Falcon Killing: ROI Analyses

Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you:

No adverts like this in the forums anymore.
Times and dates in your local timezone.
Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance.
Email notifications for threads you want to watch closely.
Being a part of the oldest wargaming community on the net.

If you are already a member then feel free to login now.

**JTS1486**

Falcon killing as a cost-effectiveness ratio measure. I'm going to use "Immobilised or Destroyed" as my measure of an effective "kill" because after that the tank and unit transported are no longer an offensive threat, and are just gumdrops that you can pick up later with any handy weapons.

Marine Lascannon - comes in a Tac squad that costs 115 points. (6 marines, Lascannon, Plasma Gun)

2/3 * 2/3 * 1/9 = 4/81 (or 1/20) - common math that we all know. Looking at the Average Kitted Falcon Cost of 210 points, let's say then that each turn/shot from this squad nets 10.5 points of effective gain. Thus, this squad, if it shot at Falcons all game, would earn 63 points back - which isn't particularly effective. The goal though is to find the MOST cost-effective anti-falcon weapon in your army which you can maximize, so maybe it won't be so bad overall.

If you add in the Plasmagun, 2/81 (or 1/40), the sum is 2/27 for a return of 15.5 pts per turn (ppt), or 93.33 pts per game. Better, but consider that you're not shooting that plasmagun at a Falcon every turn due to range restriction.

Divide your points returned versus unit cost for a Return On Investment ratio, where higher is better (over 1 is a Good Buy).
LC/PG Tac: 63/105 = 0.6 ROI

Now, I don't care so much about MEq versus Mech Eldar, because I'm not a marine jockey. What I would care most about are Tau effectiveness metrics. Check this out:

TLMP: 3/4 * 1/3 * 2 * 1/9 = 1/18 ... *210 = 11.66 ppt
MP: 1/2 * 1/3 * 2 * 1/9 = 1/27 ... *210 = 7.77 ppt
FB: 1/2 * 1/2 * 1/9 = 1/36 ... *210 = 5.833 ppt
RFPR: 1/2 * 1/6 * 2 * 1/9 = 1/54 ... *210 = 3.88 ppt
PR: 1/2 * 1/6 * 1/9 = 1/108 ... *210 = 1.944 ppt
HHRG: 2/3 * 5/6 * 1/9 = 5/81 ... *210 = 12.96 ppt
SniperUnit: 1/2 * 1/6 * 3 * 1/9 = 1/36 ... *210 = 5.833 ppt

Now that we have that, let's look at some units and configs.
That HammerHead Railgun is a classic antitank gun: 12.96 ppt means a whole game of shooting Falcons only returns 77.77 pts. Not exactly earning its keep there at 180 pts for the HHRG: 0.43 ROI.

More ROI numbers:
Sniper Unit: 35ppg / 80 pt unit = .4375 ROI - shoot your Snipers elsewhere if they have targets. Note that this does not take into account the ROI from successful marker hits on other units, because it is best to calculate this as a separate effect per enhanced unit.
Sunfire Crisis PR/FB - assume DS at close range, so RFPR/FB: 58.326ppg / 62pt model = 0.94 ROI
(pretty good, except you're only getting to do this ONCE per game, so it's unlikely to work out this way.)
Fireknife Crisis PR/MP - assume long range: same! 0.94 ROI
(THIS you could actually get, because you'll be shooting these all game if you can keep them alive.)
TLMP/Flamer Crisis is a 47 pt model: 70ppg/47pt model = 1.489 ROI

I think we have a winner - and it's something every Imperial Guard player already knew. Use Autocannons (I mean, Missile Pods) to take down Falcons.

Looking at 2 squads of Crisis Suits: 3 with TLMP/Flamer, 3 Fireknife. Points per Turn= 35 + 29 = 64 ppt. In just over 3 turns, average, you'll stop a Falcon. Is that too late? Maybe. But you'll be taking down that Falcon in a points-effective manner and not just pouring shots down a hole!

For giggles, say you DSed a Stealth Squad of 6 behind a Falcon: 1/3 = 70 ppt, which is pretty good at 2.33 ROI!

If you're able to DS the models BEHIND a Falcon:
RFPR+MP = (1/2 + 2/3) * 1/9 = 7/54, 27.22ppt;
RFPR+FB = (1/2 + 5/12) * 1/9 = 11/108, 21.388ppt;
TLMP = 1/9, 23.333 ppt;
You can see that each of these in a unit of 3 will get you about 1/3 of a dead falcon, which is nice on ROI if your units don't die after trying it. To make the DS option work, you really need to dump enough fire into the right place on a Falcon to make sure that you stop it on THAT TURN. Oh, and that you have the hatch blocked, because otherwise the guys inside will stomp on your units.

