It's been an exhausting weekend.

And as far as Prolix is concerned, it's been a productive one, too. I came in on Thursday night, exhilirated at its awesome showing at Protospiel. But at the same time, I was wary that a different group could always break the game.

Four playtests later, progress has been made.

Friday:

Playtest #1: Four players, not including me. One player, M, played at a deliberate pace. Not slow by any means, but he would take a minute or two to figure out an optimal play. The game was peppy, and finished in about 45 minutes. One of M's turns seems to slow down, but just before I propose the "unanimously agree to move the letters and start to play the turn again," M turns around with a brilliant 19-pointer.

Everybody seemed to enjoy the game. One fellow, J, ambled over in the middle of the game and was so intrigued that he asked to play it again. A friend of mine, G, also came by at this point and asked to play.

Playtest #2: Three players, including me. And suddenly: crash. J is slower than M was, and immediately realizes that interrupts aren't worthwhile enough. He can get bigger points on his turn. So the game takes more like an hour. Eeeyuk.

We discuss a postmortem, and I go up to my hotel room in a miserable mood. So much for "done;" the lockup problem is alive and well.

J is a very nice and intelligent fellow, and he designs variants for a published game, so he's not a stranger to the game design process. Still, all his suggestions for rules are much more high-reaching than I'm comfortable with. At this point, I don't want to change much more than the scoresheet and the tile mix, but J wants to see interrupted players not get a second turn. He also likes increasing penalties for interrupts, which I'd rather not go back to for reasons of simplicity. Still... does he have a point?

G thought the game was okay, but not great. J didn't seem to enjoy it... or so I thought.

Saturday:

I didn't play Prolix at all. I wonder if the 3-player game causes the downtime, considering that there's one less player to interrupt.

Sunday:

Playtest #3: Another game, with N, L, M again, and J again. Yeah, it turns out J wants to give it another whirl.

N is excited to play the game; he hasn't tried it in two years, with the old chip-based Interrupt system. I figure that between M and J, I can tell whether the lockup is specifically a 3-player problem.

It's not. The game lasts 75 minutes. The letters are incredibly difficult; lots of rare chips out there. I'm watching, in excruciating misery. I'm positive no one is having a good time, as M and J study the board.

One of J's turns lasts 20 minutes. At one point, we actually invoke the unanimous clearing rule for the first time in the game's history, but in practice, it feels like putting a band-aid on a broken leg.

L is not leaning forward. He's sitting back in his chair, seemingly disengaged from the game. He frequently doesn't have enough points to interrupt.

Halfway through the game, a bunch of nasty tiles disappear, and suddenly the pace of the game picks up. L and N start to lean forward. They start to interrupt. J loses a turn, then N loses a turn. Things are picking up.

J almost loses a second turn to interrupts, which would have been devastating, as he had only one interrupt and his turn was last in the game. He and L blurt at a word at the same time. L's interrupting word is a little worse in the tiebreaker, and J's word stays in.

Everyone exhales, and at first, I thought it was a sigh of relief that this damn, stupid game has ended. But no; L loved the game, despite his body language! N says he doesn't like any other word games, but he'd play this again in a heartbeat. J wants to play again RIGHT NOW.

All these players, who I thought were humoring me by continuing the game, enjoyed it! They all agree that the game ran a little long, but they were all engaged, and they all had fun.

So, I'm resolute. No big changes. I have to make tiny tweaks.

Playtest #4: A and S have been watching the game. I showed the game to A last year at a friend's house, and she was interested. I have no idea if she likes word games, but she seems like she wants to try it.

S has seen the game in various stages, and he's impressed by how far it's gone. He's an interesting guy, who has never actually played the game, but who glances at a board in progress when he's not playing something himself, and finds a word better than anything we see.

