Better Artillery Scatter Rules

[Alsadius]

The existing artillery scatter rules bug me - they're simple to resolve, but the results are grossly unrealistic, because it always scatters in one of the six cardinal directions. That makes it fairly easy to game, and it also results in some pretty wonky distributions of where scattered shots go.

I have a proposed rule to change this. Whatever the MOF on the attack roll is, take twice that many dice and roll them. Each die scatters the shot one hex in the direction rolled, using the same scatter direction table as the vanilla rules. (If you're rolling physical dice, you simplify this quickly by cancelling out pairs - a 1 and 4 will cancel, a 2 and 5 will cancel, etc.). This results in a vastly more plausible scatter chart - see the images below.

I'll admit that this isn't perfect. If you're playing tabletop, there is more work involved in resolving arty scatter under these rules, and it does reduce average scatter distance somewhat(it's about 4 hexes on average in vanilla and 2.5 under my proposal). That said, I think it's worth it to get rid of the gameyness of the vanilla rules.

I've drawn up some pictures below of what it looks like - this was done in Excel, so the "hexes" are rectangular, but you get the idea.

This is the status quo:


This is my proposal:

[theagent]

Huh. Hadn't thought of that before, but it make some sense.  Two question X:

• Why roll double the dice?  Was it still too wonky to roll (MoF) x 1D6?
• Should there be a rule that, on all rolls after the 1st, you have to reroll any 'reversal' results?  For example, say you need to roll 6 dice.  Roll #1 is a 4, which moves you 'down' from the target hex. Roll #2, though, is a 1...which counteracts roll #1 because you move back 'up' to the target hex.

[scatcat]

I like this.

a lot.

[Alsadius]

Huh. Hadn't thought of that before, but it make some sense.  Two question X:

• Why roll double the dice?  Was it still too wonky to roll (MoF) x 1D6?
• Should there be a rule that, on all rolls after the 1st, you have to reroll any 'reversal' results?  For example, say you need to roll 6 dice.  Roll #1 is a 4, which moves you 'down' from the target hex. Roll #2, though, is a 1...which counteracts roll #1 because you move back 'up' to the target hex.

1) Because the actual amount of scatter is fairly low if you just roll 1x the dice. With the vanilla rules, average scatter is 4 hexes. With my proposal, it's 2.5 hexes, which is already a bit low - remember that arty has splash, and missing by 1 is often almost as good as hitting directly. With MoF x 1d6, it's 1.8 hexes. I don't want arty to be too reliable, so I figured 2x MoF was a good compromise between needing fistfuls of dice and a computer to count them all on one hand or just having the scatter average out to nearly nothing on the other. (For reference, 3x MoF is about 3.1 hexes of average scatter).

Phrased differently, if you're firing a Long Tom that has splash range of 2, there's a 28% chance of doing at least some damage under vanilla rules, 74% with 1x MoF, 54% with 2x MoF, and 43% with 3x MoF.

2) That only makes sense if you roll sequentially. I figure you'll just grab the right number of dice and roll them all at once, then group and cancel. If you roll them all at once, your idea doesn't work. But if you do intend to roll one after another, you can try it. I'd only roll 1x MoF dice in that case, as eliminating cancellation will increase average drift substantially.

I like this.

a lot.

 ;D

[theagent]

Ok, that makes more sense.

Just let me make sure I have it right:

Assuming MoF 3 (6 dice), & assuming target hex is hex 0809 on a map sheet:

• Roll 1, 1, 4, 1 5, 4:  the 4s & two 1s cancel, leaving a 1 (up) & 5 (left-down). The shell falls in hex 0709 (1 hex away)
• Roll 4, 6, 1, 5, 4, 6: the 1 & a 4 cancel, leaving a 4 (down), 5 (left-down), & two 6s (left-up).  The shell falls in hex 0510 (3 hexes away)
• Roll 4, 3, 4, 3, 6, 3:  the 6 & a 3 cancel, leaving two 3s (right-down) & two 4s (down). The shell lands in hex 1012 (4 hexes away)

[Alsadius]

Ok, that makes more sense.

