Tuesday, September 25, 2012

Tweaking the Bad Quarterback League

Anyone who has ever managed a fantasy football team has suffered the indignity of watching the team's points decrease after the quarterback throws an interception, fumbles, or gets sacked for a safety. It's frustrating and can mean the difference between victory and defeat. But what if there was a way to celebrate the badness of your quarterback? What if interceptions, fumbles, and safeties became things you wanted to see? Enter perhaps the most important innovation in fantasy sports since our own Fantasy Olympics: the fine folks at Grantland have developed the Bad Quarterback League to do just that.

The premise is simple: assemble an eight-person league, and divide up the league's thirty-two quarterbacks by whatever draft mechanism you see fit. Then, your quarterbacks score points for everything that goes wrong in their games. You don't have to pick a quarterback in particular--so if you own Jacksonville, and Blaine Gabbert gets benched, you still get points when Chad Henne fumbles. Last year was the league's inception, and this year marked the introduction of the Failure Machine, an automatic scorer for your league.

I started a league, "What the Tannehill?", inspiring a few friends to join and coercing a few others, and set out to use the Failure Machine. But the more I played around with it, the less I liked it. Here are my tweaks to the Bad Quarterback League format:





Each player only drafts two teams. Nobody wanted to go through the motions of drafting teams with quarterbacks who can actually complete passes. And if we decided to go with a "starters/bench" league format, is anyone realistically going to play Tom Brady or Aaron Rodgers ever? We limited our draft to an eight-person snake with a single bend, so once the guy with the first pick also took the sixteenth, the draft finished.

Each player can "start" both teams every week. The Failure Machine builds in a limited number of starts for each team, so you can't just ride Brandon Weeden to victory every week. But that's a lot less satisfying than being able to rack up points each time he's intercepted for a touchdown. It would be like setting up a "real" fantasy football league where you draft Drew Brees but can only start him an arbitrary fraction of weeks. Furthermore, I'd sold the league to my friends--many of whom are competing in at least one "real" league--assuring everyone that BQBL would require a minimum of effort and basically no management after the draft.

Scoring is cumulative, not head-to-head. In "real" fantasy football, each player has a head-to-head matchup against another team every week. The "winner" is determined from how many of these matchups any given team wins. That's how the Failure Machine is set up as well. But with so much of the decision-making already stripped from my version of the BQBL, the head-to-head model didn't make much sense. Instead, we'll be scoring our league in at least two ways. First, all the cumulative points scored by both of a player's teams will be added together to get the league's points champion. Second, the league's consistency champion will be the player who has the highest score in the most weeks. We may give an award to the person who drafted the best, as well.

The result is a system that doesn't mandate much strategy (or reward much strategy, depending on how you look at it) but does provide a humorous look at the game's less-renowned quarterbacks. It gives a nice respite even if your actual fantasy team has collapsed--maybe Eli Manning threw four interceptions, but at least Sam Bradford did too. Most of all, it works the same viewership magic that any expression of fantasy sports does: why else would you have a reason to watch a Raiders-Cardinals game?

If you're interested in how all this works, you're welcome to follow along with our league.

Saturday, September 22, 2012

Board Game Design: Identifying Multiple Paths to Victory (Part I)


In 1985 Paramount Pictures released a live action comedy based on the popular board game “Clue”. Featuring Tim Curry in one of his best performances, the movie “Clue” received mixed reviews and was considered commercially unsuccessful and thus like many others I first experienced it on video. I was enamored by the perfect casting and I’ve always loved mysteries but what made the movie memorable for me was that after the mystery was explained and I expected the credits to roll they hit me with two additional alternate endings.

The version of “Clue” feature film in theaters provided the audience with one of three endings, but the home video release ostensibly made more sense to the viewer as it helped establish that the actual turn of events was not as important as the journey made to get there. This was certainly not the first instance of multiple endings used in film and it didn’t fulfill its intended commercial goal of selling more tickets to the same audience for an alternate ending, but nevertheless it was my first memorable exposure to the idea and it made an impact.

