• Keith

Mean it: The Debate Over Unit Ratings

I was listening to the excellent Harold on Games podcast a few weeks ago and heard Harold’s interview with Jack Greene. Jack mentioned that he used to get more wound-up about wargame unit ratings when he was younger. Greene also pointed out that beloved topics, such as World War II or the American Civil War tend to elicit strong opinions about unit ratings from a broader audience.

This got me thinking about unit values and what they actually mean or represent. Today, I’m going to provide a few thoughts on a sprawling topic that could, in and of itself, fill a book on its own. I am only coming to the topic from the perspective of a fan and not from the perspective of a designer who has had to wrestle both with the decisions and their fallout.

The Three Things…

Rather than hide this in the text, I’ll provide the three things I think of when considering unit evaluation right at the frontend so you can agree or disagree with me right away!

  1. Designers have opinions….and those opinions actually DO carry more weight than a player’s opinion.

  2. Numbers are meaningless on their own. They only carry context in relation to other unit ratings.

  3. People are going to be angry no matter what. The designer’s logic being sound is all that matters.

Let’s keep it simple and start right at the top of the list.

Designers Opinions Matter More Than Yours!

Games need to say something, first and foremost, about their topic. A game without an opinion or a unique approach doesn’t need to be made. Consequently, by the time a game makes it to a player, the designer’s opinion has had to carry the thing from idea to pitch to design and finally to production. That oversimplification belies all the debates and effort that go into each step along the way.

The designer isn’t trying to appease a gamer’s personal or widely held belief about a unit, nationality, or conflict. We’ll limit the discourse of this article, of course, to just unit ratings and say that what a single historian or widely held popular belief was about a leader or unit may not stand up to further historical research or testing.

Historians debate these things all the time, why shouldn’t that debate exist in wargames!

Instead, designers are looking at outcomes in the sense of the narrative context they have created for the game. In an American Civil War game, it’s not necessary to worship at the feet of John Reynolds giving him godlike powers had he only survived Gettysburg. Similarly, those Russian T34/76 units seemed to do pretty well on the eastern front from a logistics standpoint so maybe their ratings in a game that favors mechanical reliability does rate them better than their German counterparts.

Trust the designer if the ratings adhere to the logic of the case the designer is making with the game’s rules in short.

Numbers are Relative

We tend to base our games on pretty low numbers. They’re easier to do the math with in our heads and lend themselves well to simple charts. The closer we can get to binary evaluation as humans the better we are at making quick value judgements. This speeds play, helps foster deeper understanding, and won’t turn off the non-numerically proficient amongst us!

We’re talking about a 17% difference when a commander is rated as a 5 instead of a 6. Distilling a leader’s effect down to a number is already borderline idiotic, but it is in the service of a game so we accept it whole cloth. Were leaders ACTUALLY under 20% different from each other?

The answer is, of course, perhaps. What matters more is the context of how and why that rating was determined. Over the course of a game where players are expected to roll the dice 100 times and use that value, on its own (rare…if ever) then we’d only see this ratings difference in action roughly 9 times if players have a 50/50 chance of success on an unmodified die roll. I’m basing this on the following equation (100 chances x (.50 x .17))…so take that as you may since it’s essentially what is the probability of you NOT making the roll in 100 chances if you’re 17% deficient)

I go into this, essentially, to say that you need to do the following equation in your head before getting “wound up” as Jack Greene put it:

(Probability that the improper rating is actually breaking the game x Probability that your rage is misplaced)

As you might imagine, the actual results of this highly scientific formula and approach bear out what Jack Greene was saying. Effectively, there is no “perfect” system for ratings. The result is that people will find fault with individual values. There’s not stopping that part of a game’s critique.

Logic is how you Mean It!

Instead, the designer’s logic should be what demonstrates the validity of the rating. Jack Greene provided a few minutes of detail on how he developed gunnery ratings for Bismarck Second Edition. I had a few takeaways from that description:

  1. There is an immense amount of thought that a designer can invest in building the basic mathematical model underneath the game itself. When this is done well, it’s transparent and intuitive. When it’s noticeable…the designer probably needed some more time working through it with a developer.

  2. There is a strong parallel between the work of developing a rating and how SABRmetricians develop summary ratings like WAR for Major League Baseball players.

  3. No rating is immune from further tinkering. There comes a time when the rating and design are done, but that doesn’t come with any slammed shut dusty tome. Designers learn and evolve, so their logic has to stand the test of time for THAT design.

Designers need to make the case for their logic in the intuitiveness and broad acceptance of their conclusions. When that fails to happen, they should be including some kind of designer note that acknowledges a change from a widely (perhaps even incorrectly) held belief about the relative rating of a leader or unit.

Wrapping it Up

Designers are no infallible, but ratings are not mistakes. They must be meant when committed to cardboard. The work that goes into developing them is immense and fiercely opinionated. Consequently, players need to evaluate the logic of WHY and think about whether that one point different REALLY means anything.

In the long run, games are rarely made or broken by unit ratings. That’s why we see so many accepted combinations. There are more games with 3-3 or 4-4 units as the base value than I care to recall. However, what separates a good designer and game from a great designer and game is the logic behind the ratings. They need to mean something.