In their book Game Design [Volume 1: Theory and Practice] (published by Steve Jackson Games, 1981), game designers Nick Schuster and Steve Jackson discuss wargame design with considerable insight.
The sections in non-italic type are the words of Mr. Schuster and Mr. Jackson. My commentary follows in italic type. –Alan Emrich
Playtesting is the process of playing a new game design, over and over, in order to spot flaws and improve playability. The importance of playtesting in the production of a finished game design cannot be overstated. Ninety percent of all game defects could have been corrected by satisfactory playtesting. There is no excuse for failure to playtest a design thoroughly before putting it on the market [although you’ll hear a lot of such excuses from publishers, particularly software publishers]; it indicates either gross ignorance, total egotism, or an absolute lack of interest in giving the gamer their money’s worth.
Good playtesting takes time and goes through several stages. But, like research, it is an unavoidable part of the design process. A company or designer that is unwilling to playtest games should be in another line of work - it’s that simple.
Every professional designer and company uses the techniques used below, with variations. The advanced stages are often skipped - to the detriment of the final product - bit it all starts with the designer’s own testing.
Designer Playtesting (or "Gross Playtesting" - as in "Gross Anatomy")
This is the first stage of playtesting. You’ve fog a basic set of rules (typed), a map, counters, and whatever else you need. Fid one friend (or however many it takes for the game) and play with them. At this stage, you should not yet be relying on the rules. Just tell them how to play. Refer to the rules as necessary, but very instructions are a big time-saver. At this stage, you’re testing the systems, not the actual text of the rules.
Pay as any times as you can. Whenever a problem crops up, make a note of it. Unless you run into a really bad glitch [e.g., a ‘show-stopper’], it’s best to finished each game with the rules your started with rather than changing them mid-game. After each game, discuss the notes and decide what rule changes re necessary for the next game.
In playtesting, you are trying to build a game that meets several criteria:
1. Balance. All players should have an equal chance to win. Where one side must inevitably be wiped our or driven away, use a “Victory Point” system to allow a good player to claim victory when his forces managed a good showing in defeat.
2. Variety. The game should not be predictable. The longer it takes for the players to optimize their game strategy, the better. If you find that there is one best way to win, seriously consider changing the rules. Otherwise, your final product may be boring.
3. Realism. As a rule, a game becomes less playable as it becomes more realistic. The type of game will determine the amount of realism your gamers will require (or tolerate). In general, historical games are expected to be more complex, and therefore more realistic, then small ones. A game that plays quickly and well can get away with less realism than one whose mechanics are cumbersome.
By the playtest stage, you will already have decided how much realism you want; your research and original drafts will have been appropriately detailed. Playtesting will tell you (a) whether your attempts at realism have make the rules too cumbersome, and (b) whether the game results are ‘realistic.’
An example from my own experience: the counter values, combat rules, etc. of my One-Page Bulge game have very little detail - they are not highly realistic. This was deliberate, for simplicity’s sake. My objective was to write simple rules that nevertheless interacted to give a course of play similar to that of the actual Ardennes Offensive [by the German Army in 1944].
Playtesting of the first version proved this was not working out. Therefore, the rules were changed - not made more complex, just changed - and tested again. Eventually, a game was achieved in which the use of historical strategies will give very nearly historical results. It is therefore reasonable to assume that changed strategies will mirror the ‘historical’ effec5ts of those strategies on the battlefield, which is what simulation is all about. And extensive playtesting was necessary to achieve this.
4. Playability. This is an omnibus term, taking in many things: rules clarity, speed and ease of play, “interesting-ness” of each turn and the game as a whole, and the whole gestalt of the game. The nastiest thing you can say about a game is that it’s unplayable. An unplayable game is a failure, pure and simple.
Rules clarity will be checked in the blind testing stage, but start on it now. Any time your first playtesters tell you they can’t understand what they’ve read - listen!
Speed and ease of play will become quickly apparent. If you spend more time looking up rules and doing bookkeeping than you do moving counters - if you have to move lots of counter in very predictable fashion - if information is not centralized and cross-referenced - people are less likely to play the game.
As for interest: the harder it is for you to keep your playtester playing, the less interesting your game is!
A buyer usually expects playability in inverse proportion to realism. Therefore, fantasy and SF titles are expected to be more playable than historical titles, and small games will usually be more playable than large ones. Even its designers admit that the historical monster-size game Campaign for North Africa is not meant to be played (just admired for its research) - but people buy it anyway.
If you want a successful game, give your buyer at least as much realism as he expects from that ‘class’ of game and lots more playability - or vice-verse - and that player will love you.
Blind Testing
When you feel that you have the rules the way you want them, and that the game works, you’re ready for the second stage of playtest: blind testing. The essence of blind testing is that new playtesters are exposed to the game without the benefit of advice from the designer or their experienced players.
The purest form is blind testing by mail. Send off copies of the rules and all other materials to friends (game clubs, etc.) in other parts of the country. Ask them to play as many times as they can and to send their results, rules questions, and general reactions back to you.
A modified blind test can be achieved by recruiting some new local playtesters, handing them the game, and watching… quietly!. Simply explain that you want to see whether the rules are good enough for them to play the game without your help. You can learn more from watching a blind test session than you can by mail, but it’s hard to resist the temptation to explain things, correct illegal play, or bawl players out for misreading something you thought was totally clear.
