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.

My friend is a bit of a math geek, he likes to mathhammer his armies stats to the nth degree. I am math challenged and consider spur of the moment tactics more important for winning games than rationalizing dice probability. I mean, they're dice, their job is to mess up your plans. They are supposed to convey the chaotic elements of a real battle.

I'd say overall tactics. Probability calculations are great, but in the end they're really just for helping you gauge what you can expect out of a model/unit.

Probability calculations are used to inform your tactical decisions. It's not one or the other, you need to know both.

That question is what you call a false dichotomy. Tactics are intrinsically tied with probability calculation. Whether you do them mentally or formally, you cannot escape probability without abandoning tactics as well.

It is certainly possible to learn probabilities slowly through experience and base you decisions simply on past experience, though it is important to realize that your brain is very BAD at certain kinds of estimation, usually the kind you find people calculating, like chances of damaging a certain thing with a certain weapon. We are very bad at rapid calculation of individual variables, formal methods and computers are a way for us to make up for this.

Conversely, you could try to make all of your decisions by calculating it out, but you will very quickly hit the limits of formal methodology. In algorithmic theory we would say that true decision making is an NP-Hard, a type of algorithm that the most powerful super computers cannot handle (at least for anything beyond the trivial). Interestingly, your brain can LEARN to be very GOOD at solving these types of problems. Instead of calculating it adapts through trial and error and experience, thus side-stepping the problem of solving it algorithmically.

Every advanced military over the past hundred years has known this. The militaries of the world spend billions on mathematicians and computer scientists to calculate out the probabilities of specific problems, those which either our brains are bad at solving or for which we can't afford the luxury of experience.

They also spend money on tacticians and strategists, and spend years exposing them to historical battles and tactics and having them practice to try to rapidly advance their brains adaption process.

Do you really think you don't use probability? Do you just run one unit at another because its closer? Heck, even that would be probability, because you think that attacking the nearest target is the fastest/best tactic. Even bothering to shoot is an exercise in probability because you estimate that is the best way to win. You may not calculate them formally, but you do calculate them. The problem is your brain is prone to error when calculating certain kinds of probability without using a formal method like statistics.

Conversely, anyone who says they can win a battle strictly through mathematics has a poor understanding of the subject. Games like chess with fairly simple rules and lots of constants strain the abilities of mathematics. A game like 40k is literally incomputable with anything short of the theoretical quantum computer.

What you are really asking is, is it better to learn through trial and error and avoid formal methods, or can trial and error be done away with in favor of formal methods. If you wish to force a decision between those two extremes than I suppose the answer is you can win by pure trial and error, you cannot win by the pure use of formal methods; so congratulations, you win.

But it is a false dichotomy, and a foolish stance to take. Probability is intrinsically tied to tactics, and formal methods are a way for us to enhance that subset of tactical problems that our brains are not well equipped to handle. They shorten the learning curve. If this is a real war I would say you are a fool to do away with formal methods and learn probabilities through trial and error. All other things being equal, including tactical acumen, the commander with a statistically optimized force will defeat the one who relied on trial and error, but statistics and probability will never be enough to truly make decisions for a commander.

Then again, this is a game, not real war. The point is to have fun. If formal methods are not fun to you, if you would prefer to learn all probabilities through trial and error, if that is more fun, do it, no lives are on the line. If you are gifted in the realm of tactics than you will likely defeat those who use formal methods and are not as gifted/experienced at tactics as you. But were you to ever play an alternate version of yourself who did use formal methods, you would likely lose.

Formal methods are simply a tool for helping us in areas that we are not naturally good at, and to speed up the learning process to improve our tactics.

As Kadun said, but if you look overall it should be math that makes your decisions.

What I mean by that is this:
If you have the ability to choose to shoot an enemy squad (we'll say tactical marines in the open) with plasma guns or bolters, you will choose plasma guns (assuming no other better targets and that both squads are in equal range of target) since the plasma guns ignore the marines' armor save. If you break it down, you chose the plasma because statistically it will kill more marines than bolters.

Though making your decisions during a battle, moving what where when etc is the most important, it all is based on the mathhammer (you're not going to send a single melta gun against a land raider expecting to kill it are you? Certainly 4 melta guns will have a better chance of killing it).

And no a die's job is not to mess up your plans. A die's job is to involve probability. If you shoot 10 rapid firing bolters at a squad of marines 8 million times and average out the results, you will see that you killed a little over 2 marines. And you can always expect that. You may do better or worse, but you can plan on killing 2 marines with 20 bolter shots.


Edit: what Riplikash said too.

kadun wrote:Probability calculations are used to inform your tactical decisions. It's not one or the other, you need to know both.

QFT!!!

I just wish I didn't suck at math.

Thank feth for Math Hammer on the Iphone

Math calculations don’t mean anything by themselves – the straight probability of scoring a kill on two different targets means nothing if you can’t assess the priority of killing each target. Sure you have more chance of killing the rhino than the land raider, but how dangerous is the land raider, possibly carrying elite assault troops, and how dangerous is the rhino, which may able to drop some troops onto an objective next turn? It’s a value judgement to be made each turn.

