7th Ed. Standardized Comp Score

[Barotok]

Call me stupid if you want, I guess I am missing the point of you quote a previous post in the thread with no comments. I can see how both posts could be read as supporting two opposite opinions, but am unsure if that is the point you are trying to make. Care to clairfy?

 

[Caneghem]

"I run IE and have never had that problem despite the average length of most of my posts."

It only started when I switched to Windows Vista, which came with my new comp. I had no option of upgrading to XP, which some computer companies offer. It strikes me as funny how you have to pay more for the older OS.

 

[Caneghem]

Still in thought over this overall concept which perhaps should be renamed to "Rebalancing Warhammer"?

All joking aside, I have been trying to come up with a logical way to determine a unit's DPR(damage/round) in CC, pulling from a wide range of ideas. Some have been testing their combat skills against an enemy of equal stat lines (itself), but this does not generate varied values that seem like will be necessary to differentiate units.

Rebalancing Warhammer sounds fair enough.

As to your idea of creating a single test dummy for all units to batter against... this will inevitably favor certain kinds of attacks. And really.. even picking just 10 would do some units a disservice.

On the other end of the scale, we could just pit everyone against everyone. You could calculate average combat result, and thereby completely account for all offensive and defensive abilities.

You would run three "combats" against each target, and this will be easy to do once the program is written. First, resolve as though you charged the enemy, second, as though you were charged, and finally, as though the combat is in the 2nd round or later, where initiative matters. Against each target, you could compare movement. If your movement is lower than the target unit, assume that they will charge you (all else being equal, this is the reality of things). So you just figure out the average result of "they charge" and "later rounds of combat". Equal speed averages all three trials. Greater speed assumes "you charge". Doing it this way would handle the relative importance and probability of charging for every army, and includes things like hatred and ASF since you are just evaluating what would happen. The downside is you would need to make it dynamically able to update all values as new army books are added. Units with multiple options as to how they want to fight (spears or HW and shields?) will simply have their best combat result calculated for each option, and the best result is kept.

The beauty of resolving entire rounds is it also weighs the value of your defenses. Combat result can be positive or negative for each unit, so you would get to the end and have the program total up the final combat result against everything. Won't it be interesting to find out which is the best unit and which is the best bargain for the points? This will lead to a "ranking list" of all units in the game as well, and you can see where your favored fighters fall. This could be a fun "first project", and covers a good chunk of the basic stats and a sizable portion of the special rules.

 

[Barotok]

I would advise strongly against writing a program that uses an actual RNG (random number generator) to calculate combats. It seems that wit has many as you are planning to 'simulate' you'll end up with a few very wierd results (a goblin beating a sarus, comes to mind). Perhaps you meant this in your post and I was unable to pick it up through reading. Naturally I would assume that you meant to run these 'combats' based on average results and not actual die rolling, but just wanted to make sure I'm clear.

I've also still been thinking about how stat lines relate to one another in order to figure out a way to standardize a troop type's strength. Let me see if I can word this so it is clear what I'm talking about. It's still difficult for me to wrap my mind around.

The numbers don't really indicate how strong the unit is. As an example a weapon skill of 5 is regarded as a high weapon skill but has no effect on enemy troops when they are attacking you unless they are higher than weapon skill 5 or lower than weapon skill 3. This means that versus a WS5 target it really only affects the combat either way when the opposing model is weapon skill <2 or >6. Three, four and five all generate the same results as far as hits.

This lead me to an epiphany, if you will. The stat numbers on a profile are not what determines a unit's strength, the differences between two units stat lines determine how much better or worse they are. I know that this sounds like you're just saying the same thing 'backwards' nad it probably is, but it helps to relate those different numbers together easier in my mind. Perhaps someone with a greater understanding of math/probability will be able to explain why this matters.

Another example I can share is this: My troops strength is 7. If I am fighting a toughness 7 model, I'll wound 50% of the time on average. This is very even odds. However, when fighting a toughness 5 model my odds get considerably better. The point is, that the odds of wounding don't change after the enemy model's toughness falls to 5 of below, you can't get any better odds.

