Autofire Question

[tbeard1999]

A question and an observation:

Q1. How do "to hit" modifiers apply to autofire attacks, if they apply at all. (It would be helpful for the example to include the effects of the firer's Rifle skill).

Observation. This mechanic looks like it's defective. I can't tell for sure until the question above is answered.

To hit, you roll dice equal to the autofire rating of the weapon (usually 3 or 4). The lowest die is automatically the timing die. The remaining dice are individually paired with the timing die to produce several “to hit” rolls.

Here’s the example from the rules:

Morn fires his Advanced Combat Rifle (Automatic 4). He rolls four d6, and gets a 2, a 3, a 4, and a 5. The 2 is the lowest die, so it becomes the Timing die for the weapon. The other dice are Effect dice. Morn assembles three attacks 2+3=5, 2+4=6, and 2+5=7.

The problem is that there is a 52% chance of rolling at least one “1” on 4d6. And you cannot roll an 8+ if one of your dice is a 1! So poor Morn has a 52% chance of missing all three shots, just by rolling a 1 if he has no bonuses to the roll. Worse, if he fails to roll at least one "1", he has a 52% chance of rolling at least one "2", which is nearly as bad. If one die is a two, the other die *has* to be a "6" to score a hit.

When time permits, I'll run the statistics out. But right now, it looks like Morn has a better chance of hitting his target by firing 1 bullet rather than 12 bullets. And his chance of a miss *increases* as he fires more rounds...

I should add that *if* modifiers are added to the effect dice, then it will be nearly impossible to hit with autofire if there's a net negative modifier. A -1 modifier will make it impossible to score a hit if the timing die is a 1 or 2. There's an 81% chance of rolling at least one "1" or "2" on 4d6. So poor Morn will miss at least 81% of the time in that case.

 

[The Greek]

This is my take, and it is how I see autofire in the real world (more or less).

Blasting away with autofire is a pretty poor way of hitting something, especially without training (think of the Somalis in Black Hawk Down - lots of blasting with their AKs but very few hits (largely through volume of fire)). All would be familiar but untrained with their weapons, so the Somalis would have a nominal weapon skill of 0.

With autofire you are hosing down an area or trying to suppress the enemy (who will duck down to avoid getting hit by the sheer volume of fire coming his way). You are not aiming all that much.

Now take a trained soldier (skill 1) or a highly talented and skilled one (skill 2 or more) and the odds improve big time. Also taking time with aiming actions and the chances of hitting really start to go up.

Burst fire is far better (multiple shots into a single target - really ruins a targets day). You get the benefits of extra damage reflecting multiple hits to a target, but you are going for the single guy, aiming properly, not the far more wild shooting reflected by autofire.

The session we had today shower that aimed burst fire is absolutely deadly - as it should be, way better than single shots.

Leave autofire to Hollywood or useless mooks, burst fire is king! (Unless you are a marine with a skill of 2 or more, then you can blast away in any way you please).

 

[AKAramis]

A question and an observation:

Q1. How do "to hit" modifiers apply to autofire attacks, if they apply at all. (It would be helpful for the example to include the effects of the firer's Rifle skill).

Observation. This mechanic looks like it's defective. I can't tell for sure until the question above is answered.

To hit, you roll dice equal to the autofire rating of the weapon (usually 3 or 4). The lowest die is automatically the timing die. The remaining dice are individually paired with the timing die to produce several “to hit” rolls.

Here’s the example from the rules:

Morn fires his Advanced Combat Rifle (Automatic 4). He rolls four d6, and gets a 2, a 3, a 4, and a 5. The 2 is the lowest die, so it becomes the Timing die for the weapon. The other dice are Effect dice. Morn assembles three attacks 2+3=5, 2+4=6, and 2+5=7.

