Designer Diary: TieBreaker – The Sudden Death of Ties

I don't actually hate ties. I find games that end in ties slightly unsatisfying (moreso if I'm one of the tied players, but still it's not a big deal). I don't much like games that have three or more tie-breaking criteria either, especially because it's hard to track that criteria during the game on the off chance that I'll be tied by the end of the game – and then the result seems arbitrary and not necessarily meaningful. Ideally, every game would have one tiebreaker, like money (if the winning condition were, for example, the most VPs). That's easy to track and might actually have an impact on how you play the game.

But when a game ends with a tie and there's NO tiebreaker at all, there's a collective "ugh" at the table because it seems like the publisher could have spent 15 minutes to see what a reasonably-tied game might look like and what sorts of criteria could be used to realistically break that tie. That's where the need for TieBreaker lies.

The Concept

The idea for TieBreaker has been percolating since before the publication of the original Start Player game in 2006. As I was working on Start Player, I realized that the game could also be used to break ties. While that seemed like a nice side benefit of Start Player, there's one thing I really didn't like: Start Player was super random. The result of 80% of the Start Player cards was decided before you revealed the cards. Things like "the tallest player" or "the player with the most buttons" were fun deterministic events...the real randomness was the card that was drawn. And that was (and still is) totally okay for determining who goes first, which is why Start Player is a great way to start your games.

But if you've played a long drawn-out game that ends in a tie, having the winner determined by similar random criteria seemed wrong – and in fact it made the rest of the game pretty much irrelevant...or even worse, made the tied players feel that they really hadn't determined a single winner because the random criteria had absolutely nothing to do with their ability to win.

In 2009, when I was designing and publishing Beer & Pretzels, I ran into a tiebreaker problem with the game. In Beer & Pretzels, players toss coasters onto the table for money, and the player with the most money wins. Based on my playtests, ties happened about 5% of the time. (Scoring would be anywhere from about $10 to $50, in $1 increments.) And due to the light dexterity nature of the game, there was no reasonable way to break ties. I toyed with the idea of having whoever gets the most money in the last round win, but all that did was punish the leader of the previous round(s). Instead, since I was using a snarky tone with the game anyway, I put the following line in the game rules:

Quote:

In case of a tie for any position, all tied players say, "This is lame."

Which actually added to the fun and irreverent atmosphere of the game. While I got away with it that time, I knew that the right solution would be a "universal tiebreaker" that could be used for any tied results for any game.

When I first created Start Player, the web comic Board 2 Pieces was in its first year. Making the decision to put a Board 2 Pieces comic on each card was a flight of insanity, as I ended up making almost 200 cards for the original "kinda collectible" version – but it also resulted in a lot of interesting things for the characters in the comics to do and say that they hadn't had a chance to in their twice-weekly BoardgameNews.com appearances.

Now, Board 2 Pieces has been around for more than five years (and is currently on OpinionatedGamers) with more than 500 comics published, and the tone of the comic has changed quite a bit from its 2006 origins. Making the decision to use the comic again was a little easier, but it still ended up being a giant time suck (with much of that time spent staring at the monitor trying to come up for something at least mildly amusing for each of the 50+ criteria on the cards).

The Plan

At this point late in 2010, I had finalized the plan: TieBreaker would have the same component mix as Start Player (but in a nice heavy Amigo-style box instead of the flimsy "collectible" box it first shipped with and the tuck box that Z-Man put it in). It would have at least 50 cards, rules, and a giant meeple. The base concept behind the game was that the criteria on the cards would be task-based so that all tied players would have an equal chance to win. The tasks themselves would not be random, but since they were in a card deck, the order of the tasks would be random. That seemed like a good compromise.

One of the biggest hurdles I now face is convincing gamers who have or have played Start Player that TieBreaker isn't the same game with different looking cards. Just the other day someone posted that very question on the Strategicon website (for the Gateway 2011 conference that took place in early September). The card on the back of the box does indicate that these cards are task-based and aren't random, but I'm sure I'll have to answer that question several times over the "Launch" season in Q4 2011.

Another Giant Meeple – This Time with a Purpose

As I was starting to develop the different tasks for the game, I wanted to figure out how to better integrate the giant meeple into the game. The giant Start Player meeple was an afterthought put in by Z-Man and da Vinci, and the idea behind it was that if you were chosen as the start player you held the meeple and the next game you couldn't be the start player. Meh.

With TieBreaker, the first person to grab the meeple after completing the task is the one who wins. Big difference. This is where my love of games like Jungle Speed, Fast Food and especially Geisteblitz comes in. All of those games require some speedy grabbing of an item based on some other criteria.

