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.

So I am curious, how well does crunching the numbers on a units capabilities compare to playing it against an experienced opponent?

I ask because, while I am a very capable mathematician, I admit to not having any challenging opponents available to me. Most of the people in my group are newer than I am and aren't very good at tactics. And even when I play my friend that plays 40k semi-regularly, and is a major meta-gamer, the outcome is still heavily in my favor.

Experience is a less precise version of math. If you play enough games you know that 2 squad of marines with 2 plasma guns each probably won't kill 7 plaguemarines in cover . Math helps you get the same resoults without having to play those 100+ games to check stuff.

Crunching the numbers is (in my opinion) better than experience.
Math-hammer tells you what you should expect, and you can even calculate the chances of you having a bad day and all.

Let's say a person uses plasma and has rolled a lot of 1's in 5 games.
He than decides to drop plasma because "it is bad!".
As a mathematician you should know that 5 samples is nothing.

Makumba wrote:
Experience is a less precise version of math. If you play enough games you know that 2 squad of marines with 2 plasma guns each probably won't kill 7 plaguemarines in cover . Math helps you get the same resoults without having to play those 100+ games to check stuff.

This is somewhat accurate. Experience doesn't just mean <= math. Experience also means that you likely understand the flow of the game and the capabilities of many units. I'm willing to bet that the best players don't sit around and crunch numbers, they go out and play games and test out lists and units against varied opponents and skill levels.

At its essence the game is a dice game. The math always wins. If every move you make puts you at a mathematical advantage you will win more than you lose. Experience is what gives you the intuition to know the moves that put you at a mathematical advantage, even without doing the math.

It also helps to play a list that already has the best mathematical chance of winning.

The one drawback of relying purely on maths is that the chances are, in a competitive environment, your opponent will likely have a similar understanding of the maths as you do. Meaning that if you are aware of the capabilities of of a particular unit, the opponent is also likely to have an idea of its effectiveness, and thus will attempt to avoid/eliminate the unit in question. As such, this is where experience comes in. as you are likely more aware of how this unit works with the rest of your list, and how you maximise the effectiveness of the unit.

Maths can tell you what a unit is capable of, but experience is the only real way to know 'how' to use that capability. Relying purely on one or the other will lead to defeat.

And no amount of mathhammer will ever stop anyone from making those "Hail Mary!" moves. And sometimes, those moves will work. Example, my Rune Priest is locked with a Dreadknight. I have meltabombs, but I choose to use the Runic Weapon instead. He survives the Dreadknight by making his 5++ terminator save. Hits in combat, wounds on a 6, Dreadknight player rolls a 1 on his armor save. Force Weapon goes off, DK's gone.

If that were to happen again, I would to the same thing again.

My experience is mathhammer doesn't work, the real world will always throw a wrench in the closed environment of mathhammer, it's a good indicator, but not something to be.relied on

The concerns I have with mathhammer is that it can tell you what will happen in certain circumstances, when you compare units shooting/assaulting units. However, most times, LOS, positioning and cover specifics aren't really dealt with in a game realistic way. There are too many possible combinations.

The other thing is that you'll only get "average" results as shown by mathhammer over hundreds of games.

"So I have caused 3 wounds on your terminators - you get 2+ saves"
"three ones....well, that's bad luck"

"So I'm shooting overwatch with my pistol, 6 to hit, 6 to wound and you'll get a 2+ armour save. Oh, you died? Pity, your charge failed".

I'm for experience.

Loborocket wrote:
Experience is what gives you the intuition to know the moves that put you at a mathematical advantage, even without doing the math.

In a nutshell, this. Math helps you understand what units/weapons are effective against which targets, but you need experience to know how to put them in the right position to be used efficiently. There's also no calculator in the world to prepare you for your opponent's tactics.