The problem is that there is a 52% chance of rolling at least one “1” on 4d6. And you cannot roll an 8+ if one of your dice is a 1! So poor Morn has a 52% chance of missing all three shots, just by rolling a 1 if he has no bonuses to the roll. Worse, if he fails to roll at least one "1", he has a 52% chance of rolling at least one "2", which is nearly as bad. If one die is a two, the other die *has* to be a "6" to score a hit.

When time permits, I'll run the statistics out. But right now, it looks like Morn has a better chance of hitting his target by firing 1 bullet rather than 12 bullets. And his chance of a miss *increases* as he fires more rounds...

I should add that *if* modifiers are added to the effect dice, then it will be nearly impossible to hit with autofire if there's a net negative modifier. A -1 modifier will make it impossible to score a hit if the timing die is a 1 or 2. There's an 81% chance of rolling at least one "1" or "2" on 4d6. So poor Morn will miss at least 81% of the time in that case.

The DM is applied to each "pairing":

I'll give an example based upon how I read it:

5d autofire weapon
Rolls 1,3,4,5,6
Range DM-1
Aim DM+2
Skill 0
Net DM=+1
THis means 4 shots went down range.
The low die becomes the time die
Shot 1: t1+e6+dm1=8 Hits, for effect 7, capped at 6
Shot 2: t1+e5+dm1=7 miss, effect 6
Shot 3: t1+e4+dm1=6 miss, effect 5
Shot 4: t1+e3+dm1=5 miss, effect 4

Next burst, no aim, and target in cover,
Range DM-1
Cover DM-1
Skill 0
Net DM-2

Rolls 3,3,4,5,6
3+6-2=7 miss, effect 4
3+5-2=6 miss, effect 3
3+4-2=5 miss, effect 2
3+3-2=4 miss effect 1 (Ouch, FUMBLE!)
Timing is 3, minus recoil

Next burst, same as previous (DM-2)
4,5,5,6,6
4+5-2=7 Miss, effect 3
4+5-2=7 Miss, effect 3
4+6-2=8 Hit, effect 4
4+6-2=8 Hit, effect 4

 

[tbeard1999]

The DM is applied to each "pairing":

I'll give an example based upon how I read it:

5d autofire weapon
Rolls 1,3,4,5,6
Range DM-1
Aim DM+2
Skill 0
Net DM=+1
THis means 4 shots went down range.
The low die becomes the time die
Shot 1: t1+e6+dm1=8 Hits, for effect 7, capped at 6
Shot 2: t1+e5+dm1=7 miss, effect 6
Shot 3: t1+e4+dm1=6 miss, effect 5
Shot 4: t1+e3+dm1=5 miss, effect 4

Well, there are no special modifier rules in the Autofire section. Therefore, it's reasonable to assume that modifier apply to autofire to hit rolls like any other to hit roll.

If that is the case, it is virtually impossible to hit anything if there is even a -1 to hit modifier.

If a 1 or 2 is rolled, the to hit number cannot be 9+. And on 4d6, there's less than a 3% chance that a 1 or 2 won't be rolled.

I created a Monte Carlo simulation to roll 10K attacks at a time, assuming this is how it works. Here's the result:

-2 on the "to hit" roll, <0.1% chance of scoring a hit. Average number of hits is 0.002. Compared with 16.7% chance of hitting with a single shot.

-1 on the "to hit" roll, 5% chance of scoring a hit. Average number of hits is 0.09. Compared with 27.8% chance of hitting with a single shot; average hits 0.278.

No modifier on the "to hit" roll, 33% chance of scoring a hit. Average number of hits is 0.6. Compared with a 41.7% chance of hitting with a single shot; average hits 0.417.

+1 modifier on the "to hit" roll, 85% chance of scoring a hit. Average number of hits is 1.7. Compared with 58.3% chance of hitting with a single shot; average hits 0.583.

+2 on the "to hit" roll, 99% chance of scoring a hit. Average number of hits is 2.5. Compared with 72.2% chance of hitting with a single shot; average hits 0.722.

