Back to the Woodland: A Retrospective on the Marauders Era of Root

Welcome back fellow woodland warriors! It's been a while since my initial blog post trying to understand the complex faction interactions in Leder Games' Root. As the community prepares for the release of the new Homelands expansion, I wanted to take a more look back at the Marauders era with more complete data. We'll also look at some extra investigations on the impact of seating order, map, and deck choice on each faction's win rate.

Since we last spoke, the Marauders expansion has been added to Root Digital. Due to the phenomenal work of the Root Digital League team we now have over 9,000 recorded games of Root across all of the factions, maps, and decks available (until Homelands releases at least).

Overall Win Rates

The overall win rates of each faction are reasonably consistent with expectations and community understanding. Last time, we had a small sample size for the two Marauders factions, and most of the data came from the 2021 Winter Tournament. These were some of the first games ever played with the factions in their final forms, so there was a lot of noise in the results. This time, we have enough recorded games to be confident in the overall win rate* of each faction to within 2%:

Duchy	30.55 %
Eyrie	29.11 %
Rats	27.88 %
Badgers	27.38 %
Alliance	27.12 %
Vagabond	23.78 %
Riverfolk	22.82 %
Marquise	21.26 %
Crows	21.04 %
Vagabond2**	20.54 %
Lizard	17.53 %

* Coalition victories (the perennial problem child of competitive Root) have been removed.

** Vagabond2 is the second Vagabond drafted, so first to act.

Match Ups

Although a faction may simply be "stronger" than another, some factions have a particularly hard or easy time winning when other factions are in the mix. This can be expressed as a simple number for each faction vs each other faction, but it hides some of the complexity in a 4 player game. For instance, Crows and Moles both win less often when they're in games together, Vagabond and Cats both win more often when in games together, but when the Woodland Alliance and Moles play against each other, Alliance wins more often and Moles win less often.

Synergy

To describe these relationships, I like to make a "synergy" plot, as below:

This plot shows 3 values for each faction:

How much do each of them win on average (the solid dot)?
How much do each of them win in games that do not include the other (the small +)?
How much do each of them win in games that do include the other (the large X, surrounded by a grey oval showing our confidence in this value)?

In this case, we see that the matchup is good for Badgers and bad for Vagabond. This makes sense and can be explained with a few points:

Badgers thrive on a clear map that they can traverse easily. This is more likely to be the case against a Vagabond.
The Vagabond thrives when they are able to tax the actions of others who do not benefit from policing them. Badgers in many ways get to police the Vagabond "for free" with their retinue, as they don't have to invest specifically in actions to police that don't contribute to their points engine.
Badgers often don't put lots of cardboard onto the map for the vagabond to score with.

Results

With this construct, we can get an in-depth view of how each faction interacts with each other faction:

Seating

In addition to raw faction strength, win rates are also impacted by a player's seating. Going first is a meaningful advantage, you get a head start on the race to 30 points. Overall, each seat's win rate is as follows:

Seat 1:	29.07 %
Seat 2:	25.31 %
Seat 3:	22.74 %
Seat 4:	22.99 %

We can also look at how this breaks down by faction:

There are a number of factors at play here, but ultimately match ups are relatively balanced. It's worth noting that we have a much higher sample size for strong factions (Duchy, Eyrie, Badgers, Rats) in the later seats because they get picked first in ADSET draft much more often. The same is true in reverse for the weaker factions. A few points stand out:

The Woodland Alliance benefits disproportionately from going first. This is likely because they benefit from a clear map and the ability to get in the way of other factions before they have a turn.
Some factions win more often in seat 4 than 3 (Crows, Vagabond, Badgers). These are all factions that can burst but are easily shut down. The win rate difference could be because they win better from behind, when they are perceived as less of a threat compared to others earlier in turn order.
Riverfolk have the flattest win rate across seats. There are likely competing factors between wanting to go first to take a lead, but benefiting more than others do from going later as there are more chances they get to start turn 1 with additional funds.
The most winning factions from seat 4 have a similar win rate to the least winning factions from seat 1. This indicates that with a more cutthroat draft where the "optimal" choices are made at each point, the overall faction win rates could be more consistent than they are now. The variance (and the reason there are still 961 games where Moles made it to seat 1 compared to 1629 in seat 4) is likely due to the fact that League games aren't always the most competitive version of Root. Sometimes people may prioritise playing their preferred faction or avoid playing the same one multiple games in a row over making the "best" choice.

