Loss Aversion Among Scenario Designers

Matrix Games Forums

Forums Register Login Photo Gallery Member List Search Calendars FAQ

My Profile Inbox Address Book My Subscription My Forums Log Out

Loss Aversion Among Scenario Designers

View related threads: (in this forum | in all forums)

Logged in as: Guest

Users viewing this topic: none

Printable Version

All Forums >> [New Releases from Matrix Games] >> Command: Modern Operations series >> Mods and Scenarios >> Loss Aversion Among Scenario Designers

Page: [1] 2 next > >>

Message

<< Older Topic Newer Topic >>

Loss Aversion Among Scenario Designers - 12/12/2017 9:41:18 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

One of the most peculiar things I've noticed in community scenarios in particular, is the rampant loss aversion among scenario designers. Computer controlled forces hurl squadron after squadron into battle, such that some losses would be almost inevitable, only to have you penalized so heavily for aircraft loss that it nearly outweighs accomplishing any other goals. Often, shooting down an equivalent number of opposing aircraft results in little or no benefit scoring wise.

Perhaps this is a reflection of the overwhelming value wargamers place on the loss of imaginary life? No imaginary military/political gain could possibly justify the loss of our brave imaginary pilots. This would be the "give peace a chance," school of wargaming. They've played every war game over and over again and like the W.O.P.R. computer have concluded that, "The only winning move is not to play." It's a laudable, idealistic position. I hope we one day live in that world where even the bleeding of imaginary combatants is too much pain and suffering for our conscience to bear.

Sometimes I suspect it's a statement of value. "Our stuff is expensive and their stuff is trash. Therefore you receive little or no benefit for destroying their stuff, and terrible penalties for losing our stuff." This would be fine, so long as the stuff you're losing is actually that valuable, and the things you're destroying are actually that cheap. Too often, however, the scores associated with the platforms seem disproportionate to their actual cost (e.g. fighters being worth more scoring wise than AWACS, DDGs, CVNs, tanker aircraft, or even the airports which support them).

Sometimes I suspect scoring systems are intended to reflect military utility. Once again, however, this would seem to frequently be disproportionate to their actual military value. If the loss of an aircraft is of such critical military importance, then wouldn't it be better to just have it born out in the consequences? Perhaps you lots of have spares in the scenario? If that's the case why does the scoring loss need to be so big? Perhaps in the course of action I chose, that platform doesn't matter so much? If their stuff is so unimportant, how come destroying it impacts one's ability to achieve the scenario goals? I ought to be able to smash it endlessly and make no headway. Why would the loss of a few fighters not be worth striking a pontoon bridge, leaving an invading force cut off from heavy forces, thus ending the war? It frequently seems to be the case that scenario designers haven't really thought about the scenario they're creating.

Sometimes I suspect scoring systems are intended to reflect politics. Here I have objections. Seldom does the damage or destruction of a platform or piece of equipment result in much political fallout. Rather, it is the associated loss of life, or the potential for service members to be held prisoner associated with it. Therefore, in the case of aircraft loss, political considerations would seem to be contingent on the recovery of the pilot, not the loss of the plane, and a score penalty is not a fair reflection of the intended driver. One must also think about the scenario. What is the conflict about? How much loss of life is a nation willing to tolerate in order to achieve those ends? Too frequently it would seem the answer is "none." In that case, the scenario may be implausible. The only case where it might not be is in "Cold War" situations like the Gary Powers shoot down, where "winning" is never having to fight, or maintaining deniability. Once again, proportionality issues abound, where single seat fighters are worth more than CVNs. It's as if the scenario author is driving towards a preconceived conclusion with his penalties and scores for aircraft struck down.

The software has the "butcher's bill" screen where you can see losses and expenditures. You don't need the score to track that. What is the purpose of a scenario's scoring system and what is a reasonable way to implement it?

Post #: 1

RE: Loss Aversion Among Scenario Designers - 12/12/2017 10:23:30 PM

Excroat3

Posts: 436
Joined: 1/24/2015
Status: offline

For my scenarios, I don't score my units based on how expensive/useful they are, but rather I plan my scoring around making it difficult for the player to "cheat" my scenarios. Sure, you can throw F/A-18s with mavericks at a Soviet fleet all day, and you probably would end up sinking it, but at what cost? That isn't a strategy actual planners would use, so why should it be allowed to be played that way in my scenarios? I want players to see my scenarios as realistic as possible, and scoring is one of many ways to make sure you keep the player honest while he/she makes their way through your scenario.

(in reply to SeaQueen)

Post #: 2

RE: Loss Aversion Among Scenario Designers - 12/12/2017 10:40:47 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

Hi

Scenario authors who build scenarios for fun or to entertain have different objectives than a pro analyst. Its done for the same reason there is scoring in competitive sports. You want to give a player some validation for their efforts and a sense of glory and defeat. So its really about entertaining, providing a challenge, bringing joy etc. which I realize isn't exactly a focus of professional wargaming. It's ok though I accept it's shortcomings!

Might be a sign of my age but I'm now much less judgemental about other's work and don't get too hung up on OOB inaccuracies or some game theory they're not catching. I"m pretty sure that's a good thing!

As an aside we still need the scoring mechanism in current campaign scenarios because its the only current technical path for players to progress through a campaign. You can conclude that scenarios authors may have used it for only that.

Thanks

Mike

< Message edited by mikmykWS -- 12/12/2017 11:26:17 PM >

_____________________________

(in reply to SeaQueen)

Post #: 3

RE: Loss Aversion Among Scenario Designers - 12/13/2017 12:36:29 AM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

That makes sense, except that a scenario score should be also be agnostic about the tactics and weapons a player should use. Your score shouldn't be based on using the tactic the scenario designer thinks you ought to be using unless the scenario is intended to teach a single tactic.

