Designer: Amabel Russell
Publisher: Winsome Games / Rio Grande Games
Time: 20 minutes
Times Played: 5 times with review copy of the RGG edition, 1 time with someone else’s RGG edition
In retrospect, one of the worst Codenames clues I’ve ever given was “dazzle”. I was using several meanings of the word, but the one that had less traction than I had anticipated was a type of camouflage.
Dazzle camouflage bears a visual resemblance to Op Art more than traditional woodland camouflage. The goal is less hiding in plain sight and more ‘course deception’. The idea was that British military ships in WWI were being targeted by German U-Boats via prediction of where the ship would be when the torpedo got there, so: could you alter der seaman’s mental calculations through optical illusion?
This image, and others I’ll sprinkle throughout, are from Blodgett’s 1919 MIT thesis looking at the use of dazzle in WWI and theoretical modelling of results.
While the Bureau of War Risk Insurance and other American parties were researching the best ways to lower visibility at sea in 1917 and 1918, a British Lieutenant, Norman Wilkinson, created what came to be known as British Dazzle. Around this time, Germany had begun ‘unrestricted submarine warfare’ – which is to say survivors were not necessary – and Wilkinson’s camouflage strategy was the distortion of a ship’s size, shape, heading, speed, range, and so forth. The goal was simply to cause the U-Boat commander to mis-estimate the true course of the ship, resulting in errant torpedos – which were of limited supply and would give away the submarine’s position.
Wilkinson’s designs would be unique for each boat; tailored to the specific ship; and he would find his wife among the team that helped with the final designs. Wilkinson had a model ship and a seascape backdrop prepared and during a demonstration, King George V, who had previously served in the Royal Navy, misjudged the direction of the ship, much to his astonishment, due to the dazzle pattern.
The Armory Show was 1913. Here’s a photo of the HMS Kilbride from sometime between 1914 and 1918.
The Royal Navy ended WWI with more than 4000 ships dazzled.
Blodgett estimated that if a ship’s heading could be misestimated by 18 degrees, it would have a significant effect on the vessel’s and crew’s survival. With very intentional thought behind the paints and patterns used, Blodgett’s experienced observers’ best estimates were off between 20 and 30 degrees.
On both sides of the Atlantic, the Naval positions after the war were that the dazzle experiment had been inconclusive as measured through the potential of a ship being lost. It was, however, seen as a victory on another front as the insurance premiums on dazzle ships were less.
With the advent of radar, the optical deception of dazzle was no longer of use. But when I play Northern Pacific, I’m a flustered U-Boat commander not sure which way this train is going and not sure where to invest in the future – though rather than an optical illusion, I’m perplexed by some sort of mental illusion. For that, we’re going to take another detour, this time through the research of Tversky and Kahneman.
I was recently reading Michael Lewis’ The Undoing Project, and I didn’t enjoy it much. He’s an author for me who gets the benefit of the doubt: I’ll read his books even if they don’t necessarily interest me. This one I understood to be an apology of sorts for how much credit Lewis received for the Moneyball concept, as afterwards, folks pointed out that Tversky and Kahneman had essentially received a Nobel prize for the underlying ‘experts-schmexperts-;-let’s-do-the-math’ thesis, and Lewis wanted to bring light to a few interesting parts of their research. (That’s the part I enjoy, and that we’re going to talk about. Overall it fell flat for me as I didn’t care as much for the portions on their relationship, and that is a large focus of the book.)
In their 1974 paper from Science, “Judgment under Uncertainty: Heuristics and Biases”, Tversky and Kahneman (funded by ARPA, which much of their research was), look at many of the ways in which humans misjudge calculations of probability and statistics.
They talked about people’s judgment being limited by available representative situations that can be brought to mind. That is, a person will inadvertently assess the risk of certain diseases by calling to mind how many people they know afflicted as such. Without regards to sample size concerns (which are also covered in the paper), they demonstrated other more subtle concerns.
For instance, some information being easier to recall than others. In one experiment, they presented subjects with a list of names, some female and some male, and asked them afterwards to say if there were more male or female names on the list. The subjects could be swayed by the inclusion of celebrity names of the gender with less overall names, as the celebrity-status of the names led to a greater “availability” of the information, which skewed the subject’s answers.
In another experiment they demonstrated that some information could be more available than others because it was easier to search for in your mind. Are there more three-letter words that begin with r or have r as the third letter? The findings of such experiments are skewed as it is easier to recall words that begin with a certain letter than have it at later positions (a fact exploited by Looping Games’ Password Express).
Not restricted to mental recall, they also demonstrate issues with calculation and cultural correlations: subjects underestimated permutations that involved larger numbers and estimated that drawings which contained eyes were more likely to have come from suspicious individuals.
