Creating points values

[Ivan Sorensen]

This is NOT about whether points systems in games are the true mathematical salvation from chaos or fascist oppression of the free gaming spirit.

Rather, it’s about how values are derived.

Some games use a formula, some use eye-balled values.
Some are very approximate, some try to account for every factor.
Some are constant (the same gun always costs the same), some are relative (the gun depends on the unit using it).
Some are intended to be an integral portion of game play, some are there as an afterthought.

Share your thoughts. I’ll elaborate on some of mine later.

[Angel Barracks]

When I did mine I came up with a randon value for a basic umarmed grunt level human.
I then decided that a +1 stat would be worth X more.
etc

Then I split it into 2 camps, infantry and vehicles.

Then infantry ranges for weapons, infantry damage for weapons.
Each 10cm of range was worth 1pt.
Each dice of damage was worth double the previous number, with 1 die being 1, so
1 damage = 1pt
2 = 2
3 = 4
4 = 8 etc

Then special abilities for weapons added or subtracted a fixed value of 1pt.
So hand held miniguns can’t be aimed so -1 pt.
RDF assault rifles have Chayonets +1 pt

Then I did a similar thing for vehicles, but the values were bigger.

It is all very crude, but my workings can be seen on the website HERE

[John McBride]

PRIDE OF LIONS gives each unit a die level, from D20 (strongest) to D6.  This is used in combat and also in morale, and is the most important stat. Movement rate and missile ability are also important. So the basic equation is die level x2 plus bonus points for faster movement and for shooting ability. So halfling militia with pitchforks and sling (D6 plus another D6 missile attack) would be something like 18 points per unit, while dark elven lancers riding raptors (d20 and fast) would be something like 20 x 2 plus 16 for the speed = 56. Then there’s added points for things like FEARSOME or drilled or such, as well as point reduction s for things like Vulnerable to Shooting (unarmored elephants, for example). All rather arbitrary and requiring fiddling to balance — and even then susceptible to minimax lawyering.

It really just amounts to making a list of variables in combat performance, decided which are more important and then weighting them, and coming up with some simple equation. And being aware of the assumptions made and emphasizing the limitations of the resulting system.

PLAYTEST, playtest, playtest.

[Ivan Sorensen]

So one thing that you both kind of touch on, is whether costs should be relative or not.

If I give a unit an anti-tank weapon, but they are cowards who can’t shoot straight, that weapon is less valuable than if the same weapon is in hte hands of battle hardened veterans with ice for blood.

That’s really hard to do in a way thats easy to do quickly though.

When I did FAD, troop quality and unit traits were multipliers to the basic cost of weapon+armour. Worked well as multiple abilities would compound very quickly, but it also meant it was a bit of an awkward, math heavy process.

Old Rogue Trader had a simple multiplier that kicked in, if the cost of your figure exceeded a certain amount.

edit:
Points costs can also be used to influence players. If you want to discourage something, make it a bit more expensive. If the norm is one heavy weapon per squad, add a small premium for using two for example).

[Angel Barracks]

Yeah, it is tricky.
As noted in the other topic, I only really did them to help with my campaign games, the system is broken and you can have equal points but unequal forces.

You could have say 200 points of men with pistols vs. 1 tank also worth 200 points.
The men with pistols are not able to destroy the tank, as they have just pistols, however the tank can destroy all the men.

But the scenario may call for 3 objectives to be held at once.
As the tank can not be in 3 places at once it can not win the game despite being indestructible.

I think it shows how important a well thought out scenario should be.

[McLaddie]

I often get the impression that the reasons for points systems often gets lost in the juggling of numbers.  The points can be an attempt to produce ‘balanced’ games in whatever measurement the designer uses for one-off games or tournament play and/or an attempt to contain the players to ‘historical’ or reasonable OOBs.  Those two purposes can create very different approaches to points systems.  I’m thinking of strict points systems where each regular infantry stand is two points, particular leaders 36 points with limits of 300 point armies to the other extreme where specific unit OOBs are given with possible support units costing so much.