If the goal is to sink a Soviet SAG, then you should be scored on the ability to do that. The problem with reducing your score based on losses is that the scenario designer must then ask how much of the Soviet fleet is a fighter worth? The answer to that is, "it depends..." Few scenarios seem to take an interest in making that calculation. A lot of the time they seem to arbitrarily make the loss of a single aircraft disproportionately devastating to one's score. It's gotten to the point that in some good scenarios, I find myself disregarding the score entirely and merely judging my performance based on other metrics such as exchange ratio. In some scenarios the score functions as a proxy for exchange ratio, but those are usually oriented towards air-to-air combat. Once you start adding in a strike component it becomes more complex and I'm not sure they get that balance right.

quote:

ORIGINAL: Excroat3

For my scenarios, I don't score my units based on how expensive/useful they are, but rather I plan my scoring around making it difficult for the player to "cheat" my scenarios. Sure, you can throw F/A-18s with mavericks at a Soviet fleet all day, and you probably would end up sinking it, but at what cost? That isn't a strategy actual planners would use, so why should it be allowed to be played that way in my scenarios? I want players to see my scenarios as realistic as possible, and scoring is one of many ways to make sure you keep the player honest while he/she makes their way through your scenario.

(in reply to Excroat3)

Post #: 4

RE: Loss Aversion Among Scenario Designers - 12/13/2017 12:44:35 AM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

You're right, they have different purposes than what I do at work but I play the game for fun too. Until online multiplayer is a reality, you're really only mostly competing against yourself. That's great to me. Sadly, I don't think we sufficiently recognize the entertainment value in attempting to do something difficult better than one did before. In light of that, it's important that the score be a fair representation of one's "goodness." Sometimes I look at it and think it's a little arbitrary.

quote:

ORIGINAL: mikmykWS

Hi

Scenario authors who build scenarios for fun or to entertain have different objectives than a pro analyst. Its done for the same reason there is scoring in competitive sports. You want to give a player some validation for their efforts and a sense of glory and defeat. So its really about entertaining, providing a challenge, bringing joy etc. which I realize isn't exactly a focus of professional wargaming. It's ok though I accept it's shortcomings!

Might be a sign of my age but I'm now much less judgemental about other's work and don't get too hung up on OOB inaccuracies or some game theory they're not catching. I"m pretty sure that's a good thing!

As an aside we still need the scoring mechanism in current campaign scenarios because its the only current technical path for players to progress through a campaign. You can conclude that scenarios authors may have used it for only that.

Thanks

Mike

(in reply to mikmykWS)

Post #: 5

RE: Loss Aversion Among Scenario Designers - 12/13/2017 1:14:01 AM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

ORIGINAL: SeaQueen

You're right, they have different purposes than what I do at work but I play the game for fun too. Until online multiplayer is a reality, you're really only mostly competing against yourself. That's great to me. Sadly, I don't think we sufficiently recognize the entertainment value in attempting to do something difficult better than one did before. In light of that, it's important that the score be a fair representation of one's "goodness." Sometimes I look at it and think it's a little arbitrary.

I know the entertainment value of hard work. We wouldn't be here if I and the rest of the team didn't. Could "we" please stop being like that

I just think it really doesn't matter why things were done as long as people have fun. Most of these people are putting themselves out there with their work which takes a lot of guts. Just doesn't make sense to me to start banging on them for low rent stuff in front of the crowd but instead engage them directly like humans do.

When are you going to put your work out there? Scared?

Thanks!

Mike

_____________________________

(in reply to SeaQueen)

Post #: 6

RE: Loss Aversion Among Scenario Designers - 12/13/2017 2:34:17 AM

ExNusquam

Posts: 513
Joined: 3/4/2014
From: Washington, D.C.
Status: offline

I think the scoring penalty for player losses can also be abstracted to include force preservation for future operations. This stems from the engines inability to carry units between scenarios. If player throws all of their Hornets at a SAG with Mavericks, it might make it impossible to use those Hornets for CAS with follow-on operations (which were probably the reason you had to engage the SAG in the first place!). In these cases where the impact of player losses may not manifest within the same scenario, the player needs some feedback to emphasize that impact.

I think directly using the scoring system is probably a bit ham-fisted - using lua you could easily give the player a target loss rate where small losses are not a problem but losing entire squadrons affects the score. The other solution would be to build longer scenarios, as there's no hard limit on scenario duration. Instead of assigning an arbitrary value to units, you can show the player their value (i.e. taking severe losses on day 1 makes day 2 impossible to execute).

quote:

When are you going to put your work out there? Scared?

I think there are a few of us who don't post scenarios due to professional concerns.

(in reply to mikmykWS)

Post #: 7

RE: Loss Aversion Among Scenario Designers - 12/13/2017 3:45:09 AM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

I think there are a few of us who don't post scenarios due to professional concerns.

Fair enough. Guess we'll never know.