A finely structured board game relies on a satisfying endgame even more than defining mechanics, integration of theme or player interaction. Its the aftertaste players will remember whether they won or lost that will bring the game out of the closet at the next opportunity. The final outcome should reward player(s) that performed the strongest, have a consolidated final scoring that is not entirely transparent and keep those who fall behind early engaged and hopeful.

One of the best ways to make the victory of a particular design distinguished from its bookshelf competitors is to allow multiple ways for players to find their own paths to victory. By providing several options, players maintain a tense struggle for resources while waging a war on multiple fronts.

For the purposes of this article I am particularly interested in games offering very divergent victory goals. This excludes games with pass-through conditions such as High Society in which currency is used to buy items worth VP but the player who spends the most during the game is eliminated prior to final scoring. I’m also excluding games for which multiple victory conditions are aligned in the game-play: in Mr. Jack the person playing as Jack can win by escaping the board, evading capture on the eighth turn or an incorrect accusation by the inspector, but in all cases Jack’s motivation is to conceal his/her identity effectively.

Some common trends in games that have multiple paths to victory:
  • Games offer low player counts (often just 2 players, while 4 players is less common and typically the upper limit of the player count)
    • Fewer players allow participants to monitor progress of opponents during the game while not overwhelming players with too many potential outcomes.
  • Victory Points are often not used and one player is the winner with few methods to measure parity of the game. If VP are used it is often as a tiebreaker.
  • Games often have an upper limit on the length of the game but surprise victories are not uncommon when a player is ignored by opponents and crafty with their decisions. 

Tactical Victory:

The tactical victory is often the bread and butter of the game; the portion of which the most action resides and in which meaningful decisions are made. In Risk, your tactical objective is global domination. By virtue of being the primary focus of all participants this is often the least likely victory condition. When a particularly dominant player is fortifying their position an attack-the-leader syndrome from other players offers further balancing to the game and often prevents a tactical victory.

Set Collection:

Set collection could be considered the secondary condition following the tactical path to victory. It is tremendously important that a player maintain an active role in the tactical gameplay (to prevent opponents from achieving the primary tactical victory) but set collection is an alternative path that keeps players engaged in monitoring the progress of their opponents and increasing tension while bringing players towards the endgame.

Hourglass Victory:

In a hard fought game amongst several worthy competitors it is essential to bring the game to a conclusion rather than drag out the climax of the conflict and remove the unique experience. The hourglass is the tertiary objective filling the role of closer. When a victory cannot or has not been achieved in the designed time table of the game, the hourglass objective is the final act in determining the victor. Ideally this utilizes a different criteria than the other victory conditions; something as simple as the player with the most currency at the end of the eighth and final turn is the winner. This condition ensures a timely endpoint even if all players are playing optimally and still allows for a satisfying and definitive result as it is foreseeable.

If victory points are used in the design of a game, they hold special value here as they can be collected on the side by players and only have value if the turn limit or stalemate condition is met.

In Part II we will look at some of these victory conditions in action and see how their relationship creates depth without added complexity in game design. Did I miss a fourth category of victory conditions? If I did I would love to find out your thoughts.

Friday, September 14, 2012

Fantasy Football: the Kicker Stigma

It's conventional wisdom in fantasy football that you wait until the very end to draft a kicker. Most of the "veterans" in my league voiced the opinion during this year's draft. There's even a commercial about it, where NFL.com advises you to make fun of your college buddies for drafting a kicker before the last round. On the surface, it makes sense: while some kickers are clearly better fantasy options than others, kicker is the least likely position to come up with a massive play and pull out a surprise win for your team. But does the math hold up?