Then again, if you aren’t calculating probability then what point is there in determining priorities? If you have no idea how likely a missile launcher is of taking out the rhino or the land raider, then how could you possibly make a decision. Like riplikash says, this understanding of probability can be learned from experience or even just intuited, it doesn’t have to be calculated as a percentage to two decimal places – but it needs to be understood at some level.

Statistics is a tool to drive strategic decisions. Of course outliers (results that fall wildly outside the statistical norm) can happen, and mess things up terribly, but that’s what contingency plans are for.

In addition to 40k I play a lot of poker. The worst possible starting hand is 2 and 7 off suit and when you see that hand you should almost always fold. When you fold you do so with the understanding that 7, 7, 2 (full house) could hit the flop. 3, 4, 5, 6 of the same suit of your 7 could come out (straight flush). Also, 7, 7, 7 could hit the flop (four of a kind). Those results are outliers – certainly within the realm of possibility, but not likely enough to warrant risking ANY chips on regardless of your table position or the read you have on your opponents.

Even though everyone tells me that they are both either intrinsicaly linked or that prob. calculations are more important, I still think that solid tactics will prevail,only because ive won 8 out of 10 games with the friend I mentioned earlier [not to brag ] and he has mathhammered everything right down to bathroom brakes. I beat him because he always charges blindly at me trusting to his calculations to ensure his armies success. Ive told him many this many times and he thinks math is still king.


Automatically Appended Next Post:

Anavrin wrote:I'd say overall tactics. Probability calculations are great, but in the end they're really just for helping you gauge what you can expect out of a model/unit.

YES! thats exactly what I thought , they're fun but not something to build your army around.


Automatically Appended Next Post:

Ail-Shan wrote:As Kadun said, but if you look overall it should be math that makes your decisions.

What I mean by that is this:
If you have the ability to choose to shoot an enemy squad (we'll say tactical marines in the open) with plasma guns or bolters, you will choose plasma guns (assuming no other better targets and that both squads are in equal range of target) since the plasma guns ignore the marines' armor save. If you break it down, you chose the plasma because statistically it will kill more marines than bolters.

I get what you're saying but its still essentially a question of tactics not probability,the same way you wouldn't try to kill a landraider with a bolter,thats not what they're designed for.

incarna wrote:Statistics is a tool to drive strategic decisions. Of course outliers (results that fall wildly outside the statistical norm) can happen, and mess things up terribly, but that’s what contingency plans are for.

In addition to 40k I play a lot of poker. The worst possible starting hand is 2 and 7 off suit and when you see that hand you should almost always fold. When you fold you do so with the understanding that 7, 7, 2 (full house) could hit the flop. 3, 4, 5, 6 of the same suit of your 7 could come out (straight flush). Also, 7, 7, 7 could hit the flop (four of a kind). Those results are outliers – certainly within the realm of possibility, but not likely enough to warrant risking ANY chips on regardless of your table position or the read you have on your opponents.

Just got say this happened to me at the casion once. Had 2-7 off in the big blind. No one raised, and flop was 7-7-2. No one say it coming. The key is when an outliers comes your way to take full advantage of it. Or when Flopping Quad aces and losing to a Straight flush that you limit the damage. (had that happen once as well.)

As mentioned earlier. Statistics inform tactical decisions. The premise of your position is incorrect. You’re suggesting that tactics are more important than probability. That’s like saying a car is more important than gasoline or a lamp is more important than a light bulb.

You’re assertion that statistics are inferior to tactics because you’ve beaten an opponent who uses statistics to guide his decisions is also incorrect. I think it’s pretty clear from your description of his tactics that he’s using statistics to guide his tactical decisions in an irrational way… essentially filling the car with gas but not turning it on.

In a vacuum, from a statistical perspective, is it better to shoot a 5-man squad of marines that’s 8 inches away from a 20-man squad of guardian defenders or is it better to fleet and charge?

Imagine you have 3 rhino’s filled with 10 tactical marines. There are 2 squads of 10 khorn berserkers coming toward you. You can either move the rhono’s so that all three squads can rapid fire one squad of berserkers and ignore the other or you can send 2 to shoot at one squad and 1 to shoot at another. Which is a more efficient use of resources?



Automatically Appended Next Post:

jbunny wrote:Or when Flopping Quad aces and losing to a Straight flush that you limit the damage. (had that happen once as well.)

That exact scenario happened on the very first day of the 2009 world series of poker. I can’t imagine sitting down at a WSOP table to catch quad aces only to get stomped by a royal flush. That’s when you know lady luck’s just a sadistic and cruel fiend of a woman.

kadun wrote:Probability calculations are used to inform your tactical decisions. It's not one or the other, you need to know both.

This.

@bigbadbob: Given the scenario described below, answer the questions that follow.