I decided to call the meeple "Tie" and to paint him with a conventional tie that had been "broken." Because of the visual aspect, I then decided to put him on the cover of the box. From a publishing point of view, the giant meeple is clearly the biggest hassle involved with the game as there aren't a lot of giant meeples out there. More on the production aspect of that later.

Guidelines for Tasks

Next, I started compiling all the tasks for the game, and categorizing them. As I did so, I realized that I needed to come up with some guidelines for the tasks:

1) Tasks couldn't rely primarily on pre-existing player knowledge.

2) Tasks had to use game components (cards, box, rules, meeple), a table and chair, and nothing else. (Out went a lot of "write this" tasks I had created.)

3) Tasks had to be things where it wasn't too easy to tie. However, if there was a tie, it was broken by the player who grabbed the meeple first.

4) Tasks had to be fairly clear, but when they weren't, a clever interpretation of the task could be completed more quickly than a more literal interpretation of the task. Arguing and discussion during and after the drawing of a TieBreaker card isn't a bad thing; unlike a typical Euro where players dissect the use of "may" vs. "must" in the rules, the nature of TieBreaker is that if you use your cleverosity in order to finish first by a reasonable interpretation of the task, you not only win, but the other tied player will be asking, "Why didn't I think of that?"

5) Tasks had to be somewhat fun to do and watch.

6) Tasks typically had to be able to be completed within five minutes. No one wants a 90-minute tiebreaker. Well, some people might, but I'm not sure I would want to game with them!

Categories for the Tasks

I started listing the tasks in a spreadsheet, and as I was doing so, I came up with different categories for the tasks:

Quick: This is the most common of the tasks, where you read the card and react to it as quickly as possible, then grab the meeple. Tasks in the "quick" category include "Drink anything", "Build a two-story house of cards", "Tear something into three pieces" and "Balance something on your ear".

Say: These are cards requiring you to say something before your opponent(s), such as "Say the last seven letters of the alphabet." or "Say five nice things about the other tied player(s)."

Challenges: These are feats of skill, such as "Fling a single TieBreaker card the furthest."

Longest: These are endurance competitions; think Survivor's last challenge, but usually shorter and less painful, such as "Stick a TieBreaker card to your forehead the longest."

The original goal was to have a nice even spread of categories, and about the same number of cards for each category. However, coming up with good categories was tough, and they became heavily weighted towards the "Quick" category.

Once I had about 25 criteria in a list, I started working on the card design in Illustrator. In order to differentiate the cards from Start Player, I went with a landscape view instead of portrait. I put the criteria on the right, and the Board 2 Pieces comic on the left. The comics were slow going, but somehow I was able to slog through all 25. Then it was time for playtesting.

Playtesting

My normal rule for playtesting is that I don't playtest anything until I've done several self tests of a game. This ensures that playtesters don't hit really nasty problems with the game, and also, my thought is that if I'm excited about a game, I can play it several times all by myself without getting bored.

With TieBreaker, that proved to be a challenge, though I did test it by performing each of the tasks myself after drawing a card. The great thing about that first test session is that right away I knew there were some cards that I didn't like, while others were actually fun to do. And a few turned out to be either impossible or way too difficult, leaving me with about twenty cards for a real playtest.

Of course, I didn't wait for games to tie to test TieBreaker. Instead, I simply chose two people and flipped over a card for them, asking them to compete. I gauged their reactions to the criteria and noted how long it took to perform the task and how satisfied they were with winning/losing. It turned out the "quick" category was the most popular and satisfying, even to those people who weren't typically good at such games.

Refining

With lots of great feedback from playtesting, I started refining the existing set of cards and added new ones. The act of playtesting actually generated a lot more ideas for card criteria, and the playtesters had a few good ideas as well, so it wasn't long before I had 70 criteria in place, which was my goal. I knew I would have only 60 cards in the deck, but it was better to have more than less; I could always strip out a few.

Again, putting the cards together with the comics was a struggle. I'm used to writing two Board 2 Pieces comics a week, which suits my creative ability to come up with two amusing (of course, this is subjective) comics each week. I've had to cram and do more at a time before I go on vacation, but doing 30+ in a short period of time was taxing.

Rules

Once the comics were done, it was time to redo the rules. I typically write my rules in Word/Pages (the latter more recently), then close to production I redo them in InDesign. but I wanted the rules for TieBreaker to be super concise, and if possible, to fit into a Board 2 Pieces comic strip, as I had done for Start Player. I was pleased with my Start Player rules, though they were a bit long. They captured the essence of the game, and everyone likes reading comics as opposed to a few pages of traditional rules. Once I determined that I could fit the rules in only two panels, I was thrilled! TieBreaker was indeed incredibly simple!

Even better, I was able to include what I called the "naysayer" rule: Sometimes a player just doesn't want to do a task. Instead of setting up a potentially endless loop where each player declines a card they don't like either because they don't want to do the task or because they think they won't be good at it, I came up with a better solution: The first player to say "I'm not doing this" loses and the other player wins! This leads to temporary stalemates that end in amusing situations for everyone involved.