Let's start with the extreme case: the incomparable David Akers, by far the most valuable fantasy kicker in the league. Akers dominated the kicker landscape last year to the tune of 182 points, surpassing the second-highest scoring kicker by 32 points. His average points per game last year was 11.375 (ESPN standard scoring), which is production you'd be happy to see from your starting tight end or second wide receiver. Akers actually scored higher than fantasy luminaries Darren Sproles, Frank Gore, and Roddy White.

Does it make sense to draft Akers before you fill up all your other starter slots? Probably not. And you obviously need at least one backup wide receiver and running back to hedge against injury and plan for bye weeks. But fantasy football, at least in ESPN's system, gives a whopping seven bench slots. Even after securing a backup quarterback, running back, wide receiver, and tight end, there are still three picks remaining--and it's not clear that even the first backup is going to be worth more than your kicker.

It's tough to imagine a scenario where your backup running back is going to score more than 182 points in a season. For an eight-team league, the third RB selected by any given player can expect to score in the range of 140-160 points (120-140 for a third WR). And that's assuming the backup starts every week. For most teams, the backup is going to have a limited workload, subbing in case of injury or a bye week or a particularly favorable matchup.

Plus, by the time the draft reaches the "backup RB/WR" stage, the marginal decrease per player pick is negligibly small. For a backup running back in the range we're talking about, you can expect to lose about 3 points over the course of the season for drafting the next-best player. The drop in expected production from, say, Pierre Thomas to Brandon Jacobs shouldn't be enough to overdraft Thomas so he can sit on your bench most games.

However--and here's the big caveat--the same marginal analysis suggests that overdrafting a kicker doesn't make any sense either. We'll set aside David Akers, the clear overachiever of the 2011 kicker class, for a moment. The average marginal decrease over the second- through twelfth-best kickers was a mere 1.5 points. In other words, you're only losing about 15 points over the course of the season (or less than 1 point per game) by taking kicker number 12 over kicker number 2. That analysis suggests that once somebody drafts the clear outlier Akers, you're safe waiting as long as possible to get a kicker. You may have emotional attachment to Sebastian Janikowski over Robbie Gould, but you shouldn't have the same mathematical attachment.

The only remaining question: how much of a draft premium should David Akers have? That depends on how likely you think his chances are of repeating his outstanding 2011 performance. If you think that 182 is no fluke, it makes sense to draft him before any of your bench players, and arguably before your second starting wide receiver or your "flex slot". Only six running backs, four wide receivers, and two tight ends outscored Akers in 2011. That puts him in the top fifteen most valuable non-quarterbacks in all of football. 182 points is a lot, and you're going to like getting that production, regardless of where it comes from. If, however, you think he'll regress to the mean and score in the (more human) 150-160 range, there's little advantage to taking Akers any earlier than the last round or two.

In summary, the conventional wisdom regarding kickers seems to hold up, but it's because of the remarkably similar production among first-tier fantasy kickers rather than a systematic lack of value that the position carries. You're justified in taking a gamble on a kicker that you expect to have unusually strong production for the position, but after the lone superstar is gone, there's no significant advantage to gain from drafting any kicker over the next-best.

Friday, September 7, 2012

September Game of the Month: Dramagame Velvet Sundown

September's Game of the Month looks a little different than usual. A few weeks ago, I got an email from Elina Arponen of Tribe Studios with an invite to beta-test their game, Dramagame Velvet Sundown. (If it's a bit of a mouthful, it's only because Velvet Sundown is the first "scenario" in the Dramagame model.) Rather than a board game, this one is totally electronic.

Style and Gameplay

Dramagame is an online multiplayer roleplaying game. If one of the primary "axes" of roleplaying games is mechanical vs. narrative, then Dramagame is so far to the "narrative" side that at times it feels less like a game than like an interactive fiction exercise. That's not a bad thing, and it's intentional: there are few if any mechanics in Dramagame, and the stated goal is not to win or lose but to create an interactive, social shared experience.

