Manpower Losses - Too Variable?

[frankpowerful]

(...) a very small handful of people that come to THIS forum and post ridiculous suggestions to change the game.

speaking of ridiculous...
in a DAG game 2 days ago, a Disrupted unit of Average Jinetes fighting a full strength good order unit of superior men at arms in open terrain gave 38 losses to 1. this means that the men at arms took about 13%losses in combat against the 0,33% of the disrupted light cavalry.
NO COMMENT..

[TheGrayMouser]

I disagree with any kind of "fixed" combat results as that turns it into a gamey game instead of a realistic historical battle. Every charge into another unit didn't yield a 22% damage ratio of course neither did it 17% or 27% but luck whether you like it or not does play a big part in events and situations. I like it the way it is now and it's not important enough to muck with or change right now there are a lot more pressing things that need to be worked on like AI and random battle MAP generators.

And besides only 32 votes doesn't come close to representing the whole of the gaming community (according to Iain that is) so it's hardly a representation of the whole just a representation of a very small handful of people that come to THIS forum and post ridiculous suggestions to change the game.

Dude, even if you dont agree with the thread and the authors suggestion of the game change, why throw condescending commentary in there?

[mceochaidh]

Definition of "Forum" from the Merriam-Webster Dictionary:

1) A public meeting place for OPEN discussion

2) A medium (as a newspaper or online services) of OPEN discussion or expression of ideas.

I would imagine that the above definitions are the intent of Slitherine and the game developers in setting up this forum. I find that the regular contributors almost always have pertinent ideas to share and care about the game (and like the game) enough to share such ideas.

Of course, it is always within the perview of the developers when or if to adopt any of the ideas, but it is my impression that in some cases, ideas put forth on this forum have resulted in positive game changes.

[Scutarii]

As my grandfather says to me when he correct me "Who loves you will make you cry", the only way to see a game evolve is with the players opinions and ideas, i remember when they change the anarchy charges system (now are much less than in the previous version) and finally they return (not all but enough) to previous situation... i think that people who post in the forum is because they love the game and want see more and better, i dont know what is in the mind of "neverpost" people and only can say that if you dont like people ideas dont read and post in forums.

For me the casualties system is the main weak point of the game, one thing is luck and other very different is in situations where enemy troops or your own troops dont have any chance in tactical situation and combat bonus, only dices save the poor unit BUT the problem for is see this poor unit killing troops as Terminator in a police station

The key is prevent results where D units with all black kill 20% or more troops only suffering 1-2%, reduce the "rape" is more realistic because in a film see the hero killing 2.565 soldiers when is injuried is fine but not credible and here is the same situation, a unit in bad situation killing double enemies is a superb result, 20 times is

[76mm]

I disagree with any kind of "fixed" combat results as that turns it into a gamey game instead of a realistic historical battle. Every charge into another unit didn't yield a 22% damage ratio of course neither did it 17% or 27% but luck whether you like it or not does play a big part in events and situations. I like it the way it is now and it's not important enough to muck with or change right now there are a lot more pressing things that need to be worked on like AI and random battle MAP generators.

And besides only 32 votes doesn't come close to representing the whole of the gaming community (according to Iain that is) so it's hardly a representation of the whole just a representation of a very small handful of people that come to THIS forum and post ridiculous suggestions to change the game.

oh boy....I think you're mischaracterizing the thread again--no one is calling for "fixed" combat results, or denying that luck plays a big part in these battles.

and as for ridiculous suggestions, I think that suggesting that the devs should spend more time fixing the AI is right up there...so I guess we're even.

[Igorputski]

reduce the "rape" is more realistic because in a film see the hero killing 2.565 soldiers when is injuried is fine but not credible and here is the same situation,

You obviously never saw the "movie" Sampson & Deliah where Sampson slew 1000 phillistines with the jawbone of an ass. This is true documented story in history as well. )

as for ridiculous suggestions I think that suggesting that the devs should spend more time fixing the AI is right up there...so I guess we're even.

Well that really shows your IQ level now. Go back to romper room and tinker toys please and leave the wargaming to us MEN!

[76mm]

Well that really shows your IQ level now. Go back to romper room and tinker toys please and leave the wargaming to us MEN!