Bang, the game locks up again. I look at the board. Four uncommon tiles out. We invoke the cleanup rule, but now it's as if I've seen my enemy's face. We make sure no more than 2 uncommon tiles are out at any one point. It's a little fiddly, but it'll do for now. The game hums from there, and we finish in less than an hour.

I'm still not crazy about interrupts; both A and I have interrupted with 18-point words, and wound up losing points after the 4-point penalty. I suggest dropping the 4-point penalty to 3, and everyone thinks that might help.

So, the two tweaks I'll introduce in the game's next test are:

• No more than 2 uncommon tiles out on the board at any time. If it turns out to be too fiddly, I'll figure out some other way to address the problem. I could always add more common letters to the tile mix; that would at least make that replacement less frequent.

• Interrupt penalties will be -3 instead of -4.

Everyone really enjoyed the game, and I can't say how relieved I am. But I need more playtests!

Finally, an announcement: I just made an order for a bunch of poker chips, and I have a bunch of cloth bags. Once I'm happy with this round of tweaks, I won't be far from sending blind test copies out! I'm only 8 months behind with Seth's copy... :)

---

Permalink (referenced by: 0 posts / references: 0 posts)

Recap: In 2007, I brought Prolix to Protospiel. I thought it was more ceremonial than anything else. The game was mostly done, so what could possibly be suggested?

The game was hammered by all sorts of suggestions, some which didn't work, and some which did. I left knowing that the game wasn't done yet.

Turns out those suggestions were pretty good in the end. They led to a revised board and simpler letter-movement rules. I took these to Protospiel 2008, and battened down the metaphorical hatches for all the advice that about how the game was or wasn't yet.

I was stunned by the responses I got: "It's done." "How much are you charging?" "It's fun."

I won't say it's done just yet; I want to see how it does at Dexcon this weekend. But this game is awfully close to finished.

Some other quick observations:

• Tried 2-player Speed Prolix. It didn't work; it's too much math to count up the columns. Maybe if it was a different game...

• I posed a question about what to do about a game lockup. The unanimous answer was that a lockup would always occur if there's a problem player, who would cause problems in other games anyway. So for the boxed rules, don't worry about it. If there are ever Prolix tournaments, then maybe there should be a 3-minute timer. After all, timers are used in Scrabble tournaments, so there's a precedent there.

• I must get in touch with the game company about the new board. It's just playing too well to not let them know about it. Tomorrow, hopefully.

---

Permalink (referenced by: 0 posts / references: 0 posts)

The whole idea of Prolix is that it's a word game for people who like big words. People who are always one tile or card away from that awesome word in Those Other Word Games.

But one flip side that I've been lucky enough to avoid so far is the "Antidisestablishmentarianism Factor." If the game rewards such big words, why wouldn't "antidisestablishmentarianism" be called out at least once a game?

Well, the pat answer is that it should be called out, if it's the right word for the board. But on any other board, it should be as bad as any other ill-fitting word. If the word can perform decently on any board, well, that's a problem.

And now that I'm filling the board with more and more letters to prevent lockups, I'm noticing that I'm backing away from one problem, straight into another.

And this problem is arguably worse. Right now, lockups only happen with certain players, and they're probably players who lock up other games. But the "a..." problem is a break. It's a solution. It's an optimal strategy, or at least a single, dull strategy that will make the game Less Fun.

Ugh.

So I sat down and played 15 rounds against the word "antidisestablishmentarianism." I came up with two words every round; one for the current, 5-column board, and one for a 4-column board that I'm thinking might be a good compromise.

The outcome:

On the 5-column board, I beat antidisestablishmentarianism, 12-3. That is, out of the 15 words, "antidisestablishmentarianism" only beat my word three times.

My words averaged a score of 18.6 points, while antidisestablishmentarianism averaged a score of 15.13 (to people who haven't played with the 5-column board: yes, scores are much higher!).

So, not bad... but not great. An average of 15.13 is not a terrible score, and about what an average player will score. I do NOT want the "a..." strategy to result in "average" scores. I want it to result, over 15 words, in below-average scores.