Which brings me to the final point: if you want to be realistic about game impact, I'd say that you might want to calculate the ROI over only 3 turns (because after that, the Falcons are wherever they wanted to be anyway) and also include the points cost of the transported unit (162 pt Harlie unit or 96pt Dragons, generally) in the ROI calculation - since if you can down the Falcon before it gets in position, the unit inside is easy meat.

Does any of this lessen how tough Falcons are, or how absurdly cost-effective they are (or rather, how absurdly cost-ineffective they make your shooting)? No. It does give us a nice way by OUR unit cost to see what guns are Good Buys versus Falcons, and which guns should be targeted elsewhere if maximizing our per-turn ROI is a viable option. Of course, sometimes you just have to pour every available gun into that one Falcon, regardless of ROI...but I'd rather not. And if I know ahead of time what guns "work", and what doesn't, I can maneuvre to put myself in a good position relative to those ideas.

========

So, those of you with other armies: what is your best ROI for taking down falcons? Is it a DakkaFex (at .82%)? Or a TankHunter Autocannon Havoc squad (if these still exist)?

And, do you see usefulness in this analysis, or am I off in the weeds here?

-JTS

**Reecius**

Thanks a ton for the analysis, but the only problem is that three turns is FAR more then enough to get where it wants to go. on a standard board, you will be lucky to get 2 turns of shooting.

Supid falcons.

**Longshot**

Looks pretty good to me. Obliterators look pretty good assuming they live.

75pts for a Lascannon, which gets you 63pts for a simplified ROI of .84

10 man terminator squad is horribly pathetic - like .2

Blastmaster squad is going to be ridiculously lame.

-----------------------------------

For tyranids-
Gunfex - 167
VC - 1/2 * 5/6 * 2 * 1/9 = 10/108 * 210 = 19.4
BS - 1/2 * 1/2 * 1 * 1/9 = 1/36 * 210 = 5.83
tot: 25.23 * 6 = 151.38
ROI: .90

Dakkafex - 113
(@ 18"?) - you really need to halve this since it won't be in range that much.
Devourer - 3/4 * 8 * 1/6 * 1/9 = 1/9 * 210 = 23.33
tot: 23.33 * 6 = 140
ROI: 1.23

VC/TLDev Tyrant - don't remember the cost, but I'll rough it out based on a 150pt price tag
VC - 2/3 * 3 * 1/2 * 1/9 = 1/9 * 210 = 23.33
tot: 23.33 * 6 = 140
ROI: .93

Hashes out something I knew from experience. Dakka nids have damn good odds against Falcons, even better if they don't have Harlequins/Fire Dragons.

**Longshot**

I actually like the simplified 6 turn analysis, but more accurate would be like JTS says - do the cargo + the falcon, vs odds over the 2 turns you get before they're up your buttocks.

I think also a single turn shooting analysis would be a valid tool for helping you understand how useful your unit is shooting at a falcon. Can just divide our numbers by 6

**Hellfury**

Ahhh....its always nice to see a JTS analysis. Its been quite awhile.

Things like this are one of the reasons that dakka has been viewed as the mathhammer/powergamers paradise.

Whatever. Its nice to see an easy breakdown and JTS always delivers.

its very timely I might add. Anything to help people defeat the ubiquitous falcon is a good thing. Knowledge is power.

Thanks JTS, this is a very helpful article and it is most appreciated.

**lambadomy**

I think this is useful analysis. Especially breaking down the one turn effectiveness to judge things like "what are the odds I can stop him from getting where I don't want him? Is it worth it to shoot at that falcon again now that it has dropped its cargo?" etc. It's also nice to see what that already exists in someone's army is the best to use on any given target. But it's definitely not enough to, say, get me to put a bunch of TLMP/flamer. I did notice, however, that a 53 point TLMP + TA crisis gets a ROI of 1.565, even better than the TLMP+Flamer:

TLMP+TA: 8/9 * 1/3 * 2 * 1/9 = 16/243... *210 = 13.827 ppt

They're 53 points each...so thats 82.96 points for 6 turns 82.96/53 = 1.565

This also goes to show the targeting array is probably worth the points on those suits in general, if you don't need the flamer.

it's also nice to see that this means that you only need 16 crisis suits loaded out this way to take out one falcon a turn. Hopefully. Oh wait, thats not nice, thats ridiculous. But only one more than you can legally field...I guess I'll load out a farsight list like this and go falcon hunting.