Automatically Appended Next Post:

 laginess wrote:
I ask because, while I am a very capable mathematician, I admit to not having any challenging opponents available to me. Most of the people in my group are newer than I am and aren't very good at tactics. And even when I play my friend that plays 40k semi-regularly, and is a major meta-gamer, the outcome is still heavily in my favor.

This is simply a product of your environment. I would suggest signing up for some tournaments. You'll get an ego check pretty quickly.

Experience is selective, how often are you going to remember you passing a lot of 2+ saves compared to rolling 3-4 1s?
Experience bends what is true to what is most dominant, hence rolling poorly for a unit that shouldnt gets stuck in your mind.

Math on the other hand is absolute, following mathhammer is the way to ensure the squad performs the way they do.

 Formosa wrote:
My experience is mathhammer doesn't work, the real world will always throw a wrench in the closed environment of mathhammer, it's a good indicator, but not something to be.relied on

This. Very much this. I've seen too many units under perform and over perform and nothing stays constant like the pretty little math tables. Mathhammer is a great way to suck the soul out of the game.

MarkCron wrote:
"So I have caused 3 wounds on your terminators - you get 2+ saves"
"three ones....well, that's bad luck"
"So I'm shooting overwatch with my pistol, 6 to hit, 6 to wound and you'll get a 2+ armour save. Oh, you died? Pity, your charge failed".
I'm for experience.

But you still take Terminators.
You still charge even if one dead model would cause the charge to fail.

Why? Because math says that you will probably make it.
If experience really trumps math, you wouldn't take terminators or charge

 Formosa wrote:
My experience is mathhammer doesn't work, the real world will always throw a wrench in the closed environment of mathhammer, it's a good indicator, but not something to be.relied on

If you let your experiences cloud what the math tells you will be disappointed over time. Yes the real world does throw a wrench into "the closed environment of mathhammer". It is those events that stick in our minds, (and make the game fun actually) but should not cloud your judgement. Does making a charge over 10" make for an awesome, potentially game breaking moment? Yes of course it does, but the math tells us typically this is NOT a good idea and should probably not be attempted most times.

i prefer mathhammering things out. knowing on average whats the chance of wrecking this tank with my wave serpent is nice, because if its low, then i would use my walkers lances.

 laginess wrote:
So I am curious, how well does crunching the numbers on a units capabilities compare to playing it against an experienced opponent?

I ask because, while I am a very capable mathematician, I admit to not having any challenging opponents available to me. Most of the people in my group are newer than I am and aren't very good at tactics. And even when I play my friend that plays 40k semi-regularly, and is a major meta-gamer, the outcome is still heavily in my favor.

Number crunching helps to prepare you for games. It helps you to properly analyze what the odds are of a specific outcome when your playing.
I've got a website where I just geek out about math and 40k. That's because its also fun!
However, mathhammer and theoryhammer are no substitutes for playing. You need to get some games in to learn how to apply your mathHammer.

According to your profile you live near Seattle. I suspect that if you start digging around, you will find some opponents there who can properly challenge you.

 Lord Arturius wrote:
This is somewhat accurate. Experience doesn't just mean <= math. Experience also means that you likely understand the flow of the game and the capabilities of many units. I'm willing to bet that the best players don't sit around and crunch numbers, they go out and play games and test out lists and units against varied opponents and skill levels.

They do both.
For example, I've been playing a jetSeer build lately. The math showed me how to build the army list. The games I've played showed me how I need to place them when playing -- for example putting the Baron up front in a V formation when closing on Tau with a AP2 weapon that ignores cover. Another example is how I learned how to use hit and run to bounce out of an assault, then rush back into the assault while assaulting a second, weaker unit that I could deliver a ton of wounds to -- causing all units to be swept.
Edit -- I don't claim to be a 'best player'. I'm a good player -- but not the best. But I've seen the 'best players' crunch numbers in addition to playing.


Automatically Appended Next Post:

MarkCron wrote:
"So I have caused 3 wounds on your terminators - you get 2+ saves"
"three ones....well, that's bad luck"

A smart player won't let that detur them. Sure. you might get a bad string of luck that throws off your game, but it does not mean the math does not work.

