Jump to content
Jukey

Unit pricing algorithm (help!)

Recommended Posts

I'm trying to deconstruct units to find the point pricing, so I can create an algorithm for building new units, and at least get close to a reasonable price.

This is not my area of expertise, and so I'm wondering if anyone else has done this already.  It seems fairly straightforward until I get to the dial, where everything goes wonky.

The dial gets weird, since things fluctuate on usefulness depending on both sides of the dial.

If anyone has a better grasp of how this works, please help me out.

Share this post


Link to post
Share on other sites

It's more art than science. I really don't think you'll get a perfect algorithm, as there are simply too many variables. But if you really, really want to try, what I would do is determine what you think is the ideal "standard" command tool for infantry, for archers, for cavalry, and for heavy—a normal mix of turns/charges/initiatives/etc. For example, infantry would probably have, a move 1/2/3 on initatives 3/4/6 with the option to charge or turn-charge at -1, an attack on 4 with a +1 hit or +1 surge, a shift on initiative 6, and so forth. Then look at the proposed dials, and rate them on how much better or worse they are than the standard you've chosen, and adjust appropriately.

But as I said, no algorithm is going to be perfect, even if you managed to get the dials portion of the algorithm flawless. It can get you into the ballpark, but it's not going to be fine-tuned—that's where the art comes into play, and there's no getting around it. There are simply too many variables to consider, and you have to consider how each piece fits into the larger whole. For example, suppose you're designing a new archer unit, and you want it to have a 4x1 configuration. In this configuration, a training slot would probably add ~15% to the cost by itself, as this would allow you to get Rank Discipline. But if this unit, for some unholy reason, comes with precise 1, that training slot is far less valuable. Yes, you can still get Close Quarters Targeting, but that's actually less useful on this unit than it normally is: this unit loses threat quickly, and it's most likely to get hit by a combination late-turn charge followed by a regular melee attack, which in most units happens before missile fire. The unit is probably going to be down to threat two or below before CQT comes into play, and may even be eliminated.

It's weighing factors like this that brings the art into play. And my experience says that by the time you've gotten the unit working the way you want it to, you've got a decent feel for what it should cost, and only have to do a bit of fine-tuning to it.

 

 

Share this post


Link to post
Share on other sites

My feeling with this is that you establish benchmarks based on what other units cost and then, as @Xelto said, you have to apply the “art” from there as there’s simply too many variables that aren’t directly quantifiable.

Of course, if you choose a benchmark that’s over/under powered, then you’re stacking one problem on top of the other.  I think the community has identified which units are on the edges (or outside) of those boundaries.  In the end, you simply won’t get it exactly right for everyone’s tastes.  Look at the debate surrounding which units/upgrades were over/under powered...

Share this post


Link to post
Share on other sites

Definitely not a science, and the above-suggestions are a good template.  Part of the problem is that the game creators were learning about the game as they went, so in the early phases, it is likely that some units are costed too much and others not enough, but later units are often priced a bit more appropriately.

Where you'll have problems:

1.  Granularity:  The lower the price point, the more possibility that something gets off.   If a unit's real power value is 9.5, is that 9 or 10?  If you had 2000 points for a game instead of 200, you'd have an easy 95, and the rest of the system could account for that granularity, but when you're forced to round, you're got to get cheap and expensive outliers.

2.  At the extreme ends of abilities, it can become impossible to cost a unit successfully.  This is an extension of granularity, but is slightly different.  For example, imagine designing a siege that is worse than the Scion.  At some point, it is just so bad that it becomes unplayable, or it gets to a price-point that is spammable in a ridiculous way.  Or at the extreme end of good, you have a unit that is so powerful that it becomes an auto-include, or it effectively gets priced into a nice shelf-relic.   That's where the generic template with price point comes in as a useful mediation point.