+3 on the "to hit" roll, 99.999% chance of scoring a hit. Average number of hits is 2.999. Compared with 83.3% chance of hitting with a single shot; average hits 0.833.

So...

1. Lousy shooters are far better off using single shots than autofire. Good shooters are far better off using autofire. (The exact opposite of what seems to be happening in the Real World; highly trained Allied troops in Iraq tend to use single shots, while poorly trained terrorists use autofire).

2. A mere +1 modifier can lead to a 6X increase in the chance of hitting. A +1 net modifier will result in a hit almost always, while a -1 net modifier will result in a miss 95% of the time. And a -2 net modifier makes a hit *impossible*.

3. The more bullets you fire, the less likely you are to hit in cases where there's no modifier or a negative modifier. With no modifier, an autofire of 3 will hit 37.8% of the time. With a -1, you'll hit 9.6% of the time (vs. 5% of the time with Autofire 4).

Is this really a good way to handle autofire?

 

[Settembrini]

Let´s just be honest: The autofire rule was never put under such close scrutiny as it was in this thread, I´m sure.

The only reason it exists is because of the timing/effect dice tricks. They had the idea, rolled with it, and now seem to be stuck with it.
It´s a bit like T4 with the multiple and half dice tricks...

 

[SableWyvern]

I can't recall off-hand the difference between the actual rules atm, and a few mods I suggested a while back, but I think Gar introduced a burst fire rule seperate to autofire as being discussed in this thread.

If that is the case, then I think the burst fire rule should be considered the norm for what should be relatively accurate, short aimed bursts, and the rule under discussion in this thread be considered for use in more wild-fire situations. In that case, there is still a question as to whether it is too accurate at high positive DMs, but at the lower end seems more than reasonable.

 

[SableWyvern]

As an aside, the type of probability spread shown in this thread means the autofire rules are emininently suitable for simulating double-taps with semi-automatic weapons -- unskilled firers are very likely to be over-tense and snatching the trigger while firing double-taps, while a skilled marksman increases his chance of a hit.

 

[tbeard1999]

As an aside, the type of probability spread shown in this thread means the autofire rules are emininently suitable for simulating double-taps with semi-automatic weapons -- unskilled firers are very likely to be over-tense and snatching the trigger while firing double-taps, while a skilled marksman increases his chance of a hit.

The problem -- which seems pervasive in the combat system as a whole -- is that the autofire rules do not accurately model what autofire looks like in the Real World. In the Real World, highly trained troops tend to prefer single shots, while poorly trained rabble prefer autofire.

In this system -- ignoring the statistical absurdities -- rabble are far better off firing single shots while highly trained troops are better off firing autofire.

So while post-hoc rationalizations can be conjured up, the reality is that this system appears to be (a) illogical (mainly due to the fact that a minor modifier can make a HUGE difference in effectiveness); (b) overly fussy (a problem with the entire combat system IMHO); and (c) a very poor model of what actually happens.

 

[SableWyvern]

In this system -- ignoring the statistical absurdities -- rabble are far better off firing single shots while highly trained troops are better off firing autofire.

So while post-hoc rationalizations can be conjured up, the reality is that this system appears to be (a) illogical (mainly due to the fact that a minor modifier can make a HUGE difference in effectiveness); (b) overly fussy (a problem with the entire combat system IMHO); and (c) a very poor model of what actually happens.

Eh? Why's that a response to my post, which was specifically discussing double-taps, not automatic fire?

 

[tbeard1999]

In this system -- ignoring the statistical absurdities -- rabble are far better off firing single shots while highly trained troops are better off firing autofire.

So while post-hoc rationalizations can be conjured up, the reality is that this system appears to be (a) illogical (mainly due to the fact that a minor modifier can make a HUGE difference in effectiveness); (b) overly fussy (a problem with the entire combat system IMHO); and (c) a very poor model of what actually happens.

Eh? Why's that a response to my post, which was specifically discussing double-taps, not automatic fire?