Look at blackjack games in a casino. Just because you did not bust that one time you hit on a 16 does NOT mean its a good strategy. If you believe that it is, I advise avoiding gambling facilities.

The fault of mathhammer is that it is based on large sample sizes and probability. The problem with this is that "large samples" usually exceeds dozens and dozens - even hundreds of samples.

This requirement isn't met on tables. It governs how unit A will perform in situation B over the course of 10 games worth of shooting rounds. This is meaningful in a coarse sense, but not in a specific shooting round, where marginal results can happen without causing any problems with the overall theory of math-hammer.

How many times has everyone here seen people on tables look at a dice result or an event that happened and say "that shouldn't happen"? Or, after a combat or a shooting phase say "well if I had rolled average". They don't complain near as much when they roll over average and decimate a unit of yours with great rolls.

The assumptions of math-hammer are not met on the table top. And people who rely on math-hammer instead of experience will forever get punched in the face by it at unexpected times.

This is a primary example of what I was just talking about. Lets review Gwyidion's post and change it to apply to blackjack. I altered the wording to fit blackjack, but the core of his argument is still the same.

The fault of probabilities is that it is based on large sample sizes and probability. The problem with this is that "large samples" usually exceeds dozens and dozens - even hundreds of samples.

This requirement isn't met on blackjack tables. It governs how hand A will perform in situation B over the course of 10 games . This is meaningful in a coarse sense, but not in a specific shooting round, where marginal results can happen without causing any problems with the overall theory of probabilities.

How many times has everyone here seen people on tables look at a card result or an event that happened and say "that shouldn't happen"? Or, after the dealer hits say "well he should had drawn average". They don't complain near as much when they draw great hands and hit blackjack again and again.

The assumptions of probabilities are not met on the table top. And people who rely on probabilities instead of experience will forever get punched in the face by it at unexpected times.

Gwyidion, I'm sure your a smart guy. You would not be playing warhammer if you were not. This is not a game for the intellectually challenged.
However, would you ever go to a casino and follow the paraphrased advise above? I can see why some people would, but I would rather trust in probabilities over experience.

Mathhammer is what you do when you are not playing Warhammer

They go hand and hand. All the Mathhammer in the world won't predict what the guy across from you is gonna do. Sure the mathhammer says that your squad x should ammihilate his squad y. But the whole reason he put squad y where he did was to get you to move squad x up so he could then eliminate it with squad Z!

I hate algebra...

Anywho, Dice rolls are random, and most of the time don't even out over the course of even several games. Knowing the matthammer might prevent you from dropping a unit just because it underperformed a couple of times. Not to mention there are several factors that are just so unpredicatable it is close to impossible to math out.

example...

Realistically my Fire Prism is only gonna get to fire that big gun 3 or 4 times a game(average 6 turns, 1 turn it will probabably be moving flat out, 1 or 2 turns it might already be dead). 1 out of the 3 or 4 times it is gonna miss. But that one turn that it catches those clumped up Terminators caught inside the burning wreckage of a Land Raider, that one shot might mean the game.

You also can't factor in the psychological factor of the big template on your oponents movement phase.

I guess what I am getting at is know your math AND know your enemy.

Kangodo wrote:

MarkCron wrote:
"So I have caused 3 wounds on your terminators - you get 2+ saves"
"three ones....well, that's bad luck"
"So I'm shooting overwatch with my pistol, 6 to hit, 6 to wound and you'll get a 2+ armour save. Oh, you died? Pity, your charge failed".
I'm for experience.

But you still take Terminators.
You still charge even if one dead model would cause the charge to fail.