**onlainari**

Er, for Tau, broadsides are the best unit for shooting at falcons.

**lambadomy**

broadsides? 70 points stripped down...

3/4 * 5/6 * 1/9 = 15/216 =14.583

so 14.583 points a turn...or 87.5...87.5/70 = 1.25 ROI...worse than any twin linked missle pod crisis.

Of course, it's still pretty good, and you don't have to get within 36" to get a shot off, which is an issue with the crisis suits. But mathematically it's not actually the best, and broadsides are immobile enough (especially this one with no ASS) that falcons will have more ability to avoid their shots

**Lowinor**

I don't have much to add to this, other than a (somewhat belated) sentiment:

Welcome back, JTS.

**eaj731**

Nice work, but I'd like to add one more detail to your Hammerhead result. In the 180 point cost some points are invested into the secondary system (SMS+TL) and also part of the tank value. So I'd say that about 50 points of the hammerhead are not invested into taking out the falcon and so you should be calculating ROI based on 130 points and not 180 points.

This gives a HHRG: 0,60 ROI (vs. your 0,43 ROI)

Also I'm missing the Ion Cannon and Broadsides from your list:

HHIC: 2/3 * 1/3 * 3 * 1/9 = 2/27 ... * 210 = 15,55 ppt
BSRG: 3/4 * 5/6 * 1/9 = 5/72 ... * 210 = 14,58 ppt

As the broadside would never go to battle without at least a drone, thats 85 points per Broadside, or if taking the advanced stabilisation system, then 100 points per broadside.
BS(1SD) = 1,03 ROI
BS(ASS+1SD) = 0,88 ROI

And for the Ion cannon the value of the tank would be 95 points invested in taking out the falcon.
HHIC = 0,98 ROI (vs. 0,64 ROI, if counting the full value of the HH)

But that's just how I see it!

Espen A. J.

**onlainari**

Eh, I should be more specific.

Broadsides are the best unit for shooting at falcons.

Hmmm that's what I said last time.

Ok well what didn't I say?

Broadsides are the best unit for killing falcons.

You're correct that deathrains are better at killing falcons.

**JTS1486**

lambadomy, how do you strip a Broadside to that few points? There is a minimum when you MUST purchase extra gear. Better to calculate from a given weapon kit, unless you're just proving that even at NO extra cost it's not the best choice.

onlainari, unless you get more specific with what your position is differentiating "shooting at" from "killing", I can't tell if you're confused, or just trolling.

eaj731, I'd say don't take out the points you're spending on the remainder of any unit (vehicular or otherwise) unless you have a stated task for the rest of that unit to be accomplishing - that it can accomplish - while the single model or gun is shooting at Falcons. When you pay points for the SMS to shoot at another target, I'd suggest that the cost is an added burden to the Railgun - not to the SMS. The SMS has a value regardless, and it's about 40 points worth (4 FCW). But hey, I don't buy TL on my HH, so maybe there's something I'm missing.

Thanks for all the welcome back sentiments, fellas!

-JTS

**lambadomy**

JTS:

You're right - the minimum cost for a broadside is 73 points (blacksun filter) and in game terms they normally cost 80-95 points (counting their shield drone). I picked 70 only to show that even at the base cost with no extras it still wasn't as good as a deathrain in terms of ROI. Which of course was pointless as I made the mistake of assuming onlainari was on topic.

Onlainari: even in terms of "shooting at" instead of "killing", deathrain are still probably better, depending on the terrain. They have a better chance of being able to be placed/maneuvered to get a shot off on turn1, and after that the range advantage for the broadside becomes less important (since the falcons are almost definitely moving forward). But of course you'd also need to take into account the rest of your and your opponents army composition, likely terrain, likely placement...and still it's just informed opinion.

**Moz**

Best numbers I got were Tank hunting Havocs, 8 wound 4 autocannon:
62.22ppt 1.72 ROI

Good thing they got rid of those!

**puree**

Hypothetical scenario (i.e. exaggerated for clarity).