Tversky and Kahneman also looked at “anchoring”, where individuals’ final answers are skewed by where they start. Estimates of 8x7x6x5x4x3x2x1 where higher than those for 1x2x3x4x5x6x7x8 –with a median average of 2,250 for the first, 512 for the second, yet 40,320 is the actual answer.
They looked at predictions: select a number such that the probability that the number will be greater than the Dow-Jones average on a particular day is 90%. Some individuals were asked to select values corresponding to certain percentiles, others were given numbers and asked what percentile they would fall in. Anchor points in the two scenarios skewed the respondents’ answers.
Anyway, in Northern Pacific, players are collectively laying track for a railroad from Minneapolis St. Paul to Seattle. One continuous line, no branching, and no doubling-back. Subtleties in the art indicate which direction the track heads.
Players will be placing cubes in various cities where the track hasn’t yet reached. These cubes represent their investment in that town. If the railroad reaches that town, the player(s) will retrieve that cube and 1 (or 2 if the investment was a large cube) other from a pool. If the railroad skips that town, the player(s) has lost that cube.
On their turn, a player either extends the railroad or places a cube. There are player-count based stacking limits for how many cubes may be in a city, and each player starts with 3 small and 1 large cube. Play continues until the railroad reaches Seattle.
Record each player’s score (the number of cubes in front of them) and each player’s failed investments (cubes left on the board). Reset the board and the cubes in front of each player, and whomever has the most points after three rounds wins (with the failed investments acting as a tie-break.)
That’s all the rules.
This is the game I’ve always wanted to play.
I don’t want to oversell it, but it’s brilliant. When I ramble about wanting to like train games and games with emergent player alliances and whatnot, this is what I’ve been looking for.
There’s a certain purity to its simplicity. There’s a clear abstract game that could’ve been at home as a nestor title, but has a firm train game lineage. There’s a distillation of a game down to a single mechanic that is reminiscent of many Japanese titles.
The game’s incentives draw and quarter you: you need to invest here because player y is driving the train there in 2 turns unless player x adds a cube there first in which case you need to have driven the train over here lest it bypass this city you’ve invested in but if you do that player z will invest over yonder and you need to diversify from z’s investments in order to stratify the scores. That’s the game. Over and over.
Except, then something else happens. You missed an option that player w found. And that happens over and over. Game after game.
Now your mind has switched to thoughts of what if. Mentally undoing what has been done.
In their 1981 paper “The Simulation Heuristic”, Tversky and Kahneman looked at this. How does the mind revisit events and which parts of a tale do our neurons poke at to change an outcome? The paper centrally looks at a fatal car accident. In the narrative, details are added and omitted: the deceased woke up late; next he was out late the night before; maybe he took a different route; maybe there was traffic; the accidenter had a drug problem.
We all have these. (I mean, right?) Why is that what I chose to say to Vin – I could have said so many different things. How did I let myself write that in that card in third grade. What if I had finished my homework that day in Linguistics. Why did I not return either of their calls. Running over and over in your head what could’ve been different. (How did I buy the wrong tape!)
Tversky and Kahneman assure us that these ‘undoings’ are subject to errors and biases. (How could I have lost my temper about the film!) They assure us that more research is needed, but point in the direction of the ‘joints’ in the narrative where there are dramatic events and a bias towards a single plausible scenario over a multitude of unlikely twists:
By its very nature, a plan consists of a chain of plausible links. At any point in the chain, it is sensible to expect that events will unfold as planned. However, the cumulative probability of at least one fatal failure could be overwhelmingly high even when the probability of each individual cause of failure is negligible.
I’m glad you’re here,
Thoughts from Other Opinionated Gamers
Dale Y: I don’t love the game quite as much as James Nathan. But I do like the colored pictures of ships. For me, this game is great due to its simplicity. I still have my original Winsome copy of the game in its antiquated clamshell case – and it still gets played from time to time. The production qualities of the new version are seemingly better, though I find the font on the cover and rules to be quite hard to read. I think one of the goals of a game box should be to convey the actual title of the game as well as perhaps an inkling of the theme to the player. In typical Winsome style, the rules are brutally simple and concise. The game starts off with immediate tension, and each round, I am faced with the dilemma of wanting to play cubes in good spots OR move the train line towards places I want it to go. But, of course, I can’t do both. And whichever I choose, I’m gonna lose out on the course of the other. Sigh. The game can move brutally fast. Sometimes it feels like a blitz chess match with each successive player pounding down a move quickly and having play move onto the next player. Some entire games of this have lasted under 15 minutes, and there’s a lot of game, decision making, and screwing over of your opponents to be had in that timeframe.
Ratings from the Opinionated Gamers
I love it! James Nathan
I like it. John P, Dale Y
Not for me…