Maps and Decks

In addition to the other factions at the table, win rates can also be impacted by the map and what deck has been chosen. We can investigate the impact of these factors relative to their overall win rate of each faction:

First it's worth noting that we have a pretty meagre sample size for games with the base deck (360 compared to 9335 with Exiles and Partisans, a fair chunk of which are probably due to an accident when making the game). This makes sense, many experienced players prefer Exiles and Partisans due to the wider variety of good crafting options and the lack of the volatile "Favour of the X" cards. Other than that, there are a number of interesting takeaways:

Many traditionally strong factions (Moles, Eyrie, Alliance) have a lower win rate on the Mountain map in particular. This is likely due to the paths. Although Moles and Eyrie can take and hold the Pass, these factions often cannot afford to spend extra cards from their hand to open the paths for points, and can be hampered by the limited manoeuvrability in the early game when they are closed.
Conversely, traditionally weak factions in the Cats and Crows are not concerned with early game mobility and benefit significantly from the extra points and their ability to open almost any path on the map. Their win rate on the Mountain map is noticeably higher than any other map.
The Woodland Alliance clearly prefers the Autumn map due to the connected nature of the clearings (and potentially because some of the games use fixed clearing suits as opposed to random).
The choke points of the Winter map are an issue for Badgers, but conversely the Riverfolk are able to bridge the two halves of the map easily as swimmers.
Surprisingly, despite being the most structurally distinct map, the Lake map appears to not have a major impact on each faction's win rate, except for a small increase for Rats and a small decrease for Cats.
Cats have significant issues with the Base deck, likely due to the higher and more rigid cost of craftables.
Crows and Riverfolk have higher win rates with the Base deck, probably because of the inverse as they are some of the most flexible and able factions when it comes to crafting.

Balance

I wanted to make a short note here on balance in Root. While there are clearly some factions that could stand to be a little stronger or weaker, the overall balance across 10 wildly different factions, even in the hands of the more experienced RDL community, is a real achievement. Especially with a more "optimal" draft strategy, it's reasonably unlikely that players will be at too large an advantage/disadvantage to start the game as a result of the random draft. Praise is due to the development team and play testers for an ultimately very balanced game.

I also have high hopes for the impact of the upcoming Homelands expansion, even if it will be a while before it is implemented on Root Digital and I can get data from it. By their very nature, militant factions are going to have more control over the outcome of the game due to their higher reach and ability to interact. With Homelands though, the Bats and Knaves are insurgent factions that have been given a larger ability to impact the board (like to the Woodland Alliance) and police, (like the Vagabond). Similarly, the Frogs as a militant faction have the ability to police but have to strike a balance between doing so and scoring themselves (like Cats). I am also excited for two new maps and another deck that promises to provide a real choice (unlike the "choice" between Base/E&P).

Conclusion

I hope you have found some interesting take-aways from this data and my analysis. Once again, many thanks are due to the fine work done by the folks behind the Root Digital League, especially to Mouseman for exporting the records from the database for me. If you want to see more of this type of thing in the future, make sure to keep supporting their work (and maybe play a few more RDL games with the Base deck to improve the sample size :P ). Next time I intend to investigate how each individual Vagabond fares against other factions, now that we have a reasonable sample size for each.

If you have any thoughts or feedback, have noticed anything I didn't touch on, or have other ideas as to why certain numbers are the way they are, please let me know either here, on Bluesky/Twitter (@ArcKayNine), or in the Woodland War Machine Discord (@Rk9). If you'd like to support more of this kind of in depth analysis into Root or other games I enjoy, feel free to buy me some Root Tea over at Ko-Fi.

Until next time,

May your opponents let you pick Moles in seat 1.