Why? Because math says that you will probably make it.
If experience really trumps math, you wouldn't take terminators or charge

My examples were me talking... and I killed all the terminators next game as well.
I think experience trumps math because I still shoot at terminators, even though I should shoot at something else which I have a mathhammer greater chance of killing. (See...we can do this forever! )

ninjafiredragon wrote:i prefer mathhammering things out. knowing on average whats the chance of wrecking this tank with my wave serpent is nice, because if its low, then i would use my walkers lances.

Don't need mathhammer for that. Pick most dangerous unit. Shoot with strongest weapon. Dead? No - move to next available weapon. Can it hurt it? Yes = Shoot at most dangerous unit (repeat until "No" answer or opponent is tabled). If the answer was "No" = Pick next most dangerous unit. Can you hurt it.....etc etc.

As an example, if you know that the incoming helldrake is going to fry your troops and you'll lose the game - will you
a) Not shoot at it because Mathhammer tells you there is only a low chance of stopping it; OR
b) Shoot everything you have that can possibly pen or glance it?

If you answered (a) - wrong answer.

Loborocket wrote:
At its essence the game is a dice game. The math always wins. If every move you make puts you at a mathematical advantage you will win more than you lose. Experience is what gives you the intuition to know the moves that put you at a mathematical advantage, even without doing the math.

It also helps to play a list that already has the best mathematical chance of winning.

Math only helps in the list building. There is no math to tactical play. You still have to effectively and smartly use your optimized list to win a game. If you play like a dunderhead with your optimized list it won't matter how optimized your list is. Also if you play effectively with a non-optimized list it will get the job done quite often on the table.

Since dice rolling is still random, even the best optimized list can fall pray to bad dice performance.

Math is great if your only goal is to win no matter what. If you goal is a fun and interesting game math can add to that, but it can't provide it on its own.

Skriker

I've flirted with Mathhammer and in the end find experiance to be the better teacher. Mathhammer pushes you into very specific builds due to being mathematically the best (Nurgle and Heldrakes), but an experianced player can look at the less immediately impressive options and still make them work.

Mathhammer almost always seems to assume a unit works alone, often being played in a place with no cover.

Experiance instead teaches you how to play units together, to maximize your effects by using things together through things like area denial, counter-charge units and the like.

Plus I've seen Mathhammer has created this perception that units need to "make their points back" to be effective. It's as if that people forget that there are other things that units can be effective for.

Now could Mathhammer reliably tell me the average resault of something? Sure, but in a game where you don't roll hundreds of times the bell curve gets chucked out the window rather quickly and frankly knowing that means I don't prepare for the averages, I prepare for the extremes.

Experiance taught me that.

MarkCron wrote:
Don't need mathhammer for that. Pick most dangerous unit. Shoot with strongest weapon. Dead? No - move to next available weapon. Can it hurt it? Yes = Shoot at most dangerous unit (repeat until "No" answer or opponent is tabled). If the answer was "No" = Pick next most dangerous unit. Can you hurt it.....etc etc.

Is that really the best approach to target priority? Mathhammer will show you that's not the case.
Often your 'strongest weapon' is not the 'right weapon' to be using. Lets say that 6 wraiths are coming to smash your marine armies face in Flanking that unit is an annihilation barge.

The strongest weapon you have is a vindicator. The second strongest weapons you have are two tactical squads with bolters/PGs. According to your logic you should shoot the vindicator at the wraiths, then the bolters at the wraiths.

What you will discover through mathammer is that wraiths are vulnerable to massive STR 4 shooting. You would have been better served shooting the bolters at the wraiths and the vindicator at the annihilation barge. This maximizes your effect on the targets.

Instead the vindicator round is wasted with a small chance to kill even 1 wraith.
This is why mathhammer is important. By having a rough understanding of the proper probables your less likely to waste shots. Sure it happens when the vindiactor round will just deviate off the annaliation barge and not have any effect -- but the odds of a greater overall effect are higher. If you keep playing with better odds, you will experience a better result over the course of many games.