To me, this means that each troop type should be weighed against all other troop types that can fill the same role. This is the minimum. I think that weighing every troop type against every available troop type in the game will just take too much time and generate a lot of unneeded calculations. I'd say, stick with infnatry vs. infantry, cavalry vs. cavalry, fast cavalry vs. fast cavalry, monster (40mm) vs. monster, etc. This should render some solid starting results that you can then decide if more calculations are needed.

Perhaps another way to classify units is in terms of how much damage they do on the charge (average), how many casualties they are prone to take when they accept a charge (average) and how much damage/casualties they take during a round of static combat (when neither has charged). I know that I am reiterating your previous post but when doing your calculations based on the context set forth here, perhaps this gives us a better way to interpret and classify the numbers?

 

[Caneghem]

Yeah, no random number generators. Statistical averages are the only thing we can really use.

"To me, this means that each troop type should be weighed against all other troop types that can fill the same role. This is the minimum. I think that weighing every troop type against every available troop type in the game will just take too much time and generate a lot of unneeded calculations. I'd say, stick with infnatry vs. infantry, cavalry vs. cavalry, fast cavalry vs. fast cavalry, monster (40mm) vs. monster, etc. This should render some solid starting results that you can then decide if more calculations are needed."

Weighted averages are the way to go. It might be useful to attempt to "classify" every unit in the game, so that you can account for the fact that certain unit types will spend more time attacking certain other unit types. Heavy Infantry, Light Infantry, support, heavy cav, fast cav, flyers, monsters, etc. etc. I would say still "run the numbers" against each category.. but depending on the unit you are evaluating you would put a higher weight on things to reflect actual usage.

For example, say we are evaluating a heavy cav unit. You'd figure out a reasonable "heavy cav template" to be used by all heavy cavalry. So it might look something like %40 hvy inf %30 hvy cav 15% light inf 15% support. When finding a final average performance, you'd be placing more emphasis on how the cav performed against the heavy infantry, since that is a typical usage of heavy cav wheras the light infantry and support get a much lower weighting because they typically will flee combat rather than be destroyed, so those averages aren't as important. Obviously this is a cursory "template", but finding a good template for each type would be important.

"Perhaps another way to classify units is in terms of how much damage they do on the charge (average), how many casualties they are prone to take when they accept a charge (average) and how much damage/casualties they take during a round of static combat (when neither has charged). I know that I am reiterating your previous post but when doing your calculations based on the context set forth here, perhaps this gives us a better way to interpret and classify the numbers?"

I think you were misinterpreting when I was saying "combat result". Those are the three types of combats I want to evaluate using average numbers of course. But I'm not stopping at how much damage is done on the charge. You might as well evaluate the net wounds of the encounter... e.g. Heavy cav charges heavy infantry doing 2.4 wounds on average.. we will assume all infantry ranked up 5 across so 2.6 attacks back and let's say that produces .2 wounds on average. Net result is 2.2 in favor of the cav in terms of kills. This isn't really about figuring out who wins combat, but rather about evaluating the offense and defense at the same time.

Now when evaluating the 3 combat types (charging, being charged, and protracted combats) we can use weighted averages once again, based on the unit type. Heavy cav will have a lot of weight put on their "charging" combat scores, while heavy infantry will have higher percentages on being charged and protracted combat.

OR it might even be possible to just do something based on the difference in the two combatants' movement scores. M4 infantry will rarely get to charge M7 cav. So a lower "weight" would be given to the charging value of the infantry against that target.

Not sure which of the two ways I'd favor.

 

[Barotok]

I'm thinking a lot about how usefull numbers that calcualted the value of a heavy cav charge vs infantry would be in comparing numbers to different heavy cav units charging each other. Originally, I was wanting to run the numbers against similar units, but this becomes less favorable because these numbers don't generate real battlefield results, for the most part (because you don't see cavalry charging cavalry, at least in the heavy department, very often). I would now say measuring each heavy cav unit against what its typical charg targets would be, is far more usefull than measuring it against itself or another similar unit.

This requires a bit of a subjective look at each unit type though because evaluating combat prowress in this way for all units wouldn't work. Perhaps I'm over thinking this, but would we want to measure everything against infantry? It would make sense to measure infantry vs. other infantry for this is a common part of the game, but I don't see the usefullness of calculating say, flyers vs. infantry as thats not what their optimal role is.

