Are Fourth-Edition Kings of War Lists Getting Less Diverse?

Since fourth edition launched, I’ve seen a comment pop up on Fanatics regularly: army lists feel more repetitive. Sometimes that means repetition inside a single list (e.g. “spam”). . Instead of two or three copies of a useful unit, players are taking four, six, or even more. There is a second concern too. When several people bring the same faction to a tournament, their armies can feel like variations on the same basic package. The last 500 points might change, but the core of the army keeps showing up.

At Data & Dice, I like to take that contemporary wisdom and put it to the test. At this point, we now have enough early fourth-edition tournament lists to put some numbers behind both questions. The answer here is mixed, but the concern has some support. Early 4e lists contain more heavily repeated units, and factions with enough players to make a meaningful comparison are showing more similarity between lists. The effect is concentrated in particular factions rather than being universal.

Two different kinds of list diversity

I split the question into two parts: The first is repetition inside an individual army. How often does a list take the same unit at the same formation size several times? The second is convergence within a faction. If five Goblin players attend the same tournament, how much of their unit selection do they actually share?

Those are related questions, but they are not the same. A faction can encourage heavy spam while still supporting several different builds. It can also have relatively little spam while almost every player starts with the same six or seven units.

The data

I track and archive as much high-quality tournament list data as I can. The late-third-edition sample contains 507 lists from 10 tournaments, covering 7,041 unit entries from 2024 and 2025. The early-fourth-edition sample contains 280 lists from eight tournaments, covering 3,932 unit entries.

For the faction-convergence analysis, I compare armies only against other armies from the same faction at the same tournament. That helps control for points level, local meta, event format, and other differences between tournaments.

I measure overlap using the exact unit and formation size. A Rabble Regiment and a Rabble Horde are different choices. Multiple copies also count, so two lists containing six Rabble Regiments each are treated as more similar than two lists that merely contain one Rabble Regiment. At the edition level, tournaments receive equal weight so one very large tournament does not overwhelm the rest of the sample.

More repetition inside 4e lists

The simpler question gives us the clearer descriptive result.

The standout number is the four-copy threshold. In late 3e, about one list in five contained at least four copies of the same unit at the same size. In early 4e, it is about one list in three, a 14.8 percentage-point increase. Several other repetition measures also move in the same direction.

Goblins are the clearest example

Goblins show exactly what players mean when they describe fourth-edition lists as more spammy. In late 3e, Rabble Hordes appeared in 15 of 23 Goblin lists. Across all 23 lists, they averaged 1.96 hordes per army. Among players who actually took them, the average was three.

Early 4e looks very different. Rabble Regiments appear in 10 of 11 Goblin lists. Those 11 lists contain 70 Rabble Regiments.

There is a second common piece too. All 11 Goblin lists contain Wiz, averaging two per army. Nine of those 11 Goblin lists contain at least four copies of one unit at the same size. Individual tournament lists reach 11 and even 14 Rabble Regiments. That is a pretty convincing example of increased repetition.

It is also useful because the Goblin lists are not literally copies of each other. Players still make different choices around that core. The faction’s average same-event overlap increased by about six percentage points, which is noticeable without suggesting everyone downloaded the same army.

Are players within a faction converging too?

This question needs a little more care. If only two Ratkin players attend a tournament and happen to bring similar lists, I do not think that tells us much about how many competitive Ratkin builds exist. My primary comparison therefore requires at least three players using the same faction at the same event. The reason I compare within a particular tournament is to control for local meta variations and points differences. Under that definition, the signal is fairly clear.

The first two changes are the strongest results. A typical same-faction comparison moves from about 21% shared composition in late 3e to 26% in early 4e. The closest other list from the same faction moves from about 29% overlap to 36%. That looks more like convergence around recognizable faction builds or common chassis than complete list duplication.

There is an important caveat here. If I include every faction with only two players at an event, average overlap is actually slightly lower in 4e: 24.9% in late 3e versus 23.4% in early 4e. So the evidence for convergence is specifically strongest where there are enough same-faction players to ask a useful question about build diversity.

Basileans show what a common chassis looks like

Basileans are a good counterexample to the idea that convergence means identical armies. There are seven Basilean lists in the early-4e sample. Six of seven contain Paladin Foot Knight Regiments, with an average of three regiments per Basilean list. Five of seven contain each of these:

Ogre Palace Guard Regiments, Gur Panther Troops, and War Priest.

That is a recognizable faction package. If you are preparing to play Basileans right now, seeing those pieces should not surprise you. And yet Basilean pairwise overlap has increased by only about one percentage point from late 3e. Players are agreeing on a core while still using the rest of their points differently. In other words, a game can have narrower list construction without every army becoming a clone. Whether that’s good or bad is probably an open question. 

The pattern is not universal

Several factions show substantial descriptive increases in repetition or same-faction similarity. Ratkin, Forces of Nature, Salamanders, Forces of the Abyss, Trident Realm, and Goblins are all worth watching. Other factions move the opposite direction. Ogres, Elves, Nightstalkers, Undead, Dwarfs, and Orcs do not show greater pairwise convergence in this sample. 

I don’t think we can make a broad claim that 4e has fewer builds within factions. But the data does support a narrower claim: some early-4e factions are settling around common cores more strongly than we saw late in third edition, while heavy unit repetition has become more common across the field.

So, are 4e lists less diverse?

At this point, I think the fairest answer is somewhat, and in specific ways. The strongest descriptive change is inside the lists themselves. Four-copy unit spam has risen from 19% of lists to 34%. The second signal appears within factions that have enough players at an event to compare properly. Their armies share about five additional percentage points of unit-and-size composition, and each player’s closest same-faction list is about seven points more similar.

Goblins make the concern easy to see. Ten of 11 armies contain Rabble Regiments, those players average seven copies each, and every Goblin army contains Wiz. Basileans show the more subtle version. Most players agree on Paladin Foot Knights and several supporting units, while still building meaningfully different armies around them. 

That is where I would put the early fourth-edition evidence today. There is a measurable move toward common faction cores and heavier repetition, but it is not happening everywhere. The next useful question is whether those common builds are simply popular or whether they are also disproportionately occupying the top tables. That’s going to take me awhile to pull together enough high-quality data to make conclusions. 

In the short term: if you feel like you’re seeing more concentration, more spam, and more similar lists, you’re not crazy. The effect is real, but it’s subtle. With the changes to list-building, we’re seeing more factions spam their best core unit as opposed to relying on multiple different units. For an edition that had a stated goal to constrain spam, this is having the opposite effect. That said, there’s a difference between constraining spam of flyers or hordes of berserkers or whatever else, and constraining spam of rank and file infantry units. For a ranks and flanks game, I suspect KoW is trying to encourage “good” spam and minimize “bad” spam–which is a much thornier question to work out. 

So yes, more spam, but not necessarily more bad spam. Make of that what you will.