The 4-column board was a little better. I beat "a..." 13-2. My words averaged 17.6, while "a..." dropped to 13.4. So my scores dropped a little (to be expected, with fewer letters out), but "a..." took a bigger hit than me!

So, the next playtest of this game will be on a 4-column board. That means 8 chips out in total. I don't know if this is the last of the "a..." problem, but I hope this starts to address it.

Of course, fewer chips means that I veer back towards likely lockup situations. But it looks like it's a spectrum, and I'm either going to be close to one side, or the other. If I can find a nice balance, I'll be happy.

Oooh, one more thing... I had another playtest this past weekend. I tried -2 interrupt penalties, but even those were still too low! I'm going to go up to -4 for the next game.

-4 is a weird number, isn't it? I almost want to round it up to -5, but I worry that's going to dry up the interrupts too much.

The playtesters had never played the game before, and they were intrigued by the original, increasing penalties of the interrupts. I like the elegance of the consistent penalty, but maybe the increase will make it back, if that will slow the interrupts down.

Phew. One stinking board change, and the whole thing gets thrown off-kilter. But I think once this board is stabilized, it'll be a huge improvement, at least mechanically, over the old board.

What does the new board look like? Oh, right.

O O O O
O O O O
4 3 4 2

Again, it's just a try. Maybe 8 letters is too few. I'd like to see how it goes. And I'd love to see what the Protospielers will think next month...

EDIT: I forgot to mention that there's one other thing I can do in case 8 letters is too few. I can restore the fifth column, but make it negative. Like this...

O O O O O
O O O O O
4 3 4 2 -1

I'd suppose that if a person uses a word with a letter in the negative column and another column, he'd have to use the negative letter first. Otherwise, it may not have enough impact.

I'm not crazy about this, because it flies against what I think the spirit of the game is, which is an "unrestricted" word game. I don't want people to feel like they have to avoid letters. But I have to mention it as a possibility, even though it's quite remote.

Also, with all this score inflation, I'm keeping the letter rarity bonus the same... for now. It still feels like it's at the right level.

---

Permalink (referenced by: 0 posts / references: 0 posts)

The game is at a crossroads. I'm curious to see where it goes from here.

First, we have the "older" 3-column board. Letters move when used. There's a bit of tactics, but it's possible that an extremely slow player can lock up the game under certain conditions.

Next, we have the "newer" 5-column board. Letter movement would be much more fiddly, so I've decided that on this board, all letters move one column to the right when used. So there aren't really any tactics, other than when to interrupt.

The higher scoring on the "newer" board also means that I need to raise the value of the interrupt penalty, from -1 to -2 or even -3. There are too many interrupts with the newer board, but raising the interrupt penalty will help with that.

I may also raise the bonus of the rarer chips, though that doesn't seem to be as pressing an issue.

The newer board is much simpler to teach and play with. I'm not so sure that I'll miss the tactics, but future playtests will bear this out. There's also always something to interrupt with, making turns nice, quick, and tense.

Hopefully I can get another playtest this Saturday, during my monthly game group...

---

Permalink (referenced by: 0 posts / references: 0 posts)

If there's one thing I can't stand, it's a game that comes with several different rulesets. Everyone agrees on one of the rulesets, and off they go.

This is tough for me, because it reeks of indecisiveness. Someone somewhere came up with more than one ruleset for a game, and rather than choosing the best ruleset, he passed the choice to the players. Players who tend to take 45 minutes deciding what game to play next, to say nothing about what version of the game they want to play.

I mention this because I'm in a position where I could supply several rulesets to Prolix. Gah. I'd rather play 'em all and choose my favorite... but I have the feeling that different groups will prefer different games. What does that mean? That the groups are really different? Or that the games are so mediocre that one is just the same as another?

Anyway. Here are the three rulesets I'm considering. All these would be played on the 5-column, 2-row board.