Mike

_____________________________

(in reply to ExNusquam)

Post #: 8

RE: Loss Aversion Among Scenario Designers - 12/13/2017 11:06:49 AM

IainMcNeil

Posts: 2804
Joined: 10/26/2004
From: London
Status: offline

I tend to agree with Sea Queen. I think having the casualties affect the score would make scenarios more replayable so you can attempt to do better. I don't see how it diminishes the fun to reward good play, and I don't see how it implies scenario designers have done something wrong by not adding it. I'm not getting the negativity here :)

_____________________________

Iain McNeil
Director
Matrix Games

(in reply to mikmykWS)

Post #: 9

RE: Loss Aversion Among Scenario Designers - 12/13/2017 1:09:32 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

Agreed. They haven't necessarily done something wrong, but I do sometimes think they weight losses disproportionately to achieving other scenario goals, for various, often irrational or illogical reasons. To me, that begs the question, does there exist a rational way to attach a score value to scenario goals?

quote:

ORIGINAL: IainMcNeil
I tend to agree with Sea Queen. I think having the casualties affect the score would make scenarios more replayable so you can attempt to do better. I don't see how it diminishes the fun to reward good play, and I don't see how it implies scenario designers have done something wrong by not adding it. I'm not getting the negativity here :)

(in reply to IainMcNeil)

Post #: 10

RE: Loss Aversion Among Scenario Designers - 12/13/2017 1:19:10 PM

Cik

Posts: 671
Joined: 10/5/2016
Status: offline

I actually generally like the scenarios where "one loss is too many" though really generally i lose at least a few assets, it's nice to play for the "perfect score" and try to achieve everything without a loss (except munitions) it allows the designer to give the player overwhelming force, but it still must be applied carefully- if you get 24 B-52s and you can just lose them willy-nilly there's less tension- where if you give the player 2x CVNs, a pile of subs, a few bases full of sophisticated 4th gen+ but the overriding message is "BE CAREFUL!" even a mig21 then becomes a problem for you- despite the fact that you can shoot it out of the sky without much trouble in the forward quarter, if it DOES manage to sneak up on you, it's a tense situation.

i try to play most scenarios as if i'm actually going to have to rescue downed pilots and extricate stranded crews (even though that's often not an explicit feature) that is to say, any pilot that goes down piles on more risk because I'll have to send in some helicopters which themselves can be shot down, and risk more guys. for that reason it's best to not get shot down in the first place if you can avoid it.

the scoring doesn't need to be too harsh for this sort of thing though. generally i don't pay much attention to it anyway. i get my signals of how careful i'm supposed to be from the briefings mostly and the scoring is mostly an afterthought.

(in reply to SeaQueen)

Post #: 11

RE: Loss Aversion Among Scenario Designers - 12/13/2017 1:27:01 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

That makes sense, and I like it.

To do that right, you'd need to decide how many aircraft are necessary for the follow on operations. You can use LUA to effect the score. You could create a trigger that goes off when you have lost more than the necessary number of aircraft for the subsequent operations, and deducts points for that. I like that idea, because you could state in the scenario briefing that one of the goals is, "Preserve N strike aircraft for subsequent follow on operations." That puts the player in a position of balancing risk. They could commit (Total Number of Strike Aircraft - N) and play it safe, but they might not achieve the goal in the time constraint. They could commit more than that, and each aircraft committed represents a risk, but it also represents an increased likelihood of success. Where's the sweet spot?

quote:

ORIGINAL: ExNusquam

I think the scoring penalty for player losses can also be abstracted to include force preservation for future operations. This stems from the engines inability to carry units between scenarios. If player throws all of their Hornets at a SAG with Mavericks, it might make it impossible to use those Hornets for CAS with follow-on operations (which were probably the reason you had to engage the SAG in the first place!). In these cases where the impact of player losses may not manifest within the same scenario, the player needs some feedback to emphasize that impact.

I think directly using the scoring system is probably a bit ham-fisted - using lua you could easily give the player a target loss rate where small losses are not a problem but losing entire squadrons affects the score. The other solution would be to build longer scenarios, as there's no hard limit on scenario duration. Instead of assigning an arbitrary value to units, you can show the player their value (i.e. taking severe losses on day 1 makes day 2 impossible to execute).

(in reply to ExNusquam)

Post #: 12

RE: Loss Aversion Among Scenario Designers - 12/13/2017 2:39:12 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

ORIGINAL: SeaQueen

Agreed. They haven't necessarily done something wrong, but I do sometimes think they weight losses disproportionately to achieving other scenario goals, for various, often irrational or illogical reasons. To me, that begs the question, does there exist a rational way to attach a score value to scenario goals?

quote:

ORIGINAL: IainMcNeil
I tend to agree with Sea Queen. I think having the casualties affect the score would make scenarios more replayable so you can attempt to do better. I don't see how it diminishes the fun to reward good play, and I don't see how it implies scenario designers have done something wrong by not adding it. I'm not getting the negativity here :)

Thanks for clarifying that SQ!

Mike

_____________________________

(in reply to SeaQueen)

Post #: 13

RE: Loss Aversion Among Scenario Designers - 12/13/2017 2:47:24 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

The problem with "one loss is too many," is that it undermines the whole premise of the scenario, which is that hostilities have begun. If any losses are too many, then you probably shouldn't be risking any forces at all in a military conflict; it's a job for the guys in Foggy Bottom, not Arlington. The only time to me "one loss is too many" makes sense is in "Cold War," "pre-hostilities," or ISR sorts of scenarios where the goal is to avoid fighting.

I see what you're saying, though, regarding it as a way to add some challenge to an otherwise over-matched adversary. I typically avoid those sorts of scenarios, just because I'm less interested in conflicts with weaker powers, at least on the level that Command deals with them.