I could buy a unit of 10 Foo for 200 pts, with a mean average kills per turn of 1.2
or I could buy a single Bar for 200 pts with a mean average kills per turn of 0.9

Which would you want, The suggested ROI calculation seems to indicate that Foo is quite a bit more efficient. Personally I'd go Bar, why? cos 10 Foo at a 12% chance each still only rakes up a 72% chance of nailing the falcon but the 1 shot Bar has a 90% chance. In a situation where you have just 1 or 2 turns in any game to stop the the target doing what it wants to do I'll take high probaility per turn over a longer term average, especially where the average used is somewhat flawed. If I was going to use ROI figures then I would use the probability rather than the mean averages, as it better builds in the premium for high probability single shot weapons over lower probabilty multi-shot weapons. The chance of 2 kills in a single volloy is largely irrelevant (tank can only be destroyed once) but is part of the mean. Of course mean average is very easy to calculate, probabilities are more involved.

i don't have the later tau dex, so I can't readily provide any figures for tau comparisons based on probabilty - beyond the 53 pt crisis noted above and the bare broadside - the crisis still comes up out on top in that case.

If you are interested in the chance of cheaply stopping the falcon in just a turn or 2 before it unloads some passnegers, then 1 turn figures are what you want, Once you get into the realms of 6 turn calculations then really you are looking at the return in VP terms, thats really all that matters by turn 6.

So, we could refine the maths further. Instead we could calculate the return based on VPs gained, ie. use 210 for the kill, and 105 for the immobilise. so we end up with a ROI in terms of VPs gained to cost.

Deathrain over 6 turns:
((210 * .133) + (105 * .198) )/53 = 0.919

Basic broadside
((210 * .179) + (105 * .172) )/70 = 0.795

There are better broadsides, but without costs I can't say how efficient they are. The deathrain isn't bad, at 0.9 it is gaining almost as many VPs as you paid for it, assuming the falcon isn't doing anything unsporting like shooting back or hiding out of LOS

As to one of my favourite units:

Fex +bs + TLVC
((210 * .205) + (105 * .382) )/163 = 0.51 (ironically one of the best units in absolute terms, but you really pay for it)

for comparison

Fex +bs + TL/BS
((210 * .213) + (105 * .315) )/148 = 0.52

dakkafex (not likely to get 6 turns, but anyway)
((210 * .163) + (105 * .331) )/113 = 0.61 (a more reasonable 5 turns would drop it to .52 as well)

mmm, I need to make my preferred TLVC look better - time to account for the higher chance of non-scoring methinks, cos as they say, stats can prove anything

**winterman**

Welcome back JTS

TLMP: 3/4 * 1/3 * 2 * 1/9 = 1/18 ... *210 = 11.66 ppt

What about TLMP with targeting array? I know it goes against the more shots maxim but I think it has potential for falcon killing
BS4 TLMP: 8/9 * 1/3 * 2 * 1/9 = 16/243 ... *210 = 13.83 ppt

So new deathrain (TLMP, TA) is 53 points IIRC. Gives you an ROI of 1.57 when shooting all game.

**onlainari**

Ok let me explain.

Deathrains aren't 53 points they're 106 pts for 4 36" shots.

Broadsides aren't 80pts or whatever they're 195pts for 2 72" shots.

Broadsides are heavy support and deathrains are elites.

If you see where I'm going already then that's great.

What I'm saying is that in terms of the whole list, I would rather be taking a unit of broadsides over a hammerhead than a unit of deathrains over a unit of fireknives.

In other words, we need to take falcons into account when building our lists. Should we:

a) take a unit of deathrains, or
b) take a unit of broadsides

The answer is b).

Personally I take both and I still prefer shooting at falcons with my broadsides than my deathrains.

And I would drop the deathrains for fireknives if there was no tournament composition scoring, but would keep the broadsides specifically for killing falcons.

Of course, broadsides are 89pts more expensive. I could get at least another deathrain into the unit for that. But obviously I would not want to do that, 3 deathrains in a unit is a bad idea. And 2 units of 2 deathrains is also generally not a good idea. So we're stuck with a deathrain unit at 106pts and elite, and a broadside unit stuck at 195pts and heavy support.

In this case the actual points it takes for a kill is not the best way to look at it.

Deathrain: P(destroy) = 1 - [ 1 - (8/243)]^4 = 0.125 or just over 1/8. Very, very nice.

Broadsides: P(destroy) = 1 - [ 1 - (5/72)]^2 = 0.134 or 2/15, also very nice.

But we're not looking at the whole picture.