Just let me make sure I have it right:

Assuming MoF 3 (6 dice), & assuming target hex is hex 0809 on a map sheet:

• Roll 1, 1, 4, 1 5, 4:  the 4s & two 1s cancel, leaving a 1 (up) & 5 (left-down). The shell falls in hex 0709 (1 hex away)
• Roll 4, 6, 1, 5, 4, 6: the 1 & a 4 cancel, leaving a 4 (down), 5 (left-down), & two 6s (left-up).  The shell falls in hex 0510 (3 hexes away)
• Roll 4, 3, 4, 3, 6, 3:  the 6 & a 3 cancel, leaving two 3s (right-down) & two 4s (down). The shell lands in hex 1012 (4 hexes away)

You have it right. You can see why I say that this is both more time-consuming than a vanilla roll(where once you have the MoF, you only roll a single die - you might get a 2, for right-up, and then the shot will land in hex 1108, 3 hexes away), and why I want to have more dice than the MoF due to a lot of dice cancelling out in various ways. It shouldn't be too alien to anyone who's ever played with aero units, where similar cancellation rules apply, but it's a bit more bookkeeping. I think it's worth it, though.

[TigerShark]

Problem that I can see is that (for example) if you get enough 1 and 4 combinations, regardless of the MOF, you can eventually hit the target hex even though you supposedly missed. Ex.: 1, 1, 4, 1, 4, 4

I think you should require that, "after rolling the first die, determine the opposite side from this roll. All numbers corresponding to this roll must be re-rolled."

i.e.: MOF is 3. You get 6, 6-sided dice to determine the drift. On the first roll, you roll a 2. Rolls opposite of this should be a 5.

2, 3, 1, 2, 5, (re-roll), 6, 6

Hits 3 hexes away from the target.

[monbvol]

Or as another alternative roll a matched pair for each MoF.  The first determines direction and the second how many hexes the shot scatters in that direction.  That way it is very unlikely a shot will still scatter back into the intended hex.

[Alsadius]

Yeah, scattering back to the "missed" hex is a thing that can happen. The odds aren't huge - about 4% of shots will be "misses" that hit the target head-on - but that's not zero. IMO, just make it so that a "miss" that scatters back onto the target should not count as a hit for the purposes of making it an auto-hit hex in future, because you hit it by luck instead of skill.

It doesn't bother me much, TBH, but if it bothers you there's a simpler way to resolve it - if it lands back on the target after you missed, just throw one extra die. That guarantees that it'll wind up one hex away. (Or just up the arty penalty from +7 to +8, and the odds of hitting dead-on even with the chance of a zero-distance scatter will be lower than in vanilla.)

Edit: Also, re-rolling the opposed direction won't prevent drifting back to zero. If you roll 1, 3, 5 for example, you'll be back where you started.

[RunandFindOut]

Or as another alternative roll a matched pair for each MoF.  The first determines direction and the second how many hexes the shot scatters in that direction.  That way it is very unlikely a shot will still scatter back into the intended hex.

That's actually the way I've done things for ages, as a variant of a house rule created at my table back during D&D1e days for judging where stones thrown by siege engines landed.  Two differently colored dice(or sets of dice) one for direction one for distance thrown at the same time.

[Simon Landmine]

Nice system.
(Also, kudos on those Excel diagrams.)

[Daemion]

Or as another alternative roll a matched pair for each MoF.  The first determines direction and the second how many hexes the shot scatters in that direction.  That way it is very unlikely a shot will still scatter back into the intended hex.

Isn't this what was already done in Tac Ops?
 
Edit: One of the horrible things about it was the fact that you could accidentally return the shot to sender if you rolled all 4s, and they were just at the edge of on versus off board artillery ranges.

This keeps the drift close to target and not allow for the wind to carry the shot back to your tube if it's close enough.

[Daemion]

Question:

Have you found that order matters? Or, if you simply pick up a fist full of dice and map it out in any starting way, does it end up in the same spot?

[PurpleDragon]

Lurking:  I think the MegaMek guys might like to see this. 

[Alsadius]

Question:

Have you found that order matters? Or, if you simply pick up a fist full of dice and map it out in any starting way, does it end up in the same spot?

Order doesn't matter, for either my system or monbvol's(aside from making sure that each distance/direction pair stays paired in his system).

Lurking:  I think the MegaMek guys might like to see this.

Might be a fun little optional rule. I'm in with the MegaMekNet guys, but not MM proper. Any idea how to bring this to their attention?

[PurpleDragon]

Might be a fun little optional rule. I'm in with the MegaMekNet guys, but not MM proper. Any idea how to bring this to their attention?

I'm not sure, but I thought Mombvol was actively talking with them(if not one of them). 

[monbvol]

I can tinker at best as far as coding goes these days and I have no more direct access/influence over the MM devs than anyone else on these forums.

[PurpleDragon]

OK, I'm sorry.  :-[ I thought you were talking with them.  There is a MegaMek child board under computer games.

[monbvol]

Nothing to apologize for.

I do pop up in that area of the forum now and again and do give the odd suggestion/bit of feedback.

[idea weenie]

Very sneaky and clever method of getting scatter without getting the halo effect of the current scatter rules.