remember to update this post at the end of the year. But so far, the community continues to look pretty robust, the same thing I saw in the UK kowmasters data. The handful of anecdotes people share about the community dying off in advance of 4th edition don’t seem to be borne out at this point.
TL,Dr: Basilea is fast, hard hitting, has lots of heal, and, based on the Mantic Companion Events, is outperforming most other lists on the table over the past few years. If I could find some cool angels to print I might try them out myself after seeing the data…
For the “Meta Bubbles”
Mantic and KoW are mostly in Facebook regarding the active online community, but a lot of people are not on facebook or left it the past years/months, combined with not all events being on the Companion, the companion missing support for club leagues or other long time events and not all people playing events, things might look worse if the one platform someone uses is favoring certain builds or sees a certain development that isn’t there outside that platform
Like we have seen one part of the community rating certain factions low while at the same moment other parts of the community were winning events with that factions
Also anything closed to “outsiders” (like FB posts cannot be read by people without account) or hard to find after a while (like on Discord), makes it hard to see a development but only the current state of things
still a shame that we lost all the discussions of the old mantic forum and warseer, as although there wasn’t a lot it would still provide context for some development.
Add in that a classic forum would still provide a better overview with communities from different time zones and languages with some discussions being impossible to pick up in detail if you are 8 hours late and can’t pick up the beginning
Talking Archetypes,
I have my problem with certain terms, although they are classic in GW based games, they don’t really fit into other games.
Alpha Strike being an example here as it usually means finish the game off by turn 1 or 2, and in our case I would use “glass cannon” for the high speed + high damage lists as opposite of the “grind” lists
also something I am missing are “spam” terms in unit sizes, MSU (multiple small units), MMU (multiple medium units) and MLU (multiple large units), which would translate to KoW as using more Troops (which includes Large Infantry/Cavalry and Swarm Regiments), more Regiments (including the Large Infantry/Cavalry Hordes) and more Hordes/Legions
a Goblin or Brothermark list with only Hordes called “Trash” might give the wrong impression, same as a Regiment focused Ogre list, so at least for me TMU (too many units/too much US) might be the better term
Also Balanced and Mixed Arms is more or less the same, as you get something of each and just because a list “only” uses 17 Lightning Bolts as their shooting doesn’t make it a mixed arms list
I take a lot of the online discussion about what is viable in list building with a pinch of salt.
The point made above - about the extent to which online KOW discussion of list building is self-reinforcing and “echo chamber” in nature is well made.
I have listened to many podcasts debriefs with top KOW players and tournament winners, and often they buck conventional wisdom in terms of playstyle and list building.
Piotr Nowak is one of my favourite examples, winning COK 23 with a halflings list that wouldn’t be considered obviously “strong” by the prevailing wisdom of dogmatic online commentators. And his interviews after the fact are great, because his explanation of how he plays the game, and the tactics he uses, are so wonderfully colourful and largely at odds with some of the dry meta discussion that goes on.
In one interview, he was asked repeated questions about why he hadn’t included various “strong” halfling units (brilliant and simple answer, he didn’t want to copy his friend’s army). Later he was asked how he managed to make light cavalry work, when the conventional wisdom was that they were too fragile and not punchy enough. His answer was something like “I send the dogs forward, and they look for flanks. It’s sad when they die, but it’s good to have lots of dogs to send.” I love it.
Of course he’s a great general, and it’s not that list-building is unimportant… But it’s just far less black and white than presented online at times.
100% agree and well said. There’s a lot of discussion about lists and list building, but after looking at thousands of game results, I don’t see the numbers backing that up. It’s possible to craft a terrible list, but it’s nigh impossible to win the game at the list-building stage. Good generals tend to win regardless.