If the scoring is so uninteresting for a scenario that it's almost an afterthought, then something is probably wrong, and it needs to be improved.

quote:

ORIGINAL: Cik

I actually generally like the scenarios where "one loss is too many" though really generally i lose at least a few assets, it's nice to play for the "perfect score" and try to achieve everything without a loss (except munitions) it allows the designer to give the player overwhelming force, but it still must be applied carefully- if you get 24 B-52s and you can just lose them willy-nilly there's less tension- where if you give the player 2x CVNs, a pile of subs, a few bases full of sophisticated 4th gen+ but the overriding message is "BE CAREFUL!" even a mig21 then becomes a problem for you- despite the fact that you can shoot it out of the sky without much trouble in the forward quarter, if it DOES manage to sneak up on you, it's a tense situation.

i try to play most scenarios as if i'm actually going to have to rescue downed pilots and extricate stranded crews (even though that's often not an explicit feature) that is to say, any pilot that goes down piles on more risk because I'll have to send in some helicopters which themselves can be shot down, and risk more guys. for that reason it's best to not get shot down in the first place if you can avoid it.

the scoring doesn't need to be too harsh for this sort of thing though. generally i don't pay much attention to it anyway. i get my signals of how careful i'm supposed to be from the briefings mostly and the scoring is mostly an afterthought.

(in reply to Cik)

Post #: 14

RE: Loss Aversion Among Scenario Designers - 12/13/2017 2:56:11 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

ORIGINAL: IainMcNeil

I tend to agree with Sea Queen. I think having the casualties affect the score would make scenarios more replayable so you can attempt to do better. I don't see how it diminishes the fun to reward good play, and I don't see how it implies scenario designers have done something wrong by not adding it. I'm not getting the negativity here :)

Never said any of that and wasn't angry. My issue was that players weren't leveraging the scoring because they were missing something but that their goals were different.

Thanks

Mike

< Message edited by mikmykWS -- 12/13/2017 3:08:52 PM >

_____________________________

(in reply to IainMcNeil)

Post #: 15

RE: Loss Aversion Among Scenario Designers - 12/13/2017 9:31:05 PM

Gunner98

Posts: 5508
Joined: 4/29/2005
From: The Great White North!
Status: offline

Reflecting on it. I suppose that I generally use three different themes when setting up scoring:

-Losses, because losses hurt overall, both directly in the game (which shouldn't be penalized to harshly as you lose the asset), and more importantly in an overall campaign. Rightly or wrongly I tend to put an arbitrary value on units:
--multi engine AC are worth more than single engine (in VP loses)
--Ships (FF & larger) are worth more than AC, and the value goes up as the tonnage and capability does, CVNs are way high.
--Subs are valued about the same a CGs
--Player is usually worth more than opposition to add a tension in a knife fight - a 1:1 loss ration means you lose

-Special missions - and this is where the fun comes in. You can use these to add spice and variety to the scenario which will cause the player to balance task accomplishment with estimated losses. And some missions are tough enough that the player has to take significant risk to achieve it.

-Overall story awards - because most of my scenarios are in a mega campaign, the story is important, to me anyway. Some of the early scenarios, simply surviving is a victory, so I set points up to reflect this.

The problem with this approach is that it takes time to set up. Probably 20-30 events in some scenarios. Some authors don't go into that level of depth, and I am sometimes disappointed with the scoring, but if the scenario was fun, that's what counts.

Playtesting is important, helps with balancing the values so a victory is achievable but a you have to work at a triumph.

Anyway that's my $.02CAD

B

(in reply to mikmykWS)

Post #: 16

RE: Loss Aversion Among Scenario Designers - 12/14/2017 5:49:55 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

I like that way you order the loss of a platform's score value, but how do you weight their loss against accomplishing the mission at hand?

Example: Suppose the goal of the scenario was to safely deliver a group of logistics ships from their home port to a distant UNREP area. Protecting them are MPAs, a large surface combatant and a small surface combatant, equipped with helicopters and UAVs. Along they way they encounter mines, air, sub and ship launched ASCM raids and sub launched torpedoes. Since the scenario goal is to deliver the 4 logistics ships to the designated area, no loss of surface combatants ought to outweigh delivering those ships. Putting those 4 ships in the box with no remaining surface combatants is still a win, although it might not be the same degree of win as delivering the same four ships to the box with all escorts remaining. In that case it's pretty simple to come up with a reasonable scoring scheme.

Now consider a more complex case: Attacking an airbase. Attacking a well defended airbase means destroying the enemy CAPs, destroying their air defenses, attacking their runways, attacking their opened parking spots, attacking hangers and attacking POL storage. It's almost certain you'll take losses. How many fighters is it worth expending to damage a runway below a certain threshold? How many fighters is a POL tank or hanger worth? I haven't really been able to decide.

quote:

-Losses, because losses hurt overall, both directly in the game (which shouldn't be penalized to harshly as you lose the asset), and more importantly in an overall campaign. Rightly or wrongly I tend to put an arbitrary value on units:
--multi engine AC are worth more than single engine (in VP loses)
--Ships (FF & larger) are worth more than AC, and the value goes up as the tonnage and capability does, CVNs are way high.
--Subs are valued about the same a CGs
--Player is usually worth more than opposition to add a tension in a knife fight - a 1:1 loss ration means you lose

< Message edited by SeaQueen -- 12/14/2017 7:29:41 PM >

(in reply to Gunner98)

Post #: 17

RE: Loss Aversion Among Scenario Designers - 12/14/2017 6:56:14 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

Interesting.

Your boss gives you a task you get it done but it costs a ton. So job done well done but you're not likely to get any further tasks unless the outcome was spectacular and nobody cares about the cost.

Moving this thinking to scenario building and playing you don't have to worry about getting the next job but should the penalty for high loss (wherever) factor into win-loss so it actually does in a way?

I think so.