• Basic Prolix. Like I mentioned in my last post, this is the super-simple version of the game. All letters shift one column to the right after each word. This version may not even have interrupts, just to KISS. The idea is to be able to play it with a whole family, including 6-8 year olds.

• Advanced Prolix. This would be the "regular game," as we know it. Letter chips only move when used or forced. I'd like to keep it on the new board, as that seems less likely to lock up.

• Speed Prolix. Hoo boy. Here comes the part where I contradict myself.

Remember when I said I wasn't interested in making a puzzle version of Prolix? And that I didn't want any timers? That was what, my last post?

In the spirit of pure hypocrisy, I'm going to try these rules at this weekend's playtest...

• Start the game by having one player fill up all 5 columns and 2 rows on the board. He keeps the bag. A 30- or 45-second sand timer is kept within easy reach of all players.
• All players study the board.
• At any point, one player can take the timer, flip it in front of it, and bid a number (the bid value). That number should be the highest-valued word he thinks is out on the board.
• All players (including the player with the timer) now have until the timer runs out to come up with a word, and write it onto their scoresheets.
• All players score their words. Each player then takes the bid value and subtracts it from his score (edit: corrected from seo's advice). He applies the resulting modifier to the score.

For example, in a round where the Bid Value is 15, Player A has a 20-point word, and Player B has a 12-point word. Player A's modifier is +5 (20-15=5). Applying this to his score, he gets 25 points for the round.

Player B isn't as happy. His modifier is -3 (12-15=-3). He gets 9 points for the round.

It's a little math-y, but for now, it accomplishes what I need. The only thing left is to figure out the bonus for the player with the timer, because there needs to be an incentive to grab that thing. I was thinking something like +2 points for every player who didn't exceed the bid value, or a flat +5.

• Finally, all chips move one column to the right. Chips in the rightmost column are discarded. The bag passes one player clockwise. That player draws two new letters into the first column, and the next round starts.

• The player with the most points after six rounds is the winner.

We'll see how it plays out. I have the feeling that the Saturday group will prefer it, but my regular players will hate it!

---

Permalink (referenced by: 0 posts / references: 0 posts)

Yesterday, I played with some folks from the Albany playtest group. Several had played the game before, and they felt that the letter chip movement rules are still too complicated. Once that came up, it seemed to be a unanimous feeling from everyone in the room. The letter chip movment rules are too fiddly.

So we tried another game later in the evening, with a different board and much simpler movement rules.

The board looks like this.

OOOOO
OOOOO
34342

Each O is a letter chip space, and the third row are just point values. All letter chip spaces are full. Every round, the letter chips move one column to the right, with the last column being discarded and the first column getting new chips.

That's it. No overflow spaces, no "you have to move this chip first."

Pros...

• It's much simpler. Used chips are still flipped over to count their score, but all chips advance. It'll take about 45 seconds to explain the game rules now.
• There are always 10 chips on the board, no more, no less. There are also always 4 chips in the premium four-point spaces. So everyone will always have something to work with.
• Fewer play errors. In the afternoon playtest with the old board, players kept missing overflow spaces. No chance of that happening here.
• With this much material on the board, lockups will be much rarer. In theory, there will always be 32 points on the board, plus bonus points from rare chips. The only way the game could lock up is if everyone has scored a minimum of, say, 25 every round, no one can see a better play than that on the last turn, and the active player will not move until he sees a play that is better than his interrupts. That'll be extremely rare, and we may just be able to file it under "game group problem."

Cons...

• Just one, really. This board is completely non-tactical. Nothing you do can negatively affect your opponent.

There's some history to explain here. In the pre-Protospiel '07 version, the board had four columns instead of three. Emptying the high-point column was a key tactic, as it left the next player with little to work with. Of course, that slowed the game down terribly, as the next player took forever to come up with a word that broke ten points.

The three-column board had less tactics than the four-column board, moreso when I changed the size of the first column to make sure there were usually letters in the high-point column.