Gwyidion wrote:
The fault of mathhammer is that it is based on large sample sizes and probability. The problem with this is that "large samples" usually exceeds dozens and dozens - even hundreds of samples.

Yes and no.

Yes, because that is precisely where 99.9999999% of Mathhammer fails.

No, because it could be included in Mathhammer, if people wouldn't focus solely on the means (averages), but a complete set of statistic measures such as variance, range, standard deviation, standard error, etc.. But nobody does it, hence why most Mathhammer is - infact - fairly atrocious statistics (even if the math is - strictly speaking - correct).

 labmouse42 wrote:

MarkCron wrote:
Don't need mathhammer for that. Pick most dangerous unit. Shoot with strongest weapon. Dead? No - move to next available weapon. Can it hurt it? Yes = Shoot at most dangerous unit (repeat until "No" answer or opponent is tabled). If the answer was "No" = Pick next most dangerous unit. Can you hurt it.....etc etc.

Is that really the best approach to target priority? Mathhammer will show you that's not the case.
Often your 'strongest weapon' is not the 'right weapon' to be using. Lets say that 6 wraiths are coming to smash your marine armies face in Flanking that unit is an annihilation barge.

The strongest weapon you have is a vindicator. The second strongest weapons you have are two tactical squads with bolters/PGs. According to your logic you should shoot the vindicator at the wraiths, then the bolters at the wraiths.

What you will discover through mathammer is that wraiths are vulnerable to massive STR 4 shooting. You would have been better served shooting the bolters at the wraiths and the vindicator at the annihilation barge. This maximizes your effect on the targets.

Instead the vindicator round is wasted with a small chance to kill even 1 wraith.
This is why mathhammer is important. By having a rough understanding of the proper probables your less likely to waste shots. Sure it happens when the vindiactor round will just deviate off the annaliation barge and not have any effect -- but the odds of a greater overall effect are higher. If you keep playing with better odds, you will experience a better result over the course of many games.

I knew that without Mathhammer personally. Forcing saves through high volumes of dice is almost always the best way to kill things with good saves. But again that bell curve goes out the window when they make more saves than they should. On a 2+ Terminators shold fail at least one save in every 6 they take, but who hasn't seen a number of times where they make a considerable number more, or even less successful saves than that?

 labmouse42 wrote:

MarkCron wrote:
Don't need mathhammer for that. Pick most dangerous unit. Shoot with strongest weapon. Dead? No - move to next available weapon. Can it hurt it? Yes = Shoot at most dangerous unit (repeat until "No" answer or opponent is tabled). If the answer was "No" = Pick next most dangerous unit. Can you hurt it.....etc etc.

Is that really the best approach to target priority? Mathhammer will show you that's not the case.
Often your 'strongest weapon' is not the 'right weapon' to be using. Lets say that 6 wraiths are coming to smash your marine armies face in Flanking that unit is an annihilation barge.

The strongest weapon you have is a vindicator. The second strongest weapons you have are two tactical squads with bolters/PGs. According to your logic you should shoot the vindicator at the wraiths, then the bolters at the wraiths.

What you will discover through mathammer is that wraiths are vulnerable to massive STR 4 shooting. You would have been better served shooting the bolters at the wraiths and the vindicator at the annihilation barge. This maximizes your effect on the targets.

Instead the vindicator round is wasted with a small chance to kill even 1 wraith.
This is why mathhammer is important. By having a rough understanding of the proper probables your less likely to waste shots. Sure it happens when the vindiactor round will just deviate off the annaliation barge and not have any effect -- but the odds of a greater overall effect are higher. If you keep playing with better odds, you will experience a better result over the course of many games.

Actually, I somewhat agree. However, you've just highlighted the key issue. Experience says that if the wraiths get to charge you will lose not just that unit, but probably the chance to kill the wraiths with shooting at all (we'll assume a competent 40k player who can organise an assault to finish it off in the opponent phase - yes, I'm aware that mathhammer can be said to have a role to play in that ). So, your priority is to kill the wraiths. From your tac squads, lets assume mathhammer says that on average 2 squads of tac marines (x10) will take down 6 wraiths (I don't know if that is true by the way, but I'm SURE someone will work that out ).