Your other replies indicate that you are capable of responding tongue-in-cheek, so I will assume the same of this response, rather than the obnoxious and juvenile response that it would superficially appear to be...

[mceochaidh]

I thought it may be helpful to look at how the TT rules handle manpower losses compared to FOG PC. The TT rules define a BG as having multiple bases. An 800 point TT army (typical for tournament play) will have 10 to 15 BGs.

The range of bases per BG varies, but lets assume 4 to 8 bases or an average of 6 bases per BG (Foot will typically have more, Cav less.) To make the comparison simple, I will use opposing BGs, each of 4 bases of offensive spear (2 ranks of 2 bases), protected, and average, so no POAs or quality re-rolls. Score to hit is 4,5,6 and 4 dice are rolled for impact (2 per front rank.) One side scores 2 hits, the other 1 hit. The losing side does its death roll. This is done by rolling 1 dice. You have to score higher than the number of hits received or lose a base. Therefore if 1 or 2 is rolled, a base would be lost. There is a 33% chance of rolling 1 or 2, so there is a 33% chance of losing 1 out of the 4 bases. This is the same as saying there is a 33% chance of losing 25% of the initial strength of the BG. This is equal to 8.25% (33% X 25%).

FOG PC uses a table of 5% to 14% to calculate manpower losses for losing an impact while receiving 2 hits. The average loss is 9.5%. To equate FOG PC to the TT, the table for 2 hits would have to be reduced to 4% to 12.5%. The other tables (for example the table for 3 hits) all seem slightly skewed to the high side. I am sure the devs had their reasons for doing the conversion this way, but the result seems to be higher casualty rates (on average) than the TT version.

The bell curve I proposed is a more modest change than altering the ranges as detailed above, is probably statistically closer to the TT version and does not (in my opinion) change game balance. Lowering the ranges is another approach that seems valid, if the goal is to follow the TT version. Either change would not require much, if any, programming, so should be an easy fix.

[deeter]

A good analysis. Also, keep in mind in this example, the winner of a close combat adds two to his death roll so would have zero chance of taking losses. The thing that is annoying most players is when the win but take higher losses than the loser did which means high-quality troops suffer attrition in the PC version they would not suffer on the TT.

Deeter

[keithmartinsmith]

The PC loss ratio's actually improves greatly on the TT.
1) For example on the TT two BG', one three bases of superior, lancers, uphill, heavily armoured knights one deep fight in impact vs. other an unprotected poor mob in a line two deep.
2) The lancers get 6 attacks re-rolling 1's, needing 3 to hit and get 5 hits.
3) The mop gets 6 attacks re-rolling 6's, needing 5 to hit and get 3 hits.
4) The losses in the system are not linked to who wins, but the knights do with and the mob take and pass a cohesion test, typically neededs a 10 to pass (lost, lost big and lost against mounted in the open).
5) We now convert the hits to base losses - The Knight roll a 1 so they they lose a base = 33% losses.
6) The mob needs to and rolls a 6 to avoid any losses.
7) 33% to 0% Had the knights started as a 4 base average unit (not uncommon) they would now rout.
The PC game does use a distributtion curve to allocate losses depending on the number of hits obtained and its rarely zero. As units strengths are stored as a percentage all losses hurt a unit.
9) I like the TT a lot and play it a lot but it seems hard to improve on what we have. To guarentee losses in a combat by the better side would not be fair or historical as there are far to many battles where, until one army routs, losses are minute.
Keith

[TheGrayMouser]

No offense but I dont follow your math.

In the TT game with your above example: you have a 33% chance to lose 25% of your BG, however you have a 67% chance of losing nothing....

33*25=8.25 ?? I dont see how this value is any way indicative of anything.... The death roll is all or nothing

If anything the pc game somewhat mitigates your combat loss by only 5-14% loss with 2 hits , not 25% as per your above example of a 4 stand BG

I think the true issue is the constant wittling way of your units strength % wise, whether winning or losing a comabt but that is another story

[mceochaidh]

Mouser,

No offense taken as I was not clear enough in my example. I was trying to determine how close, on average over many examples, the PC is to the TT (pretty close). Since on the TT, you can't cut up a base to reflect casuaties, the reasonable playable solution is to remove a whole base. So if you have 3 BGs each of 4 bases and all lose the impact, each will have a 33% chance to lose one base. Out of 12 bases, on average one base will be lost in this circumstance. One out of 12 is 8.3%, close to the 8.25% in my first example (due to rounding, as I should have multiplied 33.33% time 25%.) As I said above, the average on the PC should be the mid-point of the range of 5% to 14% or 9.5%, a little higher than the TT.