Because your response does not really address the criticisms I made, so I just wanted to make that point clear.

I suppose the mechanic could be used to simulate a double tap (2 shots fired quickly at the same target). However, you'd have to apply a negative modifier, or no one would ever choose to fire a single shot. And if your modifiers slip below 0, no chance of success. The current system is not tuned finely enough for this to be a viable mechanic in my opinion. Also, you'd have to change the ammo expenditure to two shots.

And of course, you'd have to find a different autofire mechanic, since this double-tap mechanic would look much like Autofire 3.

In a different combat system, double taps and autofire could be handled easily. The timing/effect system works best for single shots; it cannot easily handle multiple shots. Which raises the question of why it's integrated into the combat system at all, since Traveller has a wide variety of automatic weaponry.

 

[SableWyvern]

Off the cuff, my "solution" would be to make the Optium range for all autofire Close or Personal. This makes automatic fire effective at very short ranges, while the range penalties quickly mount up to make it useless at longer ones.

Once the numbers have been crunched, my instints on this one might turn out to be wrong, however.

 

[tbeard1999]

Off the cuff, my "solution" would be to make the Optium range for all autofire Close or Personal. This makes automatic fire effective at very short ranges, while the range penalties quickly mount up to make it useless at longer ones.

Once the numbers have been crunched, my instints on this one might turn out to be wrong, however.

Since a -1 net to hit modifier effectively makes it impossible to hit, your idea would likely make automatic weapons fire useless at medium range+, which does not really model reality very well.

 

[Klaus Kipling]

I think you have a point with this one, tBeard.

I can think of a possible fix.

Choose the timing die just like normal - no 1's for timing then, if the player isn't asleep. Then reduce final timing by recoil + no. of dice rolled.

This would increase the number of hits (surely the point of auto-fire), but the damage each hit does would be more random, and there'd still be the inbuilt flat-footedness after the volley.

Otherwise, tBeard is correct, the more shots that are fired, the less chance any will hit.

(Auto 5 is pretty much certain to roll at least one 1, negating all shots unless there's a positive mod).

 

[tbeard1999]

I think you have a point with this one, tBeard.

I can think of a possible fix.

How many fixes are gonna be suffered before someone considers eliminating the mechanic?

At this rate, the timing/effect mechanic will be riddled with exceptions, special cases and variations. (Assuming, of course, that Mongoose implements any of the proposed changes).

And the problem with choosing the timing die is that this allows a player to virtually guarantee hits by always choosing a high die. Just as there's a high chance of rolling at least one "1" on 4d6, there's an equally high chance of rolling at least one "6" on 4d6. In an autofire-4 attack, that would mean that you'd usually score 3 hits even assuming -2 (and you're guaranteed to score 3 hits if you have a +1 modifier).

It's a valiant effort, but I don't think that this gets it done.

 

[Klaus Kipling]

Um, as far as I can see, this is really the only fix required. I don't see any other special cases.

The rest is just balancing modifiers.

And so what if some 'auto' hit. Autofire is meant to hurt. The target is lucky if it isn't hit if you're lobbing off a dozen rounds at short range.

Anyway, ALL game mechanics are 'riddled' with exceptions and special cases - it's what gives them nuance and subtlety.

And, a player might not pick that 6 anyway - with the recoil mods it's probably gonna end up a 1 for timing no matter what is rolled, and as long as it hits a 6 effect might be more useful (ie: damaging), especially against armour.

Once player choice comes into it just looking at raw probabilities isn't as helpful for analysis.

 

[tbeard1999]

Um, as far as I can see, this is really the only fix required. I don't see any other special cases.

The problem, as noted above, is that your fix does not fix anything. It merely breaks the system differently. And unless your fix is implemented in the final version of MGT, the game system is still defective.