Some games use a formula, some use eye-balled values.
Some are very approximate, some try to account for every factor.
Some are constant (the same gun always costs the same), some are relative (the gun depends on the unit using it).
Some are intended to be an integral portion of game play, some are there as an afterthought.

These choices strike me as going for balanced games and/or tournament play.  Of course, the premise is that the points relate closely to the ‘value’ of the unit’s benefits and weaknesses in game play.  I sometimes find that there was never any testing of that premise with the point system.  It can be part of the reason that a ‘super mix’ of units within the points system can be created.

[John McBride]

We have played hundreds of games of PRIDE, but we are STILL occasionally finding situations in which the rules need to be adjusted, or match-ups that yield unacceptable outcomes. It is because there are simply too many factors to ever playtest every value of every variable in relation to every other variable. I think this is especially true in fantasy and likely in sci-fi, since one is not bound by the realities of our own real world (things like gravity! or human nature!)

So points systems are never going to be more than approximate.

Playtest, playtest, playtest.

[McLaddie]

Playtest, playtest, playtest.

Yep, and then playtest some more…  

[Bandit]

BattleTech really struggled with this. Early on players used tonnage since each ‘Mech was a given weight and since different weapons and armor and such weighed various amounts the notion was that two 50 ton ‘Mechs were nominally equal.

Later this was replaced with Battle Values. A Battle Value was assigned to each component and that value of the ‘Mech was essentially their sum.

Yet later this was radically revised to Battle Value 2 because the original didn’t do much to accomplish its goal.

I use this as an example because it is easy, in short I think that point systems are really hard to get right and that games that have a lesser number of inputs into the point valuation (BattleTech had LOTS) are easier than those with many.

Cheers,

The Bandit

[Ivan Sorensen]

Yeah, trying to use the points system in Battletech was pretty gruelling.
When we played in like 9th grade,we took one attempt, realized it was far above our heads, then went back to using tonnage and just accepting it’d be broken sometimes 🙂

[Ivan Sorensen]

John McBride wrote:

<p data-wr_replaced=”true”>We have played hundreds of games of PRIDE, but we are STILL occasionally finding situations in which the rules need to be adjusted, or match-ups that yield unacceptable outcomes. It is because there are simply too many factors to ever playtest every value of every variable in relation to every other variable. I think this is especially true in fantasy and likely in sci-fi, since one is not bound by the realities of our own real world (things like gravity! or human nature!)
<p data-wr_replaced=”true”>So points systems are never going to be more than approximate.
<p data-wr_replaced=”true”>Playtest, playtest, playtest.

Playtest a lot. Then give the game to another group and realize it’s all unbalanced because their gaming table is bigger, they only play close combat armies and they use less terrain than you 🙂

Start over.
Tear hair out.

🙂

[Phil Dutré]

There are really two issues to consider when computing point values:

1. How much is a single unit worth?

This can be rather easily determined by consider a “standard” unit, give that standard unit a nominal point value of 10, and them compute (probability theory) how much units it would take to achieve a draw in damage inflicted on a set of standard units. When considering pure damage, this is not too comlplicated. Complications arise when special abilities are considered that are only have an indirect effect on damage.

2. How to put together an army?

Usually, a linear sum of point values of individual units is considered. E.g. you have an army of 200 points, that means you can pick 10 units of 5 points each and another 10 units of 15 points.

BUT! This assumes that point values of multiple units adhere to a linear model. Often, this is not the case. The underlying gaming engine often does not behave linearly, hence 2 units of 10 points are not the same as 1 unit of 20 points, although both units behave correctly vs the standard unit as computed under step 1.

The Lanchester combat model illustrates this nicely — the strength of a number of units (under certain assumptions) equals not its number but the square of its number.