Your opponent has a Land Raider and a Rhino. If you destroy both vehicles this turn you win the game. You have a Meltagun within 6" of both vehicles and a Lascannon with clear line of sight and within 48" of both vehicles.

Both weapons are capable of penetrating and destroying either vehicle.

Which weapon do you shoot at what vehicle and, most importantly, why?

kadun wrote:@bigbadbob: Given the scenario described below, answer the questions that follow.

Your opponent has a Land Raider and a Rhino. If you destroy both vehicles this turn you win the game. You have a Meltagun within 6" of both vehicles and a Lascannon with clear line of sight and within 48" of both vehicles.

Both weapons are capable of penetrating and destroying either vehicle.

Which weapon do you shoot at what vehicle and, most importantly, why?

you would fire the meltagun at the rhino because it has weaker armor and is claose enough to give the melta an ap bonus.you would fire the lascannon at the raider because its more powerful than the melta and has a better chance of destroying it.



Automatically Appended Next Post:

incarna wrote:As mentioned earlier. Statistics inform tactical decisions. The premise of your position is incorrect. You’re suggesting that tactics are more important than probability. That’s like saying a car is more important than gasoline or a lamp is more important than a light bulb.

You’re assertion that statistics are inferior to tactics because you’ve beaten an opponent who uses statistics to guide his decisions is also incorrect. I think it’s pretty clear from your description of his tactics that he’s using statistics to guide his tactical decisions in an irrational way… essentially filling the car with gas but not turning it on.

In a vacuum, from a statistical perspective, is it better to shoot a 5-man squad of marines that’s 8 inches away from a 20-man squad of guardian defenders or is it better to fleet and charge?

Imagine you have 3 rhino’s filled with 10 tactical marines. There are 2 squads of 10 khorn berserkers coming toward you. You can either move the rhono’s so that all three squads can rapid fire one squad of berserkers and ignore the other or you can send 2 to shoot at one squad and 1 to shoot at another. Which is a more efficient use of resources?

Allright I guess all of you have got got me convinced. I should ask my friend to post something on this thread, since he's the reason it exists.

were you being sarcastic bigbadblob?

or do you not know what the melta rule does?

well you shouldnt really use a mathhammer, i mean: if you have a 4+ save you just know that you have 1/3th chance not to fail it dont you? if you just thing about how big you chance is AND using tactics you will win, only calculating however wont, because with what formula are you going to calculate where your opponent is going to place his boys? and how will your mathahmmer know or you will be able to see them with your fire warriors in 3 turns?

Your post makes no sense at all.

bigbadbob wrote:you would fire the meltagun at the rhino because it has weaker armor and is claose enough to give the melta an ap bonus.you would fire the lascannon at the raider because its more powerful than the melta and has a better chance of destroying it.

You REALLY need to work on your math skills, because that is the exact opposite of what you should do.

*Actually, rereading seems to show it's more a problem that you don't understand the relevant rules. Either way, both need work.

That example failed so incredibly hard. For what it's worth, I would have assaulted the Land Raider with a Frag grenade.

bigbadbob wrote:

kadun wrote:@bigbadbob: Given the scenario described below, answer the questions that follow.

Your opponent has a Land Raider and a Rhino. If you destroy both vehicles this turn you win the game. You have a Meltagun within 6" of both vehicles and a Lascannon with clear line of sight and within 48" of both vehicles.

Both weapons are capable of penetrating and destroying either vehicle.

Which weapon do you shoot at what vehicle and, most importantly, why?

you would fire the meltagun at the rhino because it has weaker armor and is claose enough to give the melta an ap bonus.you would fire the lascannon at the raider because its more powerful than the melta and has a better chance of destroying it.

And my original assumption proves correct. Registration date 3 days ago, 12 posts, question designed to polarize, answers deliberately phrased to illicit negative responses.

I'll stop feeding the troll now.

bigbadbob wrote:

kadun wrote:@bigbadbob: Given the scenario described below, answer the questions that follow.

Your opponent has a Land Raider and a Rhino. If you destroy both vehicles this turn you win the game. You have a Meltagun within 6" of both vehicles and a Lascannon with clear line of sight and within 48" of both vehicles.

Both weapons are capable of penetrating and destroying either vehicle.

Which weapon do you shoot at what vehicle and, most importantly, why?

you would fire the meltagun at the rhino because it has weaker armor and is claose enough to give the melta an ap bonus.you would fire the lascannon at the raider because its more powerful than the melta and has a better chance of destroying it.

Steelmage99 wrote:

kadun wrote:Probability calculations are used to inform your tactical decisions. It's not one or the other, you need to know both.

This.

More of this.

Also, troll.

ONE MORE THING! Ein be4 John Locke.

Tactics is more important then calculations due to the fact that a list full of cannon fodder would find victory in the hands of a smart enough player/vulnerable enough opponent.

Although any sort of smart player would use some form of calculations in their planning.