The Box

The cost of doing an Amigo-style box (bottom and lid) vs. a standard "tuck box" is fairly significant – more than twice as expensive. But the flimsy Start Player boxes (both the one that I produced as well as the ones from da Vinci and Z-man) didn't cut it, especially if this is something that I wanted players to carry. A ratty box might mean a game is well loved, but I prefer a tough box that can take a beating and still look good. (I was thrilled with Rio Grande's current Tichu boxes compared to the previous tuck box version, so much that I bought a ridiculous number of copies; if Jay ever goes back to the tuck box, I'll have my own personal supply of boxes.)

Because of the nature of TieBreaker, the cards are going to last a long, long time through regular play, but the box is going to see a lot of action. This box will be able to handle that action.

Because I had extra space along the outside edges of the box bottom (which are inside the box when it is closed), I added a set of Board 2 Pieces characters that were representative of our game group along those edges. (Most of them playtested the game at one time or another, and some of them even provided a few tasks.) This was one of those last minute production changes I did right before submission.

Start Player had been for ages 8+, and I figured TieBreaker would be the same. However, the ridiculous rules imposed by the U.S. government in the Consumer Product Safety Improvement Act (CPSIA) of 2008 required that any products imported must have additional testing if they are sold to anyone under 13, so I changed the age range to 13+ (but not everywhere).

Cards

The standard size cards I ended up using were a bit smaller than I had originally envisioned, but going to a larger size was a significant cost increase. And since the criteria (the important part) of the card was still in huge text (24 point) at the "reduced" size, I could live with the slightly smaller cards. Of course, that meant that I had to reduce all 60 cards to a slightly smaller size. Since I was changing the cards, I also added a line along the bottom of each card that contained copyright info and more importantly, a number corresponding to the number of the card.

During the back-and-forth with the production company regarding components, they informed me that I would have space for only 55 cards, not 60. It took a long time to figure out which cards to remove from the deck, but I did it (with the help of playtesters).

Tie, the TieBreaker meeple

Next, it was on to the meeple. I provided the manufacturer with several different 3D views of Tie, the giant orange meeple, as well as the flat artwork for the "face" of the meeple:

As I was doing this, I realized I had a problem. The box size I had chosen was only so thick, so it limited the thickness of the meeple. I wanted a really large, easy-to-grab meeple. Keeping the proportions of the meeple consistent with a "real" meeple would have resulted in a meeple that was not significantly larger than a standard "large" Carcassonne meeple...definitely not what I had envisioned. So Tie went through a slimming down process in which his thickness was reduced so he'd fit in the box well, but the overall size was still what I had envisioned.

I've been using Panda Manufacturing (out of Vancouver) for several years now, and every once in a while I'll get to see something ahead of time, like the pictures of Tie's first production run sample (shown at right). There's quite a bit of time between initial submission and when you see a game in final form, so these sorts of images are totally appreciated!

One thing I didn't notice, however, was that Tie's tie is broken with a white line. I'm not sure why I didn't notice it before, but I didn't. It's not entirely bad, but it doesn't match the image on the front of the box, which is a bummer.

Sample Copies and Minor Issues

As I write this, the samples have just arrived...and they look great! However, there are a few minor issues, some caused by me, others in the manufacturing process.

1) Tie's white tie, as mentioned above.

2) Tie isn't as thick as I wanted him to be. I could have made him thicker and still fit him in the box. However, he still stands up easily.

3) There are 55 cards in the box, but the box says 60 cards. Whoops. I somehow totally missed that. Hopefully most peeps who buy the game don't notice. I made a similar mistake with the second printing of Ultimate Werewolf: The rulebook said there were cards that actually had been removed for that printing in order to include the Classic Movie Monsters expansion. I got emails from only about 0.002% of the people who got the game, so if that holds true, I should get only a few people who see the discrepancy on TieBreaker.

4) The box lid and bottom is too tight. It's not terrible, but it's definitely tighter than I would like.

5) The cards fit too tightly in the box. They still come out, but there's no wiggle room.

6) I missed the player age range on the sides of the boxes. The bottom says 13+, but the sides say 8+. Arrrrgh.

Those are all minor issues, and won't have a negative impact on 99% of the people who play TieBreaker, but they still bug me.

Final shipment of games

Now I'm eagerly awaiting the full cases of games (which I won't get until just before Spiel in October). For me, this is always the most stressful time of year. A few years ago, Beer & Pretzels got a random spot check by Customs and I didn't get the games until a few days before I left for Germany. That's all very, very bad. I have more cushion this year, but it's still closer than I'd like...

Ted Alspach
Bézier Games