Mike

< Message edited by mikmykWS -- 12/14/2017 6:57:11 PM >

_____________________________

(in reply to SeaQueen)

Post #: 18

RE: Loss Aversion Among Scenario Designers - 12/14/2017 7:37:58 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

quote:

ORIGINAL: mikmykWS
Moving this thinking to scenario building and playing you don't have to worry about getting the next job but should the penalty for high loss (wherever) factor into win-loss so it actually does in a way?

Since the tendency in a scenario is to wear multiple hats (the player needn't necessarily correspond directly to a single commander, and might actually correspond to multiple people at different levels), I'm not sure that's a good analogy, but the cost/benefit calculation is definitely a real thing. So, how does one determine the score values such that there is a reasonable valuation of the losses versus the benefit of doing whatever they're trying to do in the scenario?

(in reply to mikmykWS)

Post #: 19

RE: Loss Aversion Among Scenario Designers - 12/14/2017 8:49:16 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

ORIGINAL: SeaQueen

quote:

ORIGINAL: mikmykWS
Moving this thinking to scenario building and playing you don't have to worry about getting the next job but should the penalty for high loss (wherever) factor into win-loss so it actually does in a way?

Since the tendency in a scenario is to wear multiple hats (the player needn't necessarily correspond directly to a single commander, and might actually correspond to multiple people at different levels), I'm not sure that's a good analogy, but the cost/benefit calculation is definitely a real thing. So, how does one determine the score values such that there is a reasonable valuation of the losses versus the benefit of doing whatever they're trying to do in the scenario?

Maybe some contrived value based on the court of public opinion modified by the speed of transmission and sensitivities of the era? This might be a little too contrived. Hmm...

How about just a measure of how capable your force is to do whatever after the win/loss? I think at high-level points are doing that now but maybe more thought has to be done at the detail level (loss of logistics assets, sites, etc are major losses etc.)?

If you really want to turn things on its head get a human ref or to evaluate the win-loss

Covers the things that can't be quantified.

Mike

_____________________________

(in reply to SeaQueen)

Post #: 20

RE: Loss Aversion Among Scenario Designers - 12/14/2017 10:42:22 PM

Gunner98

Posts: 5508
Joined: 4/29/2005
From: The Great White North!
Status: offline

quote:

get a human ref or to evaluate the win-loss

Well that's sort of what I do in my day job

but tricky in a game.

To accomplish the balance is a real trick I think. I don't want to sound like I have the answer - because I don't, but by mixing the three scoring themes I think it works. This is all rough guidelines not a rule that I use, but it seems to work OK.

First thing is working in the unit point value, if you lose a capital ship - even if mission and all tasks are accomplished, you shouldn't get a Triumph. Losing a carrier to succeed on a mission should probably end up in a loss - well that's a value judgment but fits in with the story award.

Mission acomplishment should be worth a goodly sum. If you achieve your mission with minimal losses, that should equal a Triumph. Minimal is reletive, <10% or so.

In any complex scenarios you need multiple tasks (I think), some contribute to the mission, some to the story and some to both.

Eg: You've got a CVBG (or CSG if you go modern) and your mission is to strike a bunch of land targets. These targets should be specified (as they would be, you're a tactical unit and are told what needs to be hit).
-Multiple targets are better than one.
-If an airbase or a complex target, you get decent points for what you're told to strike
-total points for all mission specific targets should add up to about 3/4 of a Triumph

Tasks like - a pop up SAR, positoning assets at a certain point, delivering something somewhere, UNREP etc
-there should be several
-they should be divers
-at least some of them should be a drain on resources
-one or two have to be very risky, to the point a commander may opt not to complete them
-These should total to about 1/2 of a Triumph

Story awards, often time based, sometimes key to the next scenario (or story), sometimes distracting, always at least somewhat interesting.
These should add up to about 1/4 of a Triumph

So it is possible, if you achieve all your tasks and story awards, part of your mission set, and cause some losses without taking any - that you could get a Triumph without completely accomplishing your mission. Unlikely but possible. I rationalize that by the likely losses and situation you have put on the enemy, you will acomplish the mission shortly, or indirectly.

In most cases though, if the player doesn't take huge losses but accomplishes the mission, a Triumph will be had.

Anyway, please don't go back and check my scenarios

, as this is the first time I've actualy articulated this and most of the time its more feel than science. That's why play-testing is important.

B

(in reply to mikmykWS)

Post #: 21

RE: Loss Aversion Among Scenario Designers - 12/14/2017 11:43:45 PM

kevinkins

Posts: 2257
Joined: 3/8/2006
Status: offline

Unfortunately, giving the AI (computer controlled forces) a material advantage is the primary way to give the human player a fight once they become proficient at Command. Designers can also provide their AI a scoring advantage and program scripted AI tactics like �ambushes� since the designer knows more about the situation than the player. This issue is not unique to Command at all and has been with war games forever. This is why a player friendly H2H system for Command has to be implicated so players can challenge one another in balanced matches. Additionally, larger scenarios tend to balance themselves if the allocated victory points are set correctly.

Kevin

< Message edited by kevinkin -- 12/14/2017 11:55:55 PM >

(in reply to Gunner98)

Post #: 22

RE: Loss Aversion Among Scenario Designers - 12/15/2017 3:45:27 PM

mikmykWS

Posts: 11524
Joined: 3/22/2005
Status: offline

quote:

ORIGINAL: Gunner98

Anyway, please don't go back and check my scenarios , as this is the first time I've actualy articulated this and most of the time its more feel than science. That's why play-testing is important.

B

Very much agree. Thanks Gunner!