Deathrain: P(nothing) = [ 1 - (8/27)]^4 = 0.245 or almost 1/4. This is bad. This is a huge risk of losing the crisis suits to falcon shooting, thus less shots at it next turn.

Broadsides: P(nothing) = [ 1 - (5/8)]^2 = 0.141 or 1/7, now this is a lot more comforting.

**JTS1486**

winterman: Holy Cow! Someone still remembers at least one of my 4 "Rules of Tau"! I am so psyched!
As for Targeting Array, it's a totally valid choice depending on army composition. My current army lists all include 3 Sniper Teams, which means I'm netting out 2 Markerlight hits per turn on my chosen unit. I can use those to boost unit BS, which I calculate separately in a table that charts kills gained from BS boost based on unit boosted and unit targeted, if you follow my drift. Because of that, choosing Targeting Arrays really isn't on my list of stuff to take, because I'm getting the same effect for "free" from the Sniper units. (It's interesting to do the math on ROI of Sniper Units to get a "proper" points cost for them and see just how LITTLE you pay extra to get the Markerlight hits.) That said, you could totally do the math with Targeting Arrays to find the right cost vs effectiveness ROI point.

puree: you're right, of course. I look at things over the course of the game because I got stuck in a GT matchup 4 years ago against what might have been the worlds first "Zilla" list: 36 wounds of TMC. (Three Fexes, Two Tyrants, 3 Guards.) My Tau configuration at the time just didn't have enough of the right shots, mathematically, over the course of the entire game, to actually wipe out the forces facing me. Since then, I find it to be a good litmus test to determining if I've already dug myself into a hole it's impossible to climb out of tactically, so to speak.

Your point about looking at 1 turn kills is definitely an important one, and part of the reason why 6-turn ROI can be very misleading, especially as it concerns SunFire Crisis Teams. 1 turn, high probablity = good. Long term ROI = good. Chance of actually achieving that long term ROI due to shooting every turn: abysmally bad. In other words, it doesn't matter if your FCW squad is going to earn its points plinking 1 MEQ per turn all game if it has to shoot at a Tac squad that just tumbled out of a Drop Pod. In that case, killing ALL the nearby enemy is more important than maximizing your ROI in points for that turns shooting.

Of course, a unit with high optimal ROI is more forgiving of spending a turn or two firing at non-optimal ROI targets...

-JT

**onlainari**

Ok here's an analogy. Say you play magic the gathering and take 4 korlash, 2 tomb stalker, 2 nihilith, and 2 skeletal vampire as your creatures. The deck won't work.

Lets look at each creature individually:

Korlash - awesome creature, the fact that decks are built around him is a testament to how powerful he is.

Tombstalker - one of the best big black creatures in the game, a massive creature with flying with no drawback.

Nihilith - a great medium sized creature that fits very well into black decks.

Skeletal Vampire - nothing says "win condition" like skeletal vampire. Get him out and close the game.

Ok so a deck with all these should do well, right?

The answer is no, and it's simply because the pieces of the puzzle don't fit together. It's like taking a commander and 90 marines in a 1500pt list. Each marine is great for his points. But why will this army lose? Because it can't handle vehicles obviously, will have trouble with any unit that's different from the norm. Most armies can work around this army and pick the marines off and come out on top simply because they lost less casualties.

When you're looking at killing falcons with Tau, you have to make the pieces of the puzzle fit. Deathrains are a great unit to use just because mech eldar is going to be a problem, but broadsides are an even better unit to use, even though they cost 89pts more.

For space marines, nothing screams at me more than "take tank hunters on your terminator assault cannons". It costs, what, 15pts? You double your glances against a falcon and you're still effective against genestealers etc. The 15pts might make them in theory "less effective" against genestealers but in terms of the whole list it doesn't change anything (you can worry less about falcons with the rest of the list just because you have 2 units of 2 assault cannons with tank hunters).

For the new chaos space marines (and Imperial Guard), lascannons are statistically better than autocannons (it is a common misconception that they have the same chance!).

Lascannon: P(killing) = 0.0494

Autocannon: P(killing) = 0.0487

Lascannon: P(nothing) = 0.556

Autocannon: P(nothing) = 0.605

**Centurian99**

JTS - welcome back!

**lambadomy**

onlainari:

I completely agree that the survivability of the unit needs to be taken into account, as well as "what are my other options for that force org slot". I typically take fireknives, not deathrain, and this math isn't going to change that. Of course, I also typically take a couple hammerheads instead of broadsides, though I've been playing mostly farsight recently and end up with one of each.

