I’ve been playing Dota Underlords for the past few weeks. I liked the original Dota Autochess mod, but stopped when playing it became routine. Underlords tries to keep the game fresh by removing a new set of pieces from play every 24 hours. This is an elegant system to create a new meta every day. Let’s examine how well that is working out.
Autochess and Underlords are games about making strategical choices, balancing buying and positioning pieces against buying levels and saving money to collect on the interest or spending heavily to keep up a win streak.
The reason that gameplay becomes stale in Autochess is due to the player base figuring out the optimal compositions and plays at every stage of the game. The player is either lucky on the start of the game (getting a good composition out of starting pieces) and tries to press their advantage from there. Or they look at what their opponents are building and base their composition on that information, trying to counter the played builds or going for pieces that the opponents are not playing, increasing their chances of getting higher tier units.
When starting to play Autochess, the gameplay patterns seem dynamic, driven by chance and player choice.
The problem is that, over time, playing the game becomes routine. If sufficiently played or analysed, optimal moves become apparent at every stage of the game. Think of this as poker, at a high level, every player knows their chance of winning the round based on their hand. This boils the game down to a statistical analysis (which I personally find neither interesting or rewarding). Players in this sort of game make less and less interesting choices the longer they play.
In essence, this high level framework is what most of these kind of strategic games are (at least the ones that forgo execution skills). The things I find appealing in these types of games are learning this framework, examining the strategic options and out-thinking my opponents.
What Underlords Jail does (in theory) is force a daily re-evaluation of the entire meta game.
The Jail works by temporarily removing a number of pieces from the pool every 24 hours. Every piece’s value is determined in relation to every other piece in the game. This shifts the mid to high level gameplay to pattern analysis – finding the daily optimal combinations of pieces and counter plays, based on what pieces are available.
In addition, the Jail allows the introduction of new pieces to the game by increasing the size of the Jail (normally increasing the number of pieces that can be played with in the game, decreases the chances of finding the individual pieces a player needs). This can also be a factor of keeping the game interesting over time.
Playing the game over the past few weeks I had the following observations:
- Analysing the daily Jailbirds becomes routine but keeps the first game of the day different
- This shouldn’t increase my time investment by much (1 min analysis on a 50 min game) but feels like it is increasing the barrier to entry
- Actually making an analysis makes me more exciting to play the game
- I play less in longer sessions (multiple games) and more frequently for a single game.
- This might be due to wanting to track my engagement over time.
I would say that the Jail keeps the daily meta game fresh for at least one game per day. This also leads to needing to learn the capabilities of all pieces in the game, not just the best ones, as players might need to rely on less optimal pieces to make their strategies work. Which in turn leads to an understanding of the game in a more global sense. While the real value of the Jail may lie in the expanded opportunity to introduce new pieces to the game, I for now enjoy the daily shifting meta game.