 

[Caneghem]

This requires a bit of a subjective look at each unit type though because evaluating combat prowress in this way for all units wouldn't work. Perhaps I'm over thinking this, but would we want to measure everything against infantry? It would make sense to measure infantry vs. other infantry for this is a common part of the game, but I don't see the usefullness of calculating say, flyers vs. infantry as thats not what their optimal role is.

This is why first you'd want to categorize everything, and that would be a touch on the subjective side. You just have to figure out what separates light infantry from heavy infantry. I class heavy infantry as anything that can be made into an "anvil unit". Common sense and experience should be able to dictate which infantry types can perform this role well (like saurus) and which aren't so well suited (skink cohorts) and would be light infantry. Flyers might charge light infantry more often than heavy infantry, so you'd weigh their light infantry score at a higher rate. Of course, flyers would have the most importance placed on targets like warmachines and support units. I also wouldn't ever rule anything completely out of the final score... sometimes weak flyers are brought into a rear charge against heavy infantry just for the -2 combat swing for the rear charge. So maybe you'd only weight their heavy infantry score at 3-5%, but it should be a component.

But the first step would be figuring out what all the categories would be, then feeling out where the percentages should fall when rating each unit.

 

[Celticfire]

Call me stupid if you want, I guess I am missing the point of you quote a previous post in the thread with no comments. I can see how both posts could be read as supporting two opposite opinions, but am unsure if that is the point you are trying to make. Care to clairfy?

There are comments in the quote, which i never preveiwed, sorry. i never called you stupid, please no put words into my mouth.

Is it jaut me or are we forgetting to evaluate archers? this system that you two are attempting to create would have a difficult time evaluating any shooty Wood Elves, for example. i am not talking about a War machine spam, just normal archers. in combat they would not do well but the comp system needs to be evaluate their usefulness in the ranged phase.

 

[Barotok]

Don't worry about it Celtic, I was just confused about why you posted. Thanks for clarifying.

You bring up a great point about archers and the shooting phase. I think that this would be a tough system to evaluate primarily because of the diverse ranges and types of weapons that are available to use. For the most part, shooters are not really that good in close combat and can get mowed down by the average infantry block. Obviously their role is different than any other unit type in the army. I think that not a lot of the thread has leant itself to the discussion of how this particular phase of the game would be handled. A lot of emphasis on infantry/cavalry and close combat have come up but virutally no comments on shooting.

I'll attempt to offer some. A unit that shoots has a few factors that have to be measured. The first one is of course how powerful the shooting is. Shot strength is a big factor but range is also important. Better shooters will hit harder from further away. Weapon skill and shot strength obviously factor into the formula for this type of troop but also movement and manueverability. Take for example, our skinks. Very short range, but extremely manueverable. A group of empire archers is much less manueverable but can hit from longer distances.

The most basic calculations look something like this:

%To-Hit x %To-Wound / Range

This does not take into account movement or the flexibility of skirmishers. Obviously the formula must expand to accept these numbers, considering that skirmishers may fire from a very loose formation and may do so from a 360 degree LoS. While a group of archers must reform, wheel or perform some other type of manuever to alter their firing arcs.

Another phase that I feel is far more tricky and difficult to calculate than shooting will be, of course magic. A simple count-up-the-power-dice exercise just won't do the trick. This means, lores (sometimes army specific), spell caster's levels, their mobility and a host of other factors weigh in when determining how effective a particular caster is at delivering spells to the battlefield.

Thanks for bring up the shooting phase Celtic, pehaps Caneg can offer some comments on this particular phase of the game too.

 

[Celticfire]

i think that we would also have to take into consideration; moving and shooting, and long range. useing your basic formula, i would say take each archer unit and test them when they have

moved, and are at long range
not moved, and at short range
moved, and a long range
not moved, and at short range

these are the penalties that people run into when shooting. allthough depending on the weapon, some of the claculations might be the same. for example useing a short bow either moving and at short range, or not moving at long range, will be a -1 to your To HIt score. iguess what iam am trying to say is that you need to moke sure that you calulatiote the To Hit penalties. and you also need to take into consideration things like the slings double shot inside 9", or the crossbow mave or shoot rule. (i need to stop being long winded, when only a short breaze will do)