The new board is completely tactic-less, other than "score what you can." This is good because it makes things simpler, but I worry that I've thrown the baby out with the bathwater. What if it makes the game too light?

Tactics in Prolix are a bit of a mixed bag. The original idea was to have light, subtle tactics that reward clever play and extend replayability. I always wanted the focus to remain on long, impressive words, but some "this is what you should do here" play wouldn't be bad. I didn't want as many tactics in Scrabble (i.e. don't open up a Triple Word Score unless you get more points out of it than the play you're opening up for your opponent), but I wanted something in there to make it a little like a game.

You can look at it as a spectrum, like this.

Puzzle Game < — > Prolix < — > Tactical Game

Here, a "puzzle game" is one that all players have to try to solve simultaneously, like Ubongo or Ricochet Robot. Nothing against those games, but I find that Prolix plays better when each player has something resembling a "turn." We tried "Puzzle Prolix" at Protospiel '07, and it just wasn't as interesting.

A "tactical game" is a turn-based game where tactics underscore the game mechanics. For a word game, this becomes "good things to do" in addition to coming up with high-scoring words. In theory, a word game with strong tactics would force players to decide between a move that gives them high points with a great word, or high points with a good word and good tactical play.

I want Prolix to fall between these two categories. But where? At first, it was closer to Tactical Game, but as those rules fell away in the name of anti-fiddliness and game length, it's inched towards Puzzle Game. If I do go with these new board rules, that means less game depth, and possibly less replayability.

So, my first recourse is to ask people who have played the game: what do you think? Would you miss the tactics in the new board? Or do you prefer the emphasis on high-scoring words?

---

Permalink (referenced by: 0 posts / references: 0 posts)

After a couple of days' of reflection, I'm realizing a few things...

1) Yeah, Prolix still has a lockup problem.
2) The lockup problem is pretty rare. I've playtested this version of the game dozens of times, but this is the first real lockup I've had.
3) There's no need to throw off the equalibrium of the game for a rare problem.

So I'm going with the first solution for now. It's going to go like this...



I'll give this a try, and see if it starts to occur too frequently. My biggest concerns are...

• players somehow using this to metagame
• a slow player not agreeing to new chips, holding everyone else up.

We'll see, but I'm hoping it's unobtrusive enough that it resolves the lockup, while still letting the core of the game shine through.

---

Permalink (referenced by: 0 posts / references: 0 posts)

It's always a good news/bad news situation when playtesters break your game.

On one hand, you have some valuable data that you can use to make your game stronger.

On the other hand, you realize how ill-prepared your beautiful baby is to enter the world.

This past weekend, I was in a 3-player Prolix game that ground to a halt on the very last turn. Completely locked up. Yes, the players were a little AP-prone, but I can't forbid people from playing the game just because they like to think. The game has to be able to recover from a potential lockup.

All that needs to happen for Prolix to lock up is the following:

a) Players who have interrupted a lot early on, so they risk lots if interrupting again.
b) One player who will not make a play unless it makes a significant difference in his score.
c) No one willing to take a bullet to make the game move along.
d) A board with a lot of difficult letters.

a) and d) are entirely possible, and c) should be expected. Most of my playtesters are willing to forego b), but one player was determined to find a high-scoring word. We couldn't finish the game because of that.

I'd love to get all snooty and say, "don't play with people who take so long," but the fact is, the game needs to be stronger.

Prolix is an elegant game. This sounds like a boast, but it's actually a complaint. Any tweak I make to it has to be as tiny and subtle as possible. I swear, I'm only designing sprawling chromefests from now on.