As to your other point regarding cumulative losses, I actually like this feature and consider it more historically accurate then is possible on the TT. I consider this the total of actual casualties, wounded, fatigued or demoralized men in the unit. I just believe the percentages in the range should be toned down slightly.

For the record, I have NEVER suggested that more powerful units should ALWAYS cause more losses than less powerful ones. Even Caesar knew that fortune always played a role in battle!

Thanks to Keith from HexWar for his response. I am not sure I yet understand all of his comments, so will respond to that post separately.

[stockwellpete]

I voted in the poll for a change in the way combat outcomes are calculated. But it is only a moderate concern of mine because I do like the "chance" element in the game too. There are what might be called bizarre results in nearly every battle I fight but I take the view that these will even themselves out in the long run. And to be more specific, I find that if a unit of mine is on the receiving end of one these bizarre outcomes, then if I attack the offending enemy unit in the same turn I can often mitigate a lot of the damage - and then if you add up the two scores from the melees you usually find a much more realistic casualty rate.

I suppose, and I may be completely wrong here, that some of the differences expressed on this board over this issue can be broken down between those of us who treat FOG primarily as a game, and those of us who are a bit more into historical re-creation (like me). That is why I voted for change - basically I just would like there to be a bit more of a "wearing down" process in combats between units of different calibre than there is at the moment.

[TheGrayMouser]

ahh now the 8.25% makes more sense!

I see what your saying now.

Re: the gradual deterioration of units: hmm, this is a tough one... I like it for as you point out the small gradual attrition (whether from melee or missles (especially missles) is accounted for and adds "realism" or at least immersion..... (and obviously cant be accounted for in a TT game)

Howver the only problem i have with it is kinda abstract but falls in line with the thread that discussed lists that having MORE battle groups generally do better (nothing to do with tactics terrain player skill) just raw #s and mathemetics. So very expesnive units , no matter if they win every combat will eventually be swamped out (ie autoroute ) from the cumulative % they lose every combat roll. BTW im not using an argument of realism or such factors as fatigue, broken weapons either, more of a game play balance which is differnt than the TT. Since death rolls are harder to lose when you win a combat in the TT, theoretically highly expensive and powerfull units generally can last the game if they win their combats wheras a knight in the PC game, for example, sure maybe it can take out 2-3 medium lights spears but it will be at like 48% casualties and likly autoroute the next time it is in combat .....


Im not sure I understand all Keiths #s either but I think what he is saying is all units suffer in combat win or lose
The example I'll make is elephants that have 20 "men" per BG, if you watch the game and see your elephant take missle fire for exapmle, at ist glance you might think you are unscathed as the game still states you have 20 elephants. Howver, if you look at the % remaining it likly will be at 97% or 98%. Ie those apparently harmless missles ARE slowly damaging the Pacs.

[mceochaidh]

Keith said "The PC game does use a distributtion curve to allocate losses depending on the number of hits obtained and its rarely zero. As units strengths are stored as a percentage all losses hurt a unit."

I presume the "distribution curve" Keith is referring to is the range of losses we have been discussing, which for losing an impact and receiving 2 hits is 5% to 14%. Am I correct that there is an equal chance of each percentage in the range occurring, ie. an equal chance of 5%, 10% or 14% being used for loss calculation? In other words, my assumption is no bell curve is presently being used that would produce a result more often at 10% than it would at 5% or 14%.