And if a system results in automatic hits most of the time, why not throw out the fussy resolution mechanics and simply make the hits automatic? Seems a sloppy design philosophy to create a fussy system that will yield almost completely predictable outcomes.

 

[Klaus Kipling]

Well, not sure it breaks anything, since it corresponds to the normal method, which the autofire rules a written deviate from.

And broken as compared to what?

CT was full of auto-hits and auto-misses, depending on weapon type, armour and range. Just see Snapshot for the net targets.

In MT tasks were either average, so easy as to be automatically successful, or so hard as to be almost impossible. Not much granularity there.

What you're failing to realise is that the net modifier is not the be all and end all. That number is derived from several factors, including skill, range, dodging, visibility, etc, which make every tactical situation different, and in any situation the net modifier is likely to be varied.

For instance, the player might not need to hit to succeed at his intent - if shooting at someone forces them to dodge then that might even be more useful than hitting them for a few points of damage.

The advantages of this mechanic far outweigh any disadvantages, as is apparent to all those here who've actually playtested the game using it, rather than just coming up with often spurious probability-crunching exercises.

With timing and effect so many choices open up to the player that using just pure probability as a benchmark is inadequate.

After all, it's not as if the probabilities for success at anything in any rpg reflect reality in any way. It's good enough for a game.

In this specific case your analysis was correct, essentially because player choice was taken out of the equation. What looked good at first glance was shown to result in the opposite of the intention. I've offered one possible fix, one that is more in keeping with the base mechanic. I can think of a few others too.

At this late stage your suggestion of ripping up the rules and starting again is never gonna happen.

Most people here like the mechanic, basically because it is interesting and inherently sound. All that's left is to tweak the numbers.

And even for those that do not like the mechanic, house-ruling back to a simple 'roll 8+ on 2D' is easy as pie.

 

[tbeard1999]

Well, not sure it breaks anything, since it corresponds to the normal method, which the autofire rules a written deviate from.

And broken as compared to what?

CT was full of auto-hits and auto-misses, depending on weapon type, armour and range. Just see Snapshot for the net targets.

The games you reference are 20-30 years old. I expect a new game to be better; otherwise what's the point?

The advantages of this mechanic far outweigh any disadvantages, as is apparent to all those here who've actually playtested the game using it, rather than just coming up with often spurious probability-crunching exercises.

Still waiting for someone to explain what the purported "advantages" are. The results skew overwhelmingly to exceptional successes. Why not simply ditch the system and assume exceptional successes. That's what the system will yield in most cases.

Indeed, it seems painfully obvious that the designer is unaware that the normal d6 probability spread does not apply to the timing and effect die.

And I have played the system. Although it isn't strictly necessary to taste soured milk to realize it's gone bad; the same is true of games.

As far as "everyone likes it", well, let's see how well it sells. I think that non-anonymous comments are about as useful as non-secret ballots. Nor is there any evidence that the denizens of this forum constitute a statistically representative sample of the gaming public (or indeed of whatever target market there is for the game).

And I note that you yourself have agreed that the mechanic is defective. The difference is that you seem comfortable with Mongoose publishing (what you agree is) a defective mechanic because you think that you can fix it with house rules. The problem, of course, is that your house rules are not the system that is being published.

And while I don't mind a few rough mechanics in a game, I do get nervous when the base mechanic is defective and when even its defenders seem only to be able to say, "well, it is defective, but it can fixed if you do this <insert "fix" that does not actually fix the problem (or introduces problems that are at least as bad) but that does seriously alter the mechanic>."

It's good enough for a game.

A line that has doubtlessly been offered to excuse countless badly conceived mechanics. Hopefully the professionals at Mongoose won't resort tp this...

At this late stage your suggestion of ripping up the rules and starting again is never gonna happen.

<shurg> Not my problem. Although I am rather skeptical that Mongoose can afford to release a game that has known serious defects in its core systems. I mean, who do they think they are, Microsoft?

Most people here like the mechanic, basically because it is interesting and inherently sound. All that's left is to tweak the numbers.