The only reason I go back and check your scenarios is to find out why I lost

Mike

_____________________________

(in reply to Gunner98)

Post #: 23

RE: Loss Aversion Among Scenario Designers - 12/15/2017 4:07:11 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

quote:

ORIGINAL: kevinkin
Unfortunately, giving the AI (computer controlled forces) a material advantage is the primary way to give the human player a fight once they become proficient at Command.

That's not necessarily unrealistic, either. Only fools choose to fight fair fights. Defense is learning how to fight back against unfair fights. Offense, on the other hand, is about massing a sufficiently large force and concentrating a sufficient volume of firepower to overwhelm the defending adversary. Very few real battles are fought with one side or the other not possessing a substantial material advantage. "Fair" fights tend to be accidental meeting engagements with no decisive outcome.

(in reply to kevinkins)

Post #: 24

RE: Loss Aversion Among Scenario Designers - 12/15/2017 5:45:27 PM

Cik

Posts: 671
Joined: 10/5/2016
Status: offline

well, that's why the scoring is often relatively tilted- given overwhelming force requires the "victory" condition to be pushed beyond "achieved objective" to be challenging.

if you could do the military equivalent of just attack-moving towards the objective and win a triumph it wouldn't be hard enough to be replayable.

i mean granted in most scenarios i've played anyway, even what i would consider a rash streak of losses hardly counts as a defeat. instead you usually just win a marginal victory instead.

what scenarios in particular have you played that encourage overcaution? perhaps we're just not picking the same ones.

(in reply to SeaQueen)

Post #: 25

RE: Loss Aversion Among Scenario Designers - 12/15/2017 6:39:10 PM

pjb1

Posts: 676
Joined: 7/10/2013
Status: offline

I have made a few scenarios and am working on a campaign, in some of the scenarios I have no points for losses, the objectives are where the scoring is.
Before you use Mikes argument about completing the task but it costing a ton of resources, when I set this up I try to ensure if you have a bunch of losses
you wont be able to complete the objectives.

(in reply to Cik)

Post #: 26

RE: Loss Aversion Among Scenario Designers - 12/15/2017 8:40:02 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

quote:

ORIGINAL: Cik
well, that's why the scoring is often relatively tilted- given overwhelming force requires the "victory" condition to be pushed beyond "achieved objective" to be challenging.

That's assuming you're on the offense. If you're defending against overwhelming odds, then it would be disproportionate.

quote:

if you could do the military equivalent of just attack-moving towards the objective and win a triumph it wouldn't be hard enough to be replayable.

That might be a solution. The Vietnamese did it in the Vietnam War. The Chinese did it in the Korean War.

quote:

what scenarios in particular have you played that encourage overcaution? perhaps we're just not picking the same ones.

The one I was thinking of was "Operation Brass Drum." It's actually one of my favorites, but after playing it a few times, I started modifying it because I had some critiques. Example: Losing an F-35 results in the same point penalty as losing a CLF, an Amphib, or a DDG. YIKES!!!! To make matters worse, the only thing worth (maybe) risking an F-35 against is an SA-20. Even then, if you lose one of them destroying an SA-20, you get no net point benefit. If you lose more than one of them, you might as well not even attempt to strike them! The point system in that scenario essentially signals to the player that the F-35 is barely useful. Essentially, that's an expression of the scenario designer's opinion about the F-35. The point system essentially forces a tactical decision.

Similarly, losing a Hornet costs me 8 points but killing one of their fighters wins me only 2 points. That means I have to achieve an exchange ratio of greater than 4:1 just to not lose points in air to air combat, regardless of whether the enemy aircraft actually achieve their goal or not. So even if I defeat the enemy aircraft, I might still lose points! It raises questions about whether it's actually worth risking aircraft in air to air combat at all. The scoring system signals to the player a tactical decision. It's even worse with F-35s. I'd have to shoot down 25 other aircraft to justify the loss of a single F-35.

My opinion is that scoring shouldn't force tactical decisions on the player. It definitely ought not reflect a scenario designer's opinions regarding the tactical utility of a platform. That's what I mean by things seeming disproportionate. The scoring system shouldn't function to artificially constrain the player to make solve problems in specific ways. A good scenario (to me) is an opportunity to experiment with different approaches to a problem and then decide what works better and what works worse.

< Message edited by SeaQueen -- 12/15/2017 8:41:45 PM >

(in reply to Cik)

Post #: 27

RE: Loss Aversion Among Scenario Designers - 12/16/2017 1:33:41 AM

Cik

Posts: 671
Joined: 10/5/2016
Status: offline

well, i don't know about all that. i think the scoring is often more a "game" function than a sim function. one might ask why an F-35 costs 25x as much as a mig-21 or whatever- what's that based on? it's not based on cost (well, it could be but in this example it's not) it's mostly just based on the fact that the scenario designer thinks that given the challenge a player should not lose an F-35 often (if at all)

i guess look at it this way; you're playing a game wherein you are tasked with taking a city and you are given an infantry formation with tank support- the enemy has no tanks. while it's possible to lose the tank to fire (mines, RPGs, flipping it over or whatever) the scenario designer thinks that you shouldn't lose it and that given the calculus of the forces it should be possible for a skilled player to keep it intact the majority of the time. far removed from this is the fact that the army has thousands of these tanks or that they cost X amount a unit or whatever- mostly it's just that in this very focused arena the tank is a survivable unit and it would take a very unlucky fluke or a certain negligence on the player's side to lose it.