Yes, broadsides are more survivable against falcon shooting - but even if you can only hide the crisis suits with JSJ half the time, they end up pretty comparable.

I'd say Broadsides are more likely to be caught by the falcons and killed by the falcon's passengers.

I will admit that I have had nothing but bad luck fielding broadsides against any kind of a mobile army. I'm not sure what I'm doing wrong, but they don't tend to stay out of hand to hand past turn 3. Of course, sometimes they win that hand to hand (things without power weapons/rending take forever to kill broadside teams) but they still are tied up.

**onlainari**

Broadsides are impossible to charge, unless you take a farsight list in which case you can't take the 3 units of kroot to protect them (and in that case you've got to use fire warrior teams to try to do the same thing, and take 85pt devilfish for the free gun drones).

So I'll rephrase that, I've never had my broadsides charged.

A dark angels opponent once set up his bikes (fast attack, last unit deployed) opposite my broadsides that were pushing deployment back. Easy first turn charge. Until 2 units of kroot infiltrated in the way. He didn't bother and moved off the the right instead of down the board.

Posted By JTS1486 on 08/16/2007 11:54 AM

I think we have a winner - and it's something every Imperial Guard player already knew. Use Autocannons (I mean, Missile Pods) to take down Falcons.

JTS you must have done a lot of really good maths or an article in the past because everyone is happy that's you're back. But I still don't see the evidence that autocannons are better than lascannons for imperial guard against falcons.

My post above is pretty strong evidence in favour of lascannons.

**puree**

Posted By onlainari on 08/18/2007 8:31 PM

Posted By JTS1486 on 08/16/2007 11:54 AM

I think we have a winner - and it's something every Imperial Guard player already knew. Use Autocannons (I mean, Missile Pods) to take down Falcons.

JTS you must have done a lot of really good maths or an article in the past because everyone is happy that's you're back. But I still don't see the evidence that autocannons are better than lascannons for imperial guard against falcons.

My post above is pretty strong evidence in favour of lascannons.

I'm a little confused by your figures, were you basing them chaos or IG. For IG i get the following (for 1 weapon in 1 turn)-

lascannon
p(kill) = 0.0093
p(kill or imm) = .037
p(zip) = .66

autocannon
p(kill) = 0.0092
p(kill or imm) = .036
p(zip) = .69

so in absolute terms there is a tiny advantage to lascannon, but the thread in general was looking at how much you spend for that. Weapon costs differ per unit, but for say a sentinal, 50pts for AC, or 55pts for LC, the 'ROI' using the above figures would be

lascannon
kill = (210 * .0093)/55 = .036
kill or imm (210* .037)/55 = .141
wasted points per turn = .66 * 55 = 36.3

autocannon
kill = (210 * .0092)/50 = .039
kill or imm (210* .036)/50 = .151
wasted points per turn = .69 * 50 = 34.5

It comes down to how you want to define best, yes the lascannon is slightly better in absolute terms, but you could consider yourself to be paying 'over the odds' for that, and although you miss less often, on average you are wasting more points on misses.

That of course is also specifc to falcons, against say a hammerhead with decoys

lascannon
p(kill) = 0.0556
p(zip) = .75

autocannon
p(kill) = 0.0367
p(zip) = .84

ROI
lascannon
kill = (210 * .0556)/55 = .212
wasted points per turn = .75 * 55 = 41.25

autocannon
kill = (210 * .0367)/50 = .154
wasted points per turn = .84 * 50 = 42

So lascannon is quite a bit better at points/kill, and ever so slightly cheaper on wasted points/miss.

Against predators there just isn't a contest, the lascannon is simply in a better league altogether.

**PenguinDude**

Now someone do Orks!

Kidding, we cant kill falcons.
But, I do love reading this thread, since Eldar and Tau are two popular armies around here.

**puree**

Posted By PenguinDude on 08/19/2007 10:42 AM
Now someone do Orks!

ROI = 0
Wasted points/shots = ∞

**Phryxis**

Just to give everyone an idea of MY level of Tau tactics:

While I was reading this thread I noticed people mentioning Targetting Arrays, and I thought back to how much missing I did the last time I played my Tau, and I figured I should add some.

So I go look at my list, and it turns out I gave Targetting Arrays to pretty much all the suits in my list. All told I managed to completely forget about 80 points worth of upgrades right there, and probably would have won my last game, or at least done a lot better, if I'd remembered paying for them.

