Robo Rally 2016 board game review

This review assumes that you have played a previous edition of Robo Rally.  The 2016 edition was designed by Richard Garfield, the creator of the original Robo Rally, to improve a bunch of things about the gameplay.  It’s published by Avalon Hill.  You can read Richard Garfield’s Design Notes which explain his reasons for the changes he made.



The biggest change is the programming deck and how damage works.  Rather than have one large pile of cards, each player has an individual program deck.  This means that if you draw a handful of turn cards, you know that you’ve got some useful move cards coming up in your next hand.  Damage no longer reduces the number of cards you draw each turn, instead it adds useless “Spam” cards to your deck.  You can get rid of the useless cards by programming them, and you will draw the top card from the deck for that register.  If the drawn card is also a Spam then you can get rid of that too. Your chances of being able to remove multiple spam a register increase as you hold more spam.  You can no longer destroy a robot by shooting it, it instead gets bogged down with spam cards.  A large advantage of this system is that the dealer no longer has to deal different numbers of cards to each player.decks

rebootNo more archive markers.  There is a Reboot Token on each board.  Destroyed robots will move to that board’s reboot token where they will begin their next turn.  If multiple robots reboot, each is shunted off the token in the direction of the arrow to make room for the new rebooter.  They also receive 2 spam cards.  This mechanic is simpler to deal with than the archive marker tokens were, but intentional rebooting to a set space can give an advantage.  I’m not sure about this change, but it’s easy enough to use the old rule in the new edition rather than the reboot token.

antennaePriority antennae.  Rather than having numbers on each card to resolve move order, there is a plastic model called the priority antennae.  It normally sits on the starting board, and the closest robot goes first.  Ties are resolved by a clockwise sweep from the antennae.  This change means that moving further doesn’t automatically mean you move first like it did in previous editions.  I like the way this makes robot interactions less easy to predict.

Repair and upgrade squares no longer exist, the way to remove Spam is by programming it into a register.   Powering Down has also been removed.

An Energy Space

Shopping!  In the place of Upgrade Spaces there are Energy Spaces.  They have an energy cube placed on them at the start of the game.  End a register on them to take the cube, or end a 5th register on an empty Energy Space to get a cube from the bank.  Cubes are a resource used to buy upgrade cards.  Each player begins with 5 cubes, and upgrades cost 0 – 5 energy to purchase.  The Upgrade Phase happens each turn before programming.  New cards are turned up and each player in priority order chooses whether they will buy one.  I’ve never been a big fan of upgrade cards, and this mechanic makes it easier to gain multiple upgrade cards than the previous editions.  In Richard Garfield’s design notes he didn’t have a shop.  Instead each player is dealt 3 upgrade cards at the beginning of the game and given 2 energy.  They have to pay to equip a card from their hand, or pay 2 energy to draw another card from the top of the deck.  This sounds far more restrained!  The cards themselves seem quite good.  There is no longer the dreaded Mechanical Arm.  Some upgrade cards will add a useful card to the programming deck.  There are still Double Barrelled Laser, Pressor Beam and other old favourites, and the new cards seem to be well-thought-out.

8 of the 40 Upgrade Cards

powerupThere are two new cards in the program deck – Power Up and Again.  Power Up will gain you an energy cube.  Again will repeat the card in the previous register.  The rules say that you can’t program it in the first register, but Richard Garfield says you can and it just does nothing.  I’ve been going with Garfield’s version of this rule.  These work well with the new design, which has programming decks of only 20 cards.  There is only one Move 3, but if you get the Move 3 and an Again card, you can still move 6 squares in 2 registers.

2016 Edition below, 2005 above

Boards are now 10×10 instead of 12×12.  This makes multi-board coarses more enticing.  Games go much faster on a single board.

There are now only 6 robots instead of 8.  I’ve played games with 8 and they were all terrible, so removing the temptation is a good thing in my opinion!  Bad news if you were a particular fan of Twitch or Trundle Bot though.

Minor changes: There is no longer a limit to the number of lives.  Players now choose a spawn point on the starting board instead of going by number.  The timer now starts after the first player has finished programming rather than second-to-last.  There are 6 flags included instead of 8.

What’s in the box – production value:

There are some good and bad points to the game’s physical quality.  Let’s start with…

The good:

  • bots
    2016 Edition in front of the 2005 robots

    Painted robot models – very nice!  They’re all painted to match the colour of the player mats and the programming decks.  Not a high-quality paint job, but they certainly look a lot nicer than the solid silver colour of the previous edition.

  • The new graphic design of the boards.  In the 2005 edition I found that new players would often not notice walls and would try to drive through them.  This edition has made walls wider and given elements more distinct colours.  The design is very clean and uncluttered.
  • checkpoint2Checkpoint tokens exist!  A failing of the 2005 version was in not having a way to record which checkpoints you had visited.  As archive markers could move to other flags and repair squares, it was not a reliable method.
  • Good solid boards and reboot tokens.

The bad:

Misprint on board 4A
  • Poor error checking.  For example three of the boards have misprints and some cards disagree with the rulebook.
  • Tiny programming cards.  They are 41mm x 57mm, which is not a standard size.  I found that sleeving them was very necessary for handling, but the closest size sleeve I could find was Mini American (41 x 63).  This makes the cards too tall to fit in the box insert.  (The Upgrades Deck is Standard American, which is a good size and easily sleeved)

    New programming cards below compared with 2005 edition above.
  • scrambler
    The card is different to the rulebook, and is wrong.

    Sloppy wording and lack of clarity on some cards and rules.  For example the Upgrade Card “Scrambler” says “If you attack a robot, that player replaces the card in their next register with the top card of their deck, unless it is the final register.”  There are several issues with this wording.  It says “attack” rather than “shoot”, so does pushing with ramming gear count as an attack? (No, only shooting with the laser counts) Does this effect happen as well as damage or instead? (As well as damage) The phrase “unless it is the final register” is ambiguous.  “It” refers to the register you are currently performing, not to the card you are replacing – if Register 4 is being performed, then you can indeed replace the card in Register 5.  That’s a lot of rules clarifications required for a single card! Answers from the Rules clarification thread on BoardGameGeek.

  • No Cheat Sheets.  The 2005 edition had Factory Floor Guides and each Player Mat listed the activation order for board elements.  In this edition the back of the rulebook has a “Summary of a Round” list, with extremely basic information.  A list of board elements and activation order would have been far more useful here.  It’s easy enough to print some off, but having them included would have been nicer.
  • The checkpoint tokens are very thin.  A better way to keep track is with a dice for each player, set to the flag they are heading to.

    Better than checkpoint tokens


The new mechanics make some very good changes to the game – the individual programming decks and damage changes particularly.  The painted models are also great to have.

The only thing I’m not convinced on are the new board layouts –  the ones I have played so far seemed a little underwhelming compared to the original boards.  They are not actively bad, just a little different to play on.

I think the ideal scenario is to have both this edition and one of the older editions. That way you can benefit from the new cards and damage systems, but have the old boards to play on as well!


Having played some more, a major problem with the game is an insufficient number of Spam cards (38).  Once the spam deck is exhausted, players are supposed to pick from the 3 other damage card types, which are normally only handed out by specific option cards.  2 of these, Virus and Trojan Horse will hand out more spam cards when activated, so it’s very possible to get into a situation where so many of these are in play that it’s impossible to clear them.
Richard Garfield suggested having at least 60, but probably 120 Spam cards.
The best fix (without buying a second copy of the game) would be to buy 250 blank playing cards and create your own programming and damage decks.

Given that the individual decks are the main mechanical pro of this edition, I no longer feel happy to recommend this edition as an improvement.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s