I don't think that the F-35 being expensive is a geopolitical commentary or a statement on the expense of 5th generation fighters. I think it's simply that the scenario designer wanted you to shepherd them carefully.

as to that specific case i think F-35 being worth 25 enemy fighters is basically pushing it. in principle though i don't object to this sort of concept.

the scoring i think is a surrogate for a wider, campaign-level view- in a real war, you keep your fighters around not because it will give you +5 points and a major victory but because you will probably need them later. the score is a substitute for that- "given the average losses we expected, you did better/worse than average." in the tank analogy, the large score penalty for losing the tank is a substitute for the later campaign level "you lost your abrams and there's a T-62 here now. good luck."

for this reason i think often score is mostly superfluous- if you lose an asset early in the fight the slap on the wrist in score you get often pales in comparison to the "well, how are you supposed to spot the TEL without your U-2, genius?"

there is probably an "ideal" way to design the score, but it's going to be really hard to actually come up with because of how diverse the environment can be.

just 2 cents though.

(in reply to SeaQueen)

Post #: 28

RE: Loss Aversion Among Scenario Designers - 12/16/2017 12:08:38 PM

SeaQueen

Posts: 1451
Joined: 4/14/2007
From: Washington D.C.
Status: offline

quote:

ORIGINAL: Cik
well, i don't know about all that. i think the scoring is often more a "game" function than a sim function. one might ask why an F-35 costs 25x as much as a mig-21 or whatever- what's that based on? it's not based on cost (well, it could be but in this example it's not) it's mostly just based on the fact that the scenario designer thinks that given the challenge a player should not lose an F-35 often (if at all)

That's exactly the problem. I have no idea what's it's based on. It's an arbitrary decision that says more about what the scenario designer thinks than it actually does about the "winability" of the scenario, the capabilities of the platforms in it, their relative costs, etc. If the point of the scoring system in a scenario is to arbitrate whether you win or lose the game, and the point of the game is simulate something, then the scoring system ought to realistically reflect what a "win" or "loss" looks like. If it doesn't somehow realistically reflect the risks and rewards of the scenario, then you end up with problems like I just pointed out (e.g. no target is juicy enough in the entire scenario to justify risking the platforms most capable of taking it out - the "give peace a chance" school of wargaming).

quote:

i guess look at it this way; you're playing a game wherein you are tasked with taking a city and you are given an infantry formation with tank support- the enemy has no tanks. while it's possible to lose the tank to fire (mines, RPGs, flipping it over or whatever) the scenario designer thinks that you shouldn't lose it and that given the calculus of the forces it should be possible for a skilled player to keep it intact the majority of the time. far removed from this is the fact that the army has thousands of these tanks or that they cost X amount a unit or whatever- mostly it's just that in this very focused arena the tank is a survivable unit and it would take a very unlucky fluke or a certain negligence on the player's side to lose it.

All this amounts to the scenario designer turning the tank (or F-35) into a golden magic BB. It's enormously capable, but I can't afford to actually use it, because if I use it and lose it, I suffer such an enormous cost that I can't make it up elsewhere. That's not a usable weapons system. If a real decision maker decided that a weapons system is a golden BB, then it ought to be held back for some other conflict where deploying it is worth the risk (i.e. it ought not be in the scenario).

quote:

I don't think that the F-35 being expensive is a geopolitical commentary or a statement on the expense of 5th generation fighters. I think it's simply that the scenario designer wanted you to shepherd them carefully.

It might not be intentional, but effectively they're signaling in the scoring system that it's too expensive to use.

quote:

the scoring i think is a surrogate for a wider, campaign-level view- in a real war, you keep your fighters around not because it will give you +5 points and a major victory but because you will probably need them later. the score is a substitute for that- "given the average losses we expected, you did better/worse than average." in the tank analogy, the large score penalty for losing the tank is a substitute for the later campaign level "you lost your abrams and there's a T-62 here now. good luck."

Potentially, it could reflect higher level requirements or metrics. In fact, I think the degree of victory (and therefore the score) ought to reflect the scenario's contribution to the larger picture. To get a good scoring system, you really need to decide up front what victory actually looks like. It might be like:

-Preserve 75% of all strike aircraft
-Lost no more than 1 CVN
-Lost no more than 2 large surface combatants
-Destroy the bridge X, Y, and Z
-Destroy 25% of the enemy tanks

Then you need to decide what contribution to victory or defeat each condition makes:

-Preserve 75% of all strike aircraft - 25%
-Lost no more than 1 CVN - 100% (i.e. if I lose more than a CVN, I lose the war)
-Lost no more than 2 large surface combatants - 25%
-Destroy bridges X, Y, and Z - 50%
-Destroy 30% of the enemy tank platoons - 50%

Suppose the score represented your "percentage of victory" then the max score would be 100 (%). That represents total victory.

If you lose two CVNs then you lose 100%. Your score is zero.
If you lose more than 75% of all strike a/c then you lose 25, and you score 75.
If you lose more than two large surface combatants you lose 25, and you score 75.
If you don't lose 2 CVNs, preserve the required number of aircraft and surface combatants, plus destroy the bridges or the tanks you gain 50 points. Your score is 50.
If you don't lose 2 CVNs, preserve the required number of aircraft and surface combatants, plus destroy the bridges and the tanks you gain 100 points. Your score is 100 (i.e. total victory).

Then, to check if it makes sense, you consider total defeat and the intermediate cases short of total victory:

If I lose 2 CVNs, 2 surface combatants, and more than 75% of your aircraft without achieving the other goals, you get a -150. That's more than achieving the other two goals combined, so you're still firmly in loss territory. That makes sense.

If I lose 2 CVNs, but preserve my aircraft or surface ships, then I'm still losing more than either the other two goals are worth in combination.

If I lose 2 CVNs, but preserve my aircraft and surface ships, then I score 0 if I somehow achieve the other goals.