A bell curve can be constructed in different ways. Maybe an actual example would help. The range above has 10 possible results (5%,6%,7%,8%,9%,10%,11%,12%,13%,14%). Presently, each result (if no bell curve used) would occur 1 time out of 10 or 10% of the time. The total of the percentages add up to 100% (10 x 10%). A likely bell curve could start with the 5% result occuring 6% of the time (instead of 10% of the time.) Then the 6% loss could occur 8% of the time, the 7% loss 10% of the time, the 8% loss 12% of the time, the 9% loss 14% of the time, the 10% loss 14% of the time, the 11% loss 12% of the time, the 12% loss 10% of the time, the 13% loss 8% of the time and the 14% loss 6% of the time. Note that the total of all of the percentages of each loss occuring still and must add up to 100%. This is just an example, as the curve could be flatter or steeper. The curve idea produces less frequency of extreme results.

It would still be possible for weaker units to beat stronger units. For one thing the number of hits determine whether a combat is won or lost. The bell curve idea would just modify percentage losses and the average percentage losses would remain the same. I think the principal result would be that combats would last a little longer and battle lines would not break up as quickly, but I believe the change would be moderate, not alter game balance in any way, and best of all, not require programing.

[magobarca]

I noticed that when Of. Spears, Lt. Spears, & Def. Spears impact with an Impact Foot unit they usually take high losses, maybe too high? Spears do better afterwards, to some degree, but still usually lose. Seems Pike units behave the same way, but I haven't used pikes that much yet.

[Scutarii]

In my own experience with pikes can say that they do a good defensive job (much better than defensive or offensive spears)... but not allways, isnt strange see pike units with high casualties in the first melee even VS sword units, i think that pike units have a 66% to do a good job in the first melee when they are attacked, in offensive is a coin question, fifty fifty

PD: mceochaidh is a good idea, reduce extreme casualties results is a good option, add it as optional feature (like double moves) could be a great adition to the game.

[magobarca]

@ Scutarii, Thanx for the info. about Pikes. From historical accounts it doesn't seem that Pikes actually took a large number of casualties upon impact with the Romans, & on flat ground always pushed the Legionaires back.

[Triarii]

Keith said "The PC game does use a distributtion curve to allocate losses depending on the number of hits obtained and its rarely zero. As units strengths are stored as a percentage all losses hurt a unit."

I presume the "distribution curve" Keith is referring to is the range of losses we have been discussing, which for losing an impact and receiving 2 hits is 5% to 14%. Am I correct that there is an equal chance of each percentage in the range occurring, ie. an equal chance of 5%, 10% or 14% being used for loss calculation? In other words, my assumption is no bell curve is presently being used that would produce a result more often at 10% than it would at 5% or 14%.

A bell curve can be constructed in different ways.

First I have voted for no change as this is not something that I feel very strongly about.

However Deeter's point about culmulative damage set me thinking and I do agree that there may be too great a frequency of moderately high casualty results that prevent sustained, repeated, attacks by higher quality troops against lower quality - the tendancy of the high volume low cost forces to do better against the high cost low volume forces seems to support this.

Apologies for stating the obvious but Mceochaidh is correct and bell curves can be constructed in different ways. Furthermore a curve for frequency distribution does not have to be the traditional symmetrical bell.
Here I assume but; If the current distribuition curve Keith referred to is based upon RNG that follows the traditional die roll the curve is likely to be a typical, symmetrical, bell with sides closer to constant gradient i.e. /\. One aid to game play to address Deeter and GM's comments about culmulative attrition might be to moderate the frequency distribution so that one side become more of a 'J' curve. This can be scripted fairly easily in the digital game as opposed to the die reliant TT game.
An asymmetrical 'J\' frequency distribution might be a way to make fewer casualties more common for winning units without making absolutely crippling high losses an equal and opposite chance.

I am of course a 'know nothing' about the game design and am merely guessing that this is practical in the coding. If it is, and given support on these forums, an alternative casualty calculation of this type may be worth trying at some point in the future. For me though the current system works and I am equally content if this observation sinks without trace

[mceochaidh]

I would just like to thank all who participated in this poll; I set it up for a ten day response and I believe this period is over. I hope the discussion has stimulated some thought among the forum participants and development group. I fully realize the dev group are working on many issues and new projects and have a priority list for same.

Personally, I believe the game is very much worth the effort to continue improving and has the potential to become even more historically accurate as well as remaining the very playable game it is presently. I salute the design and development team for their efforts!