"It's broken and inherently sound." Right. Still waiting for that magical "tweak".

And even for those that do not like the mechanic, house-ruling back to a simple 'roll 8+ on 2D' is easy as pie.

Except that you have to toss out the entire MGT combat system in the process. If Mongoose Traveller can't provide us with a good combat system and a functional task system, then what good is it, exactly?

Do we really need another crappy version of Traveller? As someone who'd buy everything if the new version was even adequate, I'd happily wait an extra month or two to get a decent game. But I get bitter when a company foists a half-baked design on me that has known, serious flaws in it. Maybe you don't mind a company treating you like an idiot; but I (and many consumers I suspect) do.

Now, if you honestly believed that the mechanic was fine, then we could agree to disagree. I could be wrong that it is not enjoyable. But you have already conceded that the mechanic is defective. Which makes your subsequent equanimity rather perplexing. As I noted to another person in a different forum:

[Y]our fixes are not part of the game that Mongoose is publishing. So your defense is equivalent to someone saying "yes, I agree that this car is badly designed and poorly made. But I have a machine shop and am willing to re-engineer it to my taste, so I think it's a fine product. I am also untroubled by the fact that the manufacturer is producing a product that it knows (and that I concede) is defective."

I find such a position absurd on its face.

 

[AKAramis]

In MT tasks were either average, so easy as to be automatically successful, or so hard as to be almost impossible. Not much granularity there.

Wrong. MT's task system has an autofail on natural 2; and a natural 2 aggravates any failure where a 3 would fail.

So no matter how good the character, under MegaTraveller, 1/36 of tasks fail.

 

[Klaus Kipling]

No, I have not once conceded that the mechanic was broken. If you're going to misrepresent what I've said it doesn't help your credibility. :roll:

I said, that in the case of autofire, your concerns were correct.

This, in no way, is saying that the base timing/effect mechanic is broken, just misapplied in this instance.

At no point have I said my fix is part of the rules, just a suggestion on my part, as is proper on a playtest forum.

As you are in the vast minority on this playtest forum in decrying the mechanic, then my statement "everyone likes it" can stand. I see no evidence here to the contrary.

If most people on this playtest are not complaining about it, I can safely assume most folk here are happy with it. I only see 1 lone voice claiming the mechanic is broken.

The advantages to timing/effect are pretty obvious. The player has some control over a random roll, actions can be weighed against possible outcomes (I can duck but lose my opportunity to fire or risk the hit to make an aimed shot), Tactics and Leadership have some real effect without fuss, recoil is incorporated into combat, and all this without complex bookkeeping.

There is some more work to do. The effect die descriptions could do with some work, it's possibly too easy to fumble, and weapon damage needs tweaking.

I'm adding my thoughts and ideas to fixes to the email I will send in to Mongoose as per request for feedback. Whether they are taken up is up to the powers that be. I'm not going to spit out my dummy if they're not.

The basic concept of the mechanic is sound. What's more all the rules in this version of Traveller are highly compartmentalised, meaning that if there's a particular rule that a particular group does not like, it can be swapped out without messing up other parts of the game.

If you don't like timing and effect, you could just use a regular CT style mechanic, which does not break combat (you'd use the mods and actions that was in your old CT mechanic). For weapon damage, just roll a die for every 1x. Simple.

[Y]our fixes are not part of the game that Mongoose is publishing. So your defense is equivalent to someone saying "yes, I agree that this car is badly designed and poorly made. But I have a machine shop and am willing to re-engineer it to my taste, so I think it's a fine product. I am also untroubled by the fact that the manufacturer is producing a product that it knows (and that I concede) is defective."

I find such a position absurd on its face.

It's pretty sad you have to stoop to this level of dishonesty to try and prove your point. This misrepresents my viewpoint beyond breaking point. If you're trying to persuade me of the merits of your argument then you're going about it in exactly the wrong way.