A session of Dramagame--called a "show" to emphasize its narrative quality--lasts 45 minutes and can accommodate between 4 and 11 players. Each player is randomly assigned one of 11 characters, so chances are good that each individual game will turn out differently than the last. Unlike in many roleplaying games, there are zero NPCs; all the characters are player-controlled, and all the stories are player-driven. Much of the game is spent simply chatting with other characters, and the experience is the best when the most people are confident roleplayers.

To keep it a "game" at all, there are specific "quests" and tasks that each character is given. All of them involve interaction with other people: trading items, verifying identities, stopping trouble from breaking out, even uncovering hidden conspiracies. The only mechanical thing about it is that each item has a specific function, and deciding whether or not to use them can drive the story in vastly different directions.

Each show lasts only 45 minutes, so Dramagame emphasizes quickly establishing a character and jumping right in to make progress on your quests. As simple initial quests are completed, other more complicated ones can emerge; the best games of Dramagame weave a web that unites all of the characters into some dense tangle of plot lines. Shows are short on purpose; the game is designed for people who can't necessarily dedicate many hours per night to gaming but want to have a meaningful game experience anyway.

Analysis and Anecdotes

I've only been able to play through one show of Velvet Sundown, and even then I couldn't finish it because of some internet issues on my end. Still, it was great to see the game "work" as advertised--everyone quickly dug right in to the roleplaying and aggressively pursued their characters' goals. I got randomly assigned to play Mary, the first mate of the ship the game takes place on, and I was immediately alerted to three things I was supposed to work on: checking passports in my capacity as an officer of the ship (which seemed legitimate enough), scanning the IDs as an undercover agent of a corporation (which seemed a bit less so), and getting my flirt on with a dude named Cooper and convincing him to write me some poetry.

And that was only me--each of the four other characters presumably had their share of secret agendas too. It was a whole lot, almost an impossible amount, to get done in 45 minutes. That's probably intentional, too. If every character simply and quickly completes their quests, then there's no more narrative tension. As long as everyone still has something to work on, then something is still happening in the story. There was a lot of "you scratch my back, I'll scratch yours" that emerged in my game, which makes enough sense but was still fascinating to watch given that "winning" isn't exactly the point of the game.

Elina says she and her team were influenced not only by computer games and RPGs but also by board games and even improv theater. There's a strong flavor of point-and-click adventure games, where one of the goals is to use items in combination or on other characters in creative ways. And the roleplaying side reminded me a whole lot of Burning Wheel, where each character is motivated by a set of personal beliefs and goals, and the game is about working with other characters to get that accomplished.

It's obvious that Velvet Sundown, and the whole Dramagame model, is still a work in progress. Tribe has clearly thought about their setting, and its details and back story, but so far not much of that detail has made it into the game. Similarly, the luxury yacht that the game takes place on doesn't have a ton of personality of its own; it would be nice to have an excuse to explore the ship and interact with it somehow. Tribe's plan is to work a lot more with Velvet Sundown, developing new features and content, and perfecting the model before moving on to develop new scenarios.

Overall Impressions

Dramagame is probably not going to appeal to everyone. There's not a lot of "play" in it, so Euro-game aficionados who delight in mathematically optimizing each turn of a board game probably won't find much here. Neither will RPG fans who are looking to play through eighty hours of an epic story and attain godlike powers by the end. But that's not exactly who the game's developers are targeting. Finally, if you're a hardcore console gamer interested in cutting-edge graphics and a lot of action, Dramagame might not be your thing.

On the other hand, if you're time-constrained and would like to have a meaningful, entertaining experience in a short time, you'll probably like Dramagame. Same goes if you've ever played a point-and-click adventure and wished it were multiplayer, or if the most fun RPG sessions you've ever played didn't involve a single die roll. Right now, the community is small and friendly--you might just as likely play with strangers from Denmark or the game's lead programmer--and everyone is excited and eager to hear feedback. At this point, there might be sort of a lot of feedback to give, but once Dramagame has time to grow, it should only get more fun.

Online, 4-11 players per game, 45 minutes, 13 EUR (about $17)