**JTS1486**

onlainari-

I have to agree with Puree - I'm more than a little confused at where your lascannon and autocannon numbers are coming from. I need you to explain your formulas and assumptions in more detail, or understand that I cannot accept any of your results as valid.

In order for your math to work, you must be defining P(destroy) as "probability to immobilise or destroy a Falcon in a single turn", which is okay. The Broadside Railgun number is just fine. The Deathrain number you got makes less sense to me: TLMP is 3/4*1/3*1/9 to Immo or Destroy, which is 1/36 per shot - not 8/243. (2^3 / 3^5 ??? Where did this come from?) If you look at the P(no immo or destroy, single shot) as 35/36, and assume that you've got 4 shots (from two suits), your P(no immo or destroy, 4 shots) is (35/36)^4 = 89.4%, or 10.6% chance of "stopping" the Falcon.

10.6% chance of Stopping a Falcon for 106 points (odd coincidence) is, in my book, a LOT better buy than 13.4% chance for 195 points. THAT is one purpose of ROI analysis. Note that this number gets even more skewed when looking at Probability on a more closely matched point-for-point level. If you can buy 4 TLMP suits for about the same cost as 2 Broadsides (which you can), then the math for comparing effectiveness apples-to-apples is 1-(35/36)^8 to Stop the Falcon: 20.2% versus 13.4%.

If you define Broadsides: P(nothing) as "the probability of 2 Broadsides not achieving a Glancing Hit in a single turn", then the number is (3/8)^2, not 1-(3/8)^2. THAT number is 14.1%, alright. Why care about that, though? It's not the metric I need. Most glancing hits result in no effect, and I need to be looking at results that materially impact game play. Against Broadsides, the Falcon just has to get close enough to drop cargo.

Again, your P(nothing) for Deathrain looks pretty flawed from where I stand. I understand you not wanting to get the suits shot at by the Falcon they are targeting, but the math is off. A glance is struck 1/4 of the time on every shot, and every glance will prevent them from being shot at (or shot at as much). So, P(nothing) is (3/4)^4, or a 31.6% chance that the Falcon can go hunt the unit of 2 TLMP Crisis suits with impunity next turn. That's alarming, but not unexpected. At 8 shots (the 4 suit apples-to-apples) it's only 10% chance that the Falcon made it though the turn still able to shoot.

===================

I understand that you are making the assumption that your only choices happen within Force Org Slots - you are balancing Hammerheads versus Broadsides (and Fireknife versus Deathrain) over a wide variety of targets.

I am planning for Falcons here and now, and working on the rest downstream.

I can accept that, in **your** army composition, Broadsides are what you shoot at Falcons. It may be the best unit in your army (on an ROI basis) to shoot at Falcons, in which case it is ABSOLUTELY CORRECT that you use them in that manner.

What **I** am looking to get at is whether or not Broadsides are the best thing for me to buy in an Army List for shooting at Falcons. The answer would appear to be a resounding "No".

Is 2 units of 2 Deathrains a good idea? Not in my army, as I need AP2. I'm going to be hybridizing, with a unit of Deathrain-like suits and a separate unit of Fireknife-like suits - and that hybrid WILL end up being more cost-effective than slapping down two Broadsides against EVERY target class except AV13+.

I look at balanced army composition in terms of finding the best unit for addressing the armies of metagame opponents I might face. If I can construct an army that is flexible out of a variety of units, understand the optimal configurations across target classes and expected results, and gameplan reaction models, then I will be prepared to make good tactical decisions (movement and fire planning) in-game.

===================

Is there a role for Broadsides in an army List? I'd say yes, depending on how they're deployed. They still make great bait. Nice, juicy unit that can put out good Anti-Tank shots, use Field of Fire to channel enemy movements, draw in mobile Close Combat or Deep Strikers, etc. I'm just saying they aren't the "best unit Tau can shoot at Falcons", because the math indicates that they aren't.

-JTS

**winterman**

JTS, FYI deathrain is now considered to be a TLMP and TA suit for 53 points. 8/243 is the chance of one BS4 TLMP shot immob or killing a falcon. 8/9 * 1/3 * 1/9 = 8/243. Looks like his calcs were for 2 suit teams.

Is 2 units of 2 Deathrains a good idea? Not in my army, as I need AP2. I'm going to be hybridizing, with a unit of Deathrain-like suits and a separate unit of Fireknife-like suits - and that hybrid WILL end up being more cost-effective than slapping down two Broadsides against EVERY target class except AV13+.