3.  Will ceteris ever meet paribus?  That's the Latin for "with all things being equal."  We love to say that phrase so that we can eliminate variables and compare two items on supposedly equal footing.  Yet all things are almost never equal.  While there seems to be a basic idea behind the classes (infantry, cavalry, siege) and the races (Daqan, Uthuk, etc.), there's quite a bit of diversity going on.  This comes in the dice thrown, the mix of melee/range attacks, the movement on the dial, the stats on the card.  In fact, there are almost so many variables going on that I suspect the design process was more along the lines of:  2 trays of non-siege are going to be around 17-18 points, which we are then going to walk up or down based on how the dials/cards turn out.  That just seems like an awful lot of variables for an algorithm to capture.  The best bet will probably be to not let the algorithm act as a substitute for playtesting and thinking.  It can probably get you within 2 points, but the rest has to come from good judgment.

Share this post


Link to post
Share on other sites

Thanks for the replies!  This helps me a great deal, as I'm thinking I should focus more on what is in the actual unit for points, and let the dial develop more on a 'feels right' aspect.

For basic building blocks, I've been doing hp×armor + dice + unit abilities + natural skill (if applicable).  Costing dice at W=3, R=2, B=1 has been coming out fairly accurately.  Things like brutal and precise seem to be about 2-3 points, and unit abilities range from 3-8 depending on power and effectiveness.  Still working on what a slot is worth.  At a quick glance, it seems most slots are about 1 point, possibly with the exception of uniques.

In the end I think it will largely come down to putting the units on the table and costing them where they feel good, and less on a statistical spreadsheet.  

Share this post


Link to post
Share on other sites

I've got a working algorithm that is placing within 2-3 points proper values.  It gets weird with seige units, but everything else seems to be coming together proper.

I'll post the results once I run a few more units thru the deconstruction process.

Share this post


Link to post
Share on other sites

Things are not as cut and paste as I'd hoped, I can get Latari units to within 2 points of their actual cost using the system I made, but Waiqar consistantly gets overpriced by about 4-6 points on the same system. I have some reworking to do.

Dials must come into play a good deal, so I'm thinking there are aspects for scoring to dig out of the dial.  Like @Xelto said, there are certain initiatives for actions that appear standard, so when one is not in a regular position, it likely is modifying the overall points of the unit.

Share this post


Link to post
Share on other sites
24 minutes ago, Jukey said:

Things are not as cut and paste as I'd hoped, I can get Latari units to within 2 points of their actual cost using the system I made, but Waiqar consistantly gets overpriced by about 4-6 points on the same system. I have some reworking to do.

Especially since most Waiqar are overpriced to begin with.

Share this post


Link to post
Share on other sites
14 hours ago, Jukey said:

Things are not as cut and paste as I'd hoped, I can get Latari units to within 2 points of their actual cost using the system I made, but Waiqar consistantly gets overpriced by about 4-6 points on the same system. I have some reworking to do.

 

13 hours ago, Xelto said:

Especially since most Waiqar are overpriced to begin with.

My interpretation was that @Jukey found that according to his algorithm Waiqar was too expensive, and not too cheap? So that would be what @Xelto (and myself) expected?

Share this post


Link to post
Share on other sites
Posted (edited)

Runewars creators assume combo (waiqar unit + heros waiqar unit) > (others faction units + their heros units), so waiqar units recieve a cost penality to buy ? 

Perhaps there is a layout faction cost ? or just a wet finger ! (I mean loosely artistic estimation)

Edited by hakooh

Share this post


Link to post
Share on other sites

Based on my latest added variables Waiqar, Latari, and uthuk are coming within 1-5 points, daqan has some quirks of balancing on the small units but coming up way shorter of the points for the larger unit sizes.

This makes me think each faction has some 'faction' variables that apply to synergy between units.  There also seem to be some elements that work on a slight sliding scale.

This is the flat cost chart I have been working off of lately.  It's about 70% accurate.