So, I'm looking at a few possible solutions to the lockup bug...

a) Have the players, by unanimous consent, add two letters to the board whenever they want.
b) Include a... include a... (sigh) timer.
c) Increase the number of letters added to the board later in the game.

a) is a pretty straightforward solution, but it's also very much a hack. There's something I don't like about it, but I can't really put my finger on it.

b) isn't quite that bad, even though it's been my intention to avoid a timer for the game. I might be able to pull it off, without causing the brainfreeze that a timer normally causes, and that I'm desperately trying to avoid.

b1) If a player doesn't think anyone can come up with a good word, he flips the timer. That player cannot come up with a word while the timer is running.
b2) If no one comes up with a word by the time the timer is done, the player who flipped the timer gets 1 point. Two new letters are added to the board, and the active player starts his turn again.
b3) If a player comes up with a word while the timer is running, his word is worth 1 extra point.

It's a little fiddly, and it adds a dimension to the game that I don't necessarily want to add.

I just thought of c) while writing this, and it's intriguing. It would add an arc to the game, because the extra letters would make the endgame feel different from the opening. It probably won't add a whole lot of overhead to the rules either. But it doesn't directly address the lockup problem, so I'd have to be sure the extra letters are worth it.

If this was a gamer's game, I'd tie the extra drawn chips to the number of interrupts made, but that's a little too crunchy. I might have players draw 3 chips starting in Round 4. I might just have the player who starts each round draw 1 or 2 extra letters. Or I might trigger an extra draw every time the 3-point column was empty, though that's a little heavyhanded.

I may even increase the number of board spaces later in the game. That would certainly increase the number of chips on the board.

Tough choices. I'm leaning towards c) for now, because it's the simplest fix, and it's firmly in the spirit of the game.

Time to sleep on it. Hopefully I'll have an answer soon.

---

Permalink (referenced by: 0 posts / references: 0 posts)

No, don't get excited. I spoke with them a couple of weeks ago.

Apparently, the game has made it out of the closet and onto a table.

Progress, I suppose.

---

Permalink (referenced by: 0 posts / references: 0 posts)

I've had a couple of playtests lately that have shown me that there's some room for me to tweak the Interrupt system. Obviously, at this point, I'm just making the tiniest of changes, because even those can send enormous ripples throughout the whole game. These changes will almost always be cuts, not adds. I'm looking to streamline and simplify, and hopefully make the design more elegant.

Recall that there's an increasing penalty for interrupting players. This means that players will almost always max out at 2-3 interrupts (though I once saw a player interrupt 4 times, and he still won the game!).

I'm realizing that this increasing penalty isn't really necessary. There's already an increasing penalty that emerges from the game. If you interrupt once, you'll replace your lowest score. The second interrupt replaces your second lowest score, and your third replaces your third lowest score. At that point, you're most likely risking losing points already, and the increasing penalty ensures that a player will likely be disengaged from the game flow by the last round. That might make the game drag for some.

So I'm going to try a flat -1 modifier for all interrupts. I don't want to eliminate the penalty just yet, because I'm worried about the game starting to become a puzzle game, which we've already found isn't as much fun. Maybe at some point, I'll try eliminating the modifier, because perhaps the replacement mechanism is enough to discourage wanton interrupting. It would certainly be a more elegant game without the penalty, but I don't want the skill levels to become too stratified.

I had an interesting game today. I saw all sorts of potential issues come up; many letters leaving the board, difficult letter combinations, and two times when we had to use the tiebreaker. One of those tiebreakers wound up with interrupting players with identical scores, which is resolved by the player closest to the active player breaking the tie.

Through it all, the game bent but never broke. No one "gamed" the system, everyone had a good time, and it lasted a quick, punchy 30 minutes. It's an incredible thing, to see a fragile game that's broken by the tiniest bit of metagaming slowly become a solid, strong design that can withstand a wide variety of experiences.

I'm really starting to respect designs like Zooloretto, which can be played casually or seriously. You can approach it with heavy strategy, or by the seat of your pants, and it's still a fun game. That takes a hell of a lot more work than I ever expected.

---

Permalink (referenced by: 0 posts / references: 0 posts)