 

I once had the idea of having players bid for units. E.g. many different units with different capabilities are available, give players a budget, let them bid, and after a whole series of games a “fair price point” for every unit would be the result. However, this would take a lot of bidding phases and games, but it could converge quite quickly. Collectible card games such as Magic The Gathering have shown that this can work in practice.

[John McBride]

Yes, non-linear. Skirmish units in PRIDE are extremely important, except when they are not. Opposing skirmishers must match up and more or less offset one another. But their movement makes them VERY flexible — they can shoot gaps between units and get into the enemy rear (which I allow mainly as the mechanism to encourage generals to keep a coherent battle line) and one skirmisher unopposed and in position to hit an enemy rear while it is engaged to the front is a game winner. But such units are only effective under the right circumstances. And their power is more a function of movement than of combat power which is generally relatively low. And like mages and spirits, they are really only very effective if one side has more than the other. So you can spend a lot of points to little effect except just to stay even.

[Phil Dutré]

Another issue with determining point values that is often conveniently forgotten is the non-transitivitiy of the “being stronger than”-relation. The classic example is rock-papers-scissors. What is the point value of rock? paper? scissors?

One way to look at the problem is saying that when given a random choice, each option is as likely to win. Hence, all choices have equal point values.

However, once the choices have been made, the probability of winning is not equal anymore. If I pick rock for 1 point, and you pick scissors for 1 point, the game is balanced in my favour. Whatever you do, you can’t win anymore. So the “fairness” of the point system in this game lies in the selection of forces, not in the gameplay itself. The counterargument is that over the course of many games, everything will average out – and that is correct! But this does not imply “fairness” when playing a single game. A variant of this counterargument is that when one has to choose a large force of 100 units (each being rock, paper, or scissors), the total battle (consisting of 100 individual skirmishes) will balance out. But even then forces of 100 units will not be equal. E.g. an army of 100 rocks (100 points) will still beat an army of 100 scissors. However, it will draw against an army that has an equal proportion of all three.

Translated to the wargaming table: are points determined to provide a fair balance during troop selection, or during gameplay? There is a difference … (Cfr. also Magic The Gathering. Is the winner determined during deckbuilding? Or during actual gameplay?)

And things get more complicated if in your made-up fantasy world there is an abundance of stones and a scarcity of scissors… 😉

 

As an amusing additional example, consider the following six-sided dice, each with different numbers:

Dice A: 3 3 3 3 3 3
Dice B: 2 2 2 2 6 6
Dice C: 1 1 1 5 5 5
Dice D: 0 0 4 4 4 4

If you do the math (highest roll wins), then A beats B with probability 66.6%; B beats C; C beats D; and D beats A. What point values would you assign to each die?

This is not to say that point values do not have meaning (it is not my preferred style of wargaming, but that is another discussion 🙂 ). But, it is important to realize that points do no necessarily imply that both forces have an equal chance of winning the game.

[John McBride]

Phil, those dice are cool!

Someone mentioned random armies, which can be a lot of fun. We used to do that with JOHNNY REB: each side got a certain number of regiments and batteries, then diced for size and weapon type and basic morale point. Games were ROUGHLY even, but lots of devils in those details. Combined with hidden unit markers and visibility, made for some great games.

[Bandit]

Phil Dutré wrote:

One way to look at the problem is saying that when given a random choice, each option is as likely to win. Hence, all choices have equal point values.

…

However, once the choices have been made, the probability of winning is not equal anymore.

I think the reason to conclude the point values are equal is because the choices are made blind.

From an all knowing perspective in a scissors vs rock perspective the valuation changes to:

Rock 1
Scissors 0

Due to the impossibility of scissors to win as you say. But since the choices are made in the blind with no notion of what is being confronted, the equal valuation is sensical.

This can be a problem with systems that apply essentially a rock, paper, scissors arrangement where some troop types can only win against given opponents and others can only lose. Their respective valuation becomes more a question of the means the players have of setting up the combinations they seek to.

Cheers,

The Bandit

[Author]

Posts