Whichever way you do it, blind testing is necessary. [In fact, blind testing is absolutely vital and is the very essence of ‘critical’ playtesting.] It will tell you which of your rules are ambiguous or unnecessary. It may also reveal some strategies you hadn’t considered, or (if realism is a criterion) bring up problems there. Strangely, some of the biggest companies don’t blind test much, if at all (or don’t listen to the blind testers!) You can tell which ones just by reading the rules they print; they’re the ones that don’t quite work and raise questions that they fail to answer.
Proof Playtesting
This is the last stage of playtest - the stage that is omitted by almost every company and that would eliminate 95% of the ‘errata’ in the hobby if manufacturers would just take the time.
Unfortunately, time is money - and the very nature of proof playtesting requires an extra delay of two weeks to a month before the game is published.
Proof playtesting is an extra stage of blind testing and another round of designer playtesting using only ‘proof copies’ of the finished rules, maps, and other game components. In other words, once the game is totally ready to print, and all of the final art and layout is done -once everything is ready for the printers -you stop, make a dozen copies, and playtest it again.
If you’ve done everything right so far, there will be no design flaws left at this stage. What you’ll catch with proof playtest are the killer typographical and layout errors: a location left off the map, a line left off a chart, a paragraph left out of the rules - or just a number typed in wrong, messing up your setup rules or your victory conditions. These small errors can drive players mad - and more proofreading won’t find them all! Proofreaders are good at finding misspelled words and sentences that end in the middle of an idea. They’re not good at noticing total omissions. You, the designer, might notice an omission… but by now you’re so familiar with the game that you see what you expect to see [whether it’s actually there or not].
Proof playtesting is the answer. Make those proof copies and turn some new play testers loose on them. If your basic design is good, they’ll be playing just as though they had bought the game in a store. And when (not if, but when) they run into a rules typo, you’ll still have time to fix it. No player frustrations, and less expensive, embarrassing errata.
Special Problems
The easiest game to playtest is a game where both sides have identical forces, attacking on a symmetrical map. More complicated setups provide more richness - and correspondingly more playtest is required. Some things to watch for:
Building scenarios. If players may choose their own beginning forces, or build reinforcements, according to some kind of ‘point system,’ you must test every possible combination of forces they can achieve. Some types of units become disproportionately power in large numbers. You can control this by making units more expensive, changing their abilities, or just limiting the number of counters available.
Monster games. If a game takes two full days to play through, you know you won’t get through many playtests. Make the most of the ones you have!
Multiplayer diplomatic games. Such games can be self-balancing, in that the weaker players will tend to combine against the strongest. Just make sure they’re not too balanced and that someone can eventually win. Cosmic Encounter offers good solutions to this problem.
RPG supplements. The bigger the role-playing game, and the more material that is already available for it, the more closely a supplement must be checked. A Dungeons & Dragons supplement ought to be compatible with literally millions of words of existing game material.
Refereed games. If solution of some problems is left to the referee, you should playtest with many different referees to see whether things can get out of hand under some interpretations of the rules.
Playtest Techniques
Several techniques are especially valuable, both in designer playtesting and in blind testing. The designer should try all of these things, but blind testers should be encouraged to try them, too. Show them a copy of this article…
Try dumb strategies. This is my biggest single piece of playtest advice; I’ve been saying it for years and I’ll say it for years more. Just becomes something seems ridiculous in real life, don’t assume it won’t work in your game. Try it! If you Civil War simulation constantly lets infantrymen charge uphill and take entrenched positions from equal forces, something is terribly wrong. A good civil War gamer would never try that at all - so make sure you have a couple of ignorant (or at least patient) playtester to help you.
Idiots make good playtesters. Some of my best playtesters have been people I would never play with for pleasure. They nit-pick at rules, find impossible meanings in simple statements, botch setups. A couple of them have been cheaters. If people like this can play a gamer properly, without driving the others at the table totally up the wall - that game is probably airtight.
Geniuses are helpful, too. If you know any true experts in the field your gaming (even if they don’t play games themselves), ask for their help. They’ll be complimented; they’ll also be useful. And a hard-core, full-time, fanatic gamer is the best playtester you can get. He’ll play for blood, wringing every possible advantage out of the rules. And, unlike the idiot, he’ll have some constructive suggestions to make after he’s through.
You can never playtest too many times. I think one of the reasons that micro-sized games became popular [during the early 1980s] was that, being small and quick, they had often received more testing than larger games. And, as a rule, more testing means a better game. Not always! I know of a tremendously popular game that was never played by anyone but the designer until it had been published. (Sure, it’s got problems, but people still play it and like it.) Playtest a game until you drop.
Keep up with your changes. When you find an unplayable rule, note it on the rule manuscript. When your manuscript becomes illegible, update it… as often as necessary. If you fail in this, your playtesters will be playing a game that has nothing to do with the current rules draft you’re working from.
Blind testing will catch this, if you blind test. But I’ve read of one game project that totally bombed because the designer failed to notice that the playtesters had changed all his rules as they went along. The playtesters had a pretty good game worked out, but the designer never found out what it was… and the designer’s version was published and fell flat.
Playtest Credits
I hope I’ve convinced you that playtesters perform an invaluable service, and are invaluable aids to a professional designer. As such, they deserve appropriate recognition and compensation. Any players who provide significant input should be listed in the game credits and should receive a free copy of the finished game. They’ve earned it.