Torgoch posted a hybridish build awhile back for a suit heavy 1500 point list that may be of use: 3 suit team, 2x deathrain, 1x leader w/ PG/MP/TA and HWMT. Was what I was looking at fielding in my (now on hold) tau army.

PS I remember much of your old tacticas. Infact, I had to join dakka after reading your infamous batrep against mauleed way back when. Also, funny you mentioned snipers. They've always seemed to me to be a perfect addition to the ole Nikken Ryu (sp) style of list that you championed.

**JTS1486**

Ah, okay - thanks for catching me up on the terminology, winterman. Now I see the Math.

I really do like Sniper Teams right now - they seem to be highly annoying (hard to kill) cost-effective units with force multipliers built into them. They're the same kind of bait I used to use Broadsides for in Nikkenryu.

Torgoch was good with Math for optimal suit team builds, I'm not surprised to hear he's still putting out good work. I take it he's moved on from Dakka too?

Just a quick bit of math to satisfy my curiosity:
TLMP/Flamer = 47pts
TLPMP/TA = 53 pts
So 6 points to get the +1BS, giving up just a bit of flexibility that would only get used in **rare** situations. Is the 6 pts worth it?
MEq: 5/12 vs 40/81 18.5% effectiveness boost for 12.7% cost increase. Same vs AV12. Okay, I like it.

All of which makes TLMP/TA even tastier vs Falcons. 12.5% chance of immobilizing or destroying the falcon with 2 suits is even more worth it than the single Broadside those points could buy.

============

I'm curious about the viability of Nikkenryu in v4. Mech Tau is a LOT more viable than it was in v3 thanks to the RF rules, and the ridiculous cost-effectiveness of Broadsides has dropped off significantly now that it seems you have to kit them with Shield Drones and lose the MTs (which hurts the old-style Nikkenryu list that banked on them). I think if it works now, it'll still have a few static elements that are hard to kill and cost-effective across the board, with a mix of units that are effective against the certain target classes mixed-and-matched.

-JTS

**onlainari**

I use 2 deathrains in my list, as well as 2 broadsides.

I'm not a big fan of hits or casualties per point because I don't think it's a good idea. Deciding what is the most effective unit at killing marines will not win you games. Working out the firepower needed to kill all 40 marines your opponent fields is useless. Selecting units only on their ability to make their points back is a poor decision. You only need to lose less than the opponent and score more points on objectives.

I believe you need to take certain units into account when building your list.

These units are:

Space Marines
Falcons
TMCs
Guardsmen

In my experience these are the units that you need to be able take care of.

Land raiders, genestealers, and terminators are often added to this list but are not needed. Your list's ability to deal with these things should be a side effect, these units do not tend to dominate games where they are unkillable (especially the land raider).

You need to work out the probability distribution of your units when you shoot at those 4 units. For each of those 4 units, you need at least one unit that can handle the job.

We've done some probabilities on both deathrains and broadsides. Which unit will you use to take care of falcons? Maybe you will lose falcon killing for increased marine, TMC, and guardsmen killing fireknives and hammerheads give us. Maybe you will decide to pick one of these two units to help you out against falcons. Maybe it will be both.

I say go with broadsides first, deathrains second.

By the way, broadsides are even more expensive than deathrains than what their points make them out to be. They cost an extra 15-30pts over a hammerhead, while deathrains cost 33pts less than an FK4/FK3 team. Even then, you cannot forget to take kroot into account. Broadsides do not function properly without speed bumps. Kroot are great speed bumps, but they still cost 70pts a unit. Two units of kroot dedicated just to protect the broadsides, that's 335pts for 2 S10 shots.

But as I say, the efficiency doesn't matter. These broadsides are guaranteed to shoot for 6 turns without fear of assault. Most importantly, that's 195pts your opponent cannot get.

The deathrains also have a good chance of shooting 6 turns which also makes them a very good unit.

The other factor is gravity. Broadsides create gravity, which most armies have to deal with, but certain mech Tau lists have been able to lose it. A unit approaching the broadsides gets put right on top of the target priority list, and will pull your firepower from the rest of the list. Deathrains have no gravity, they are able to operate on their own. This is another advantage of deathrains.

In the overall feel of the list, I have chosen to say I prefer broadsides shooting at falcons, even though they cost more in points, and even more in protection and gravity, simply because I price risk very high. Broadsides are less risky and that can be mathematically proved.