If I preserve all my CVNs and ships, but lose more than 75% of my aircraft, while destroying the bridges then I score -25+50 = 25. Not great. A tactical level victory even if an operational loss.

If I preserve all my CVNs and ships, but lose more than 75% of my aircraft, while destroying the bridges and the tanks then I score -25+50+50 = 75. Better, but less than total victory. Still maybe an operational level loss, albeit maybe a more justifiable one.

That all makes intuitive sense!

All this would be easy to implement in LUA by implementing counters. That's not usually how people do it though.

quote:

for this reason i think often score is mostly superfluous- if you lose an asset early in the fight the slap on the wrist in score you get often pales in comparison to the "well, how are you supposed to spot the TEL without your U-2, genius?"

If it's superfluous then someone did something wrong. That means a player can't realistically judge how well they did based on their score, which defeats the purpose of having a score. A player ought not essentially "throw away" part of the experience because it's perceived to be useless, or it's unclear what it's actually simulating.

quote:

there is probably an "ideal" way to design the score, but it's going to be really hard to actually come up with because of how diverse the environment can be.

If don't know if there's an ideal way, but there seems to be a more rational one as I just outlined, provided one has a clear idea up front what the scenario is attempting to achieve. I suspect some of the disproportionate costs and risks I see in some scenarios might reflect a lack of forethought regarding what player victory looks like, or what the player is really trying to achieve in the scenario. Rather than writing the scoring triggers to reflect the destruction of individual units or platforms, it makes more sense to write them in terms of slightly more abstracted conditionals. Otherwise, the score amounts to just a weighted exchange ratio, which potentially looks distorted when you ask how many of platform X is platform Y worth risking to destroy.

< Message edited by SeaQueen -- 12/16/2017 12:09:10 PM >

(in reply to Cik)

Post #: 29

RE: Loss Aversion Among Scenario Designers - 12/18/2017 11:04:26 PM

BeirutDude

Posts: 2625
Joined: 4/27/2013
From: Jacksonville, FL, USA
Status: offline

Niger, There is a perfect example how we as a society have determined that any loss is too high in political and real capital. Now consider that 73 years ago today we were fighting the Battle of the Bulge and each day we were incurring casualties we would consider intolerable in today's combat. Lets switch our focus to just the conventional situation on the Korean Peninsula and how NKPA artillery on the DMZ and the political, economic and human cost it could take has stayed the U.S. hand for four decades! I could go on and on, but when I design a scenario I consider that there is more at play than just unit value.

Now looking first at units, I'll use an F/A-18F, represents...

1. The cost of the unit and the cost of it's eventual replacement
2. The cost of training and experience of the Pilot and possibly Radar Officer
3. A tangible loss of capability for the duration scenario.

Now while I agree with some of your points, I do try to use a logic into the point penalties/values. More or less from Med Tsunami 2019...

C-2 Greyhound/SH-60 10 pts
F/A-18 25 points
FFL/PCM type 100 to 150 pts
E-2D 200 Pts
DDG/SS/SSK 200-250
CG/SA-22 400-500
SA-400 1500
CVN 2500

Now the primary target, each dock in Latika is worth 1,000 points not because the docks are so valuable militarily but they represent so much in Russian prestige in the Middle East. The airbase in Khmeinen is really there for the defense of Latika which is a large part of why Putin is in Syria (not the only reason but a major one). The points for the F/A-18s are also there to discourage players for using them in a battle of attrition against the SA-400s. Without a penalty for loosing aircraft (FA-18s & Tornadoes in this case) they can be used to soak off the SA-400's Growlers and then the task groups TLAMs can ravage the Russian position once the Russian SAM missiles run out. So the points also prevent a gamey tactic that no real commander would consider. Heck I've even done it and used the British Tornadoes to cause the Russians to launch and as soon as I detect launch took off on an evasion course to use up some of the Russian ADA defenses. Sometimes it works and sometimes it costs me aircraft and victory points.

Another "gamey tactic" is moving all of your ADA to guard your primary assets (Latika) at the expense of important tactical units (an airbase). Thus all of it is worth points but the primary target more so players are not tempted to leave one or the other uncovered. I ask in Syria which would Putin either have defended Khmeinen AFB or Latika Naval Base or both? These are the choices a designer must make.

Some points assigned are political/public relations considerations. You're some high level analyst in D.C. if an American Harpoon hit a Disney cruise liner in the cross fire what kind of political damage would that cause? What would our enemies do with that in the realm of world opinion (especially if there are fatalities!)? So maybe that cruise liner being hit is worth 1,000 points (while a CVN is worth 2,500), not for the military value of the liner but the political and PR cost of the action/mistake.

So there is a logic and hierarchy to the point structures I use. Now YOU may not agree with their decisions, but every designer is doing their best.

See my point, it's not just what unit X is worth in a sterile spreadsheet but what the designer is trying to get across and how they see the world. One other thing to consider, but designing a scenario represents weeks of spare time dedicated and everyone does their best to produce scenarios they believe are both entertaining and hopefully thought provoking.

< Message edited by BeirutDude -- 12/19/2017 12:07:46 AM >

_____________________________

"Some people spend an entire lifetime wondering if they made a difference. The Marines don't have that problem."
PRESIDENT RONALD REAGAN, 1985

I was Navy, but Assigned TAD to the 24th MAU Hq in Beirut. By far the finest period of my service!

(in reply to SeaQueen)

Post #: 30

Page: [1] 2 next > >>

All Forums >> [New Releases from Matrix Games] >> Command: Modern Operations series >> Mods and Scenarios >> Loss Aversion Among Scenario Designers

Page: [1] 2 next > >>