The Fall of the House of Tank Drive

I’m pretty sure this topic has been brought up a million times, but I’m going to bring it up again.

Tank drive is dying.

What used to be the purest form of FRC robot transportation is being replaced by robots driving sideways at incredible speeds. It’s almost poetic, actually.
Let’s take Einstein in the past 2 championships. From what I remember in 2019, I saw about 5-7 total swerve teams across all 4 Einstein divisions. This year, in 2022, there were only 2 tank drives that I could see in total across all 24 teams.
Now, what does this mean for regular driving devotees like my team? Honestly, I’m not sure. All the signs I saw this year point to that swerve will start to overshadow tank AND mecanum drivetrains as the big way for robots to get around the carpet. However, I think that the kitbot chassis will remain as tank for simplicity with rookie teams and veterans alike.
In the near future, I honestly predict that tank drive will be as rare as when teams (ex: 1717, 16,) started to prototype swerve drive back in the early 2010s.

Something must be done to save this dying drivetrain. Question is, what?


[Insert comment about obvious CD and championship bias/correlation]


I believe this is what will save a “dying drivetrain”

Seriously, though, speaking as a mentor on a team who used swerve a couple years in a row, there are games that will exist (hopefully) which will be disadvantageous for swerve, and games where swerve will have not nearly as much as an impact.

I bet teams who decide, no- matter-what, that they are swerving next year, may open themselves up to more heart ache than intended by the GDC.


But does something really need to be done?


I got an idea. Send every new team a free one, get them hooked. Swerve industry won’t know what hit 'em.


As @vargoose84 mentioned, any field that required teams to go over many bumps will be the only way. I think it will take a few years for this correction to take place and FIRST to adapt. Until then tank can’t compare to swerves, especially the simplicity and reliability of the MK4’s or any iteration of them.


I really hope we don’t see this kind of game (2016) ever again.

Why? I want the vanilla KOP chassis to be a viable dt choice. I believe teams should focus on the stuff on top of the DT and this game really killed that philosophy.

There were many teams (good teams) that struggled in 2016 because of failed custom drivetrains. That feels very uninspiring to me.

Just my 2 cents

The other side is that if the vanilla KOP chassis can play the game unmodified, the SDS MK4i can too, but better.


I think this could be the greatest idea in FRC right behind the introduction of the CIM. I think while we are holding onto the past, let’s start sending out a PDP’s worth of CIM motors. We can’t let the CIM motor die. Big Brushless won’t know what hit 'em.


You can start by taking ours. We’ve got mountains of the things.


Contrary to popular belief, swerve was possible in Stronghold. On 3419 we had 8” pneumatic tire swerve AND we fit under the low bar:


Yes, it was “possible”. But it offered fewer advantages and far more difficulty.


This is something that I feel was very cool about 2016. The drive train was a real mechanism, that could score points and was more relevant to your teams’ success than just whether you can move around on the field. Similar to other years, teams could also ignore this scoring section of the game by building a shooter that could go under the low bar and have a much easier time playing the game.

I understand the logistical issues with incentivizing custom drive changes, but this is already happening with the ridiculous number of teams putting together cots swerve lately.


The vanilla KOP chassis could cross the moat and rough terrain in almost any configuration. With a little thought about dimensions it could cross the ramparts as well. Depending on what was on top of it, it could cross the low bar, chival de fries, portcullis, drawbridge, and sally port (and the drawbridge and sally port could also be scored with just a drivetrain provided a good enough driver, some extra time, and a ref that was carefully watching your bumpers as you spun). The only defense that “required” a custom drivetrain was the rock wall, although some other defenses could be crossed at a wider range of velocities and conditions with a custom solution.

The biggest point, however, is that drivetrains themselves were earning points in that game. Focusing on a drivetrain was also focusing on a core gameplay mechanism.


Tank drive on Einstein is the new mecanum on Einstein.


One word: terrain.
As in, the soldiers Charged Up the hostile terrain. Tank drive will be back.


I know you’re making a joke here, but both the brushless revolution and swerve revolution pose the same issue with keeping FRC inspiring for teams that struggle financially (inb4 “just switch to VEX or FTC”, pretend for sake of conversation that FRC is the established program of choice for a given low-budget team for reasons independent of finances). Both swerve and brushless motors provide HUGE advantages in terms of robot capability, but they are both VERY expensive to implement. The number of teams we had ask us about the cost of doing swerve only to immediately become crestfallen and say “that’s more than our past two robots combined” is heartbreaking.

Swerve has always been expensive even before it was available as COTS (in-house innovation is expensive), but it used to be inaccessible as well due to the need for technical resources to figure it out yourself.

Brushless motors give teams more weight and space to work with and can allow for more complex mechanisms because of it, but the prices of the Falcons and/or SparkMAXes are out of reach for so many teams.

FURTHERMORE, even though there has always been an advantage to teams with more resources, vanilla KOP drivetrains used to be capable of playing shut-down defense and that made them capable of contributing to a match outcome. Swerve is hard to defend with a tank drive, but it’s do-able and usually on the back end of the draft, that was all you had to choose from, so teams with a kitbot drivetrain and no top functions had a chance there if they had a good driver (and preferably a 50lb weight or a box of our inventory of old CIMs). Now, we’ve seen VERY CLEARLY this year how big of an advantage swerve has when defending swerve, which is great for teams that have some money but lack technical/fabrication resources (which is a good chunk of third robots at the regional/district level), but is a big middle finger to teams that are struggling financially regardless of technical resources.

So there used to be two barriers to swerve (price and technical resources), and we have now removed one…but it is the same barrier the KOP drivebase is intended to remove (technical resources), so on-field success is now much more of a money game than it used to be. I think we can all agree that on-field success is not the only way to inspire students, but it’s a lot harder to inspire students if you have almost zero chance of on-field success from the get-go.


You folks had a great robot that year. We enjoyed playing with you at Tech Valley that season.

1 Like

THIS. This is the biggest crux of the problem. The increasing advantage of money in FRC. As more and more things can simply be bought.

Swerve is the worst of them all because it is so expensive and gives such a large advantage. Its prevalence has also forced many teams (mine included, to go to it).

Other COTS stuff like Greyt products, WCP MCC stuff, Everybot kits, etc. all raises up the floor of FRC. They are not too expensive, provide a starting point for teams, and can all (usually) be done better with enough experience and design effort.

The problem with Swerve is the fact that for 95% of teams, the best possible option is to buy a super expensive kit to only meet the competitive standard.


What do you mean by this type of game? I honestly don’t think this evolution in the drivetrain meta is primarily caused by the game design. It’s a conflation of factors, including the increased number of PDP slots, availability of reliable, top-tier COTS modules, and open source software availability. On top of this, this year had more competitive parity at champs between the mid and top tier than I’ve ever seen before. I think this is just a continuation of a trend that has been going on for decades: teams get better at building robots over time. It will only continuously get more difficult to compete at the top end of this program.

Missed context whoops.


I’m pretty sure they were referring to stronghold and how the kit chassis wasn’t great that year.