Skip to Content

Can Design Decisions be made without having to Test?

5 replies [Last post]
Joined: 12/31/1969

Continuing with my chariot racing obsession, I have been looking at a bunch of racing games to give me some ideas. I stumbled across a interesting genre of games called "replay games." Researching further, these games are typically solo sport game simulations based on actual player or horse statistics.

I looked specifically at two of Terry Goodchild's games: "Metric Mile" and the "Sport of Kings." Both of these games are not strictly replay games, and thus are tailored for having multiple players interact together.

The way Terry has designed these games is truly brillant from what I have read (I recently just ordered these games and will soon be playing them).

Anyway, from what I can gather of the rules, it appears that in both games the sprinter or the horse has running modes that use up a finite amount of stamina each sprinter or horse is given. Obviously, the faster mode you go, the more stamina is used up. This is a gross simplicfication with many factors left out, but this gives an idea of the chief mechanic of the games.

For the "Metric Mile", there is five modes from "ease off" to "sprint" while the "Sport of Kings" has only "normal" and "under pressure" modes. In both cases, it sounds as if you go to the highest mode (under pressure or sprint), you must stay in that mode until you run out of stamina.

This rule then encourages players, if they decide to go at the fastest mode, to only do so towards the end of the game (if you do it too soon you are guaranteed to run out of stamina and become exhausted as you must stay in the fastest mode until you run out of stamina).

Obviously this lends to interesting decisions of if you should go the fastest mode and if you decide to do so then when; you've got only one opportunity to do so.

Something irks me a little bit about this entire idea from a realism and game design standpoint. I don't know about athletes, but with horses if you go at their ultimate max speed, they can only maintain that speed for three furlongs at the very most. If you examine the throughbred world records and plot the distance versus time of a race for all recorded distances, you have an almost linear relationship, except for the fact that there is a big drop in feet per second (fps) from 3 furlongs to 3.5 furlongs (3 fps). Note I am making the assumption that in both races, all the horses are going at a sprint pace right from the beginning of the race.

It appears that with Goodchild's design, you may opt to sprint to begin with. If you did so, you would travel a much further relative distance to three furlongs before being tired. In horse races, different horses break at different parts of the race, sometimes going at their top speed, but these durations are short and the horses are not immediately exhausted after going at these speeds.

The problem from a gaming standpoint is why only allow a horse team to sprint towards the end?

To change this would mean that there would have to be a rule change that limits the player to only use this mode once, but only for a certain time or use of stamina.

A solution I've come up with, is to have special "reserve stamina" that can be used for all modes while the "normal" stamina can not be used to sprint. Once a player does go to sprint mode, all of the reserve stamina must be used up. This now limits how long a horse can go and allows the player to sprint at any point in the race.

Sounds good, right? But between my idea and Terry's idea, how can one decide between the two? Must one always playtest to find out, or can one use gaming design principles to decide between the two. My method, I believe, adds a bit of complication for the added bonus of adding extra decisions and realism.

This is my dilemma these days, not just coming up with a mechanic that is simple and seems to work, but being able to analyze the mechanic from a decision/fun design standpoint.

Any input would be appreciated.


Joined: 12/31/1969
Can Design Decisions be made without having to Test?

I think you can make design decisions without having to test them first.

it a matter of asking yourself questions

Below is a couple of questions that I have asked myself while desiging Runic that apply to your issue

A) Does this add an element to the game that it would be otherwise incomplete without, or in simpler terms is it nessasarry or whould I just like to have it.

B) Does it add an unnessasary amount of complexity to an issue ie. sacrificing fun for realisim

as always this can only get you so far and while its possible to make desisns this way testing will always be better

Zzzzz's picture
Joined: 06/20/2008
Can Design Decisions be made without having to Test?


I want to tell you yes, but I really feel it all depends on the change being made.

As with any project, it is always a good thing to test out changes, you just never know. But even testing does not guarantee that the game is not broken. That is why there is errata for so many games these days.

I think you might be able to combined the two, jot down the changes in design, analyze the new alterations, removing those that dont stand up to the design criteria you set. Once you get down to a few options that you really like, I think you need to go back to playtesting.

Joined: 08/03/2008
Can Design Decisions be made without having to Test?

I agree with the others, that it's almost certainly a highly situational question. I've found it's very easy to generate a very long list of "it would be nice to try this" items for any game. Between one's own crazy ideas, and suggestions made by playtesters, it's easy for this list to grow quite long. When approached scientifically, in which each test represents only one substantive change, it could easily take many years to try out all the permutations.

A few guideposts I use to rein in this sensibility. The key overarching principal, which I admit I've had to learn from Scurra, is that you can't let the perfect be the enemy of the good. If you've got a good, fun game, trying to create the absolute optimum superlative great uber-game that is perfect in every way may be overreaching and unnecessary; why not just play your good, fun game and be happy?

I also tend not to be shy about making several changes all at once; testing only one variable at a time is just too slow, and the overall impact of one variable may be too slight; or, conversely, the impact may be too great, and other parameters must be changed to compensate.

I think all changes I make are typically evaluated with an eye towards what about the change clearly will result in a better game. If there's a change that will make the game different, but it's not obvious that it will be better, it's maybe worth giving it a solo playtest to try it out, but it's not, to me, the end of the world if I don't evaluate all of those possibilities.

A good example is my Disciples game, in which there is a deck of cards, each of which is tied to a specific location and allows an action to be taken at that location. One of my playtesters suggeted that the game could instead work by having some randomizer determine where these cards would go (the idea being, presumably, that there would be more game-to-game variability). Now, this could probably be done, and would make the game different. But it would require changing so many other things that I haven't tried it; it's just not clear how it would make the decisions more interesting, and that, I think, is what ultimately matters.

So, I think it's not always easy to evaluate which changes will make a game better, but I don't really feel it's necessary to try every permutation of everything for the sake of trying it. Just pick something and go with it; if it works, don't look back!


sedjtroll's picture
Joined: 07/21/2008
Can Design Decisions be made without having to Test?

I think it's definitely possible, and perhaps desireable, to make such decisions without haveing to test. Obviously in some cases you cannot reasonably make the decision, and other times you can but you'll be wrong, but a lot of tweaking can be thought through and resolved as a series of simply thinking about it, running through sample turns in your head, and talking it over with people.

- Seth

Scurra's picture
Joined: 09/11/2008
Can Design Decisions be made without having to Test?

I'll echo Seth here: sometimes it is clearly unnecessary to run a practical test on one individual mechanic tweak when the theoretical runs* show it to be fine. That's not to say that theoretical tests are fool-proof, since they clearly aren't, but they save you a lot of time.

(*NB in my terminology here, theoretical tests aren't quite the same as solo play-tests. A theoretical test is saying "rolling 2d6 will on average give me a one-in-six chance of rolling a 7, which is more reliable than using 1d6", whereas a solo play-test will show you that in real life this doesn't happen ;-))

Syndicate content

forum | by Dr. Radut