Login Register
RSS Feed Twitter Facebook YouTube

DEV JAM #4 – Simmons on Party Matchmaking



The Fast and Formidable update has been live for a couple of weeks now and whilst we still have some balancing to do to maximise the range/diversity of viable Robots possible in the new meta, we’re happy with how the initial launch went.

F&F was developed to lay the foundation for better matchmaking (as explained in the first Dev Jam here). Good matchmaking is the foundation of all good Arena games and Robocraft has frankly sucked in this area for a long time; so we’re focusing on getting our act together in preparation to be able to launch a solid competitive Clan League and exit Early Access in 2017.

Although there is still more work to do on reducing subdivision by making improvements to Battle Arena and the other Game Mode options, we’re are not planning any more massive changes to the core building mechanics like F&F going forwards. So the improvements and balance changes made to the game going forwards will be less likely to render Robots in your Garage non-viable in future.


Now we have all Robots available to be match-made together – as both low and high CPU Robots are more viable to be pit against each other in battle, we can now focus on matchmaking Parties properly, and once that’s done we can then focus on ELO.

Our next update is currently scheduled for Thursday 1st December (this may change so stay tuned to the website for updates during this week) and will rework the way we match-make Parties.


I often say there are two sides to matchmaking, ‘Real matchmaking’ and ‘Perceived matchmaking’. ‘Real’ is when you actually effectively matchmake the players fairly.

With Thursday’s update ‘real’ matchmaking of Parties will be present in all team game modes. This is how it will work:

“The total number of players in two opposing teams that are in a Party (not necessarily the same Party) will not differ by more than +/- 1”

Example 1: Team 1 has 5 randoms, Team 2 has 3 randoms and 1 Party of 2 (randoms are considered as being a Party of 1).
Example 2: Team 1 has a Party of 5, Team 2 has a Party of 4 plus 1 Random.
Example 3: Team 1 has a Party of 5, Team 2 has a Party of 3 and a Party of 2.

Now, this isn’t 100% perfect, as obviously it would be better to have the two teams exactly equal at all times, but this will ensure that the balance is a lot fairer whist not pushing queue times too high and still allowing us scope to match-make players and Parties by ELO in the next stage of development.


In our case this is all about being transparent about the matchmaking. So, we’re going to display the players that are in each Party in the loading and countdown screens at the start of battle so you can see it’s being fairly matched up. This is something we had to remove in the past due to simply not being able to match-make parties properly (due to too much subdivision) and is something we can put in now only because of the changes made in F&F.


We’ve been listening and reading your feedback on the forums, Steam discussions, FB, Twitter, Reddit, and i’ve even been having some voice comm chats with some of the Clans (will do more of that too). The feedback has been great and is really helping us to improve the balance post-F&F. With that in mind we’ll be making some more changes for Thursdays launch. The main things we’re trying to address are:

  1. Reduce the friction and remove the frustrating ‘snagging’ that occurs when Robots ‘scrap’ along the side of parts of the environment
  2. Improve the Energy to CPU curve to try to ensure all CPUs are viable for maximum diversity
  3. Improve the effectiveness of mid-range Plasma and SMG so all weapons are viable for different purposes
  4. Increase the cooldown of the EMP Module
  5. We’ll be looking at Drones – having played this style a lot recently I feel a well-built Drone is overly dominant in BA at the moment and this needs to be addressed as it ultimately makes playing other styles of Robot less competitive

Thanks for reading!

Mark Simmons, Game Director and CEO, Freejam