Hp×Armor= x
Dice: W=3, R=2, B=1
Brutal=3
Precise=3
Protected=1 per
Lethal=0.5 per
Steadfast=2
Impact=1 per
Champion=2
Equipment=2
Music=1
Heraldry=2
Training=1
Seige=3
Unique=2
Artifact=3

This is where things get odd.  I listed the average action initiative, and added a plus or minus point if it varies from the average.
March i+/-=1 (i3)
Melee i+/-=1 (i4)
Range i+/-=1 (i5)

Armor up +1 seems to add a solid amount, this one might scale based on size.
Armor up=3 (right dial)

Surge effects vary based on ability, but seem to range between 2-6 depending on effect.

Card text ranges from 1-6 as the effect can vary as well.

I'm sure I'm missing a lot of big modifiers, like I said before, this is not my area of expertise,  but its proving to be a fun exercise, even if its a fools errand 😁

 

Share this post


Link to post
Share on other sites

It's a lot more complex than that.  Combos is where it's at, so a unit with a training slot and a equipment slot should be more than simply added the cost of one to the other.

Think of this:

Ardus 'Ex in the army with

6 trays of reanimates, with Death Caller, Support Carrion Lancer, Metered March, and Simultaneous Orders.

It's a 2 unit combo that's using 4 upgrade cards, all together.  It lets the unit initially active late in the turn (march on 6 or shift on 7), but stop just within range of the enemy (range 5), and then deliver a blight token, and then immediately remove that blight token to do wounds.  In the following rounds, that unit can sit put and continue to spit out blight = wounds.

While some combos are better than others, you need to look at how a unit will work within an army when assigning points.  You might have a crazy good synergy that requires a unit to be more expensive, not based on it's own merit, but on how in impacts the game over all.

On the topic of spam, I wouldn't make anything new under 14 points, no matter how bad it is.  Very few things in the game strike an area, so having units too cheap becomes a real problem.

Share this post


Link to post
Share on other sites

I think slots change price per faction, and likely scale with the unit.  I would guess it accounts for synergy in faction that way.  

I'm thinking my hpxA might be slanted too, I'm going to try subtracting tray number from hpxA to account for decreasing price for increasing trays.  This will allow for more fluctuation on slot pricing and dial manipulation.

If I can get an algorithm for each faction sorted out I'd say it's good enough to call a theory, even if it doesn't exactly work across all factions.

Share this post


Link to post
Share on other sites

The shape of the formation matters too.

A 3 wide 2 deep infantry block is better than a 2 wide 3 deep block.  You cannot simply assign point values to attributes and upgrades.

 

I think a better approach is look at existing costs for each unit option.

We have 4 shooting infantry and 5 fighting infantry.  

For fighting infantry, everything lands in the 16 to 21 point range for 2 trays; 26 to 32 for 4 trays, and 35 to 43 for 6 tray (all 3 wide and 2 deep).

 

The point ranges and variation in units is pretty small actually.  I'd say have everything new land in the range of existing point values is a really good start.  If anything seems like it should be cheaper; it's most likely not right for the game.  If a unit has so many abilities piled on it needs to be above this range, it is likewise not right for the game.

Share this post


Link to post
Share on other sites

Thinking about such things also melts my brain a little (game balance is probably the hardest part of game design as every new element added to a game exponentially increases the variables based on player choice, luck and so on) but part of it may be thinking about what armies players may wish to put together and what is a good balancer in terms of armies having units that are balanced versus the units other armies may have in play. It's often a bit like a complex paper-scissors-stone in that certain units are usually better suited to taking on other units, although other factors can come in to play such as fighting in numbers to take down or outsmart a major unit and how you approach it's challenge etc. Heard that one game (can't remember the name) has a mechanic where you can draft units during a battle, players may wish to experiment for this in RWM, if unit options are expanded it could bring these strategic choices of which units to field and where into the combat, but it might also prevent the blind pick aspect that you might lose just based on which units you chose if you got to the point of having lots of possible unit choices, if you could make it work the in-battle drafting means players can try to respond to what the opponent throws at them, would change up the combat so may be it's an optional way of playing, but some other things to think about.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...