Mathhammer doesn't tell you the order to shoot though. So, you could shoot the Vindi at the AB first and it would make no difference to the mathhammer - the wraith squad should die from Tac marine fire anyway. Experience says - shoot the tac marines first - because if the wraiths don't die, you need to keep shooting at them or you could lose the game.

That's why I prefer experience over mathhammer - knowing that on average 20 tac marines with bolters kills 7.65 wraiths (made that number up by the way - figured I would put that in before everyone races to correct my math!) doesn't take away the pain when the wraiths don't die.

ClockworkZion wrote:
I knew that without Mathhammer personally. Forcing saves through high volumes of dice is almost always the best way to kill things with good saves.

That is an excellent point, yet that's not always the case. Let me give you another example. Lets say that instead of 6 wraiths there are 10 plague marines (units with good saves) Each bolter shot you fire has a 4/81 (just under 5%) of killing a PM. In this case, knowing the RPPvalue of PMs to bolters you would know that shooting bolters at them is a waste, and they should instead be used on a different target. On the other hand, a PMs RPP to vindicators is extremely low so shooting them with is bank.

You may have noticed that your experience has shown the same results as Mathhammering. This is a good thing! This provides key point indicators that our mathhammer works, giving us encouragement that further items that we can gleam from mathhammering will be valid.

ClockworkZion wrote:
But again that bell curve goes out the window when they make more saves than they should. On a 2+ Terminators shold fail at least one save in every 6 they take, but who hasn't seen a number of times where they make a considerable number more, or even less successful saves than that?

Your right in the sense that you cannot throw 30 wounds on 5 marines and expect them all to be dead -- but they will probably be dead or very hurt. What the mathhammer tells us is that we can expect that 30 wounds are what it will take to get that result. Sure, it might happen after 5 wounds, but odds are it will happen when your getting closer to 30.
This is a separate item that we can gleam other from the other item -- what weapon to shoot at a target.

There's a difference between using Mathhammer as a guideline and using it as your sole source of consultance for everything ever.

MarkCron wrote:
Actually, I somewhat agree.

I think most of the time people get intimidated by mathhammer because it seems complicated or does not make sense. I've been trying to focus on a 'mathhammer for everyone' concept where I explain how the mathhammer applies to players in 40k. Listen to the 11th company podcast, I've been doing weekly segments on mathhammer for the past 2 months.

MarkCron wrote:
Mathhammer doesn't tell you the order to shoot though. So, you could shoot the Vindi at the AB first and it would make no difference to the mathhammer - the wraith squad should die from Tac marine fire anyway. Experience says - shoot the tac marines first - because if the wraiths don't die, you need to keep shooting at them or you could lose the game.

Right. What it tells you is your best chances to complete your task.
By knowing that the wraiths have low RPP values to bolters, you can start with the bolters and see the results. If the wraiths were badly injured and you can take the assault, then you can focus on the barge with the vindicators. However, if the wraiths are unscratched, then you need to hit them with the vindicator to soften them up a touch.

Mathhammering does not give you an absolute 'play this way'. Instead its a tool to add to your understanding of the game that assists you while playing.

 labmouse42 wrote:

ClockworkZion wrote:
I knew that without Mathhammer personally. Forcing saves through high volumes of dice is almost always the best way to kill things with good saves.

That is an excellent point, yet that's not always the case. Let me give you another example. Lets say that instead of 6 wraiths there are 10 plague marines (units with good saves) Each bolter shot you fire has a 4/81 (just under 5%) of killing a PM. In this case, knowing the RPPvalue of PMs to bolters you would know that shooting bolters at them is a waste, and they should instead be used on a different target. On the other hand, a PMs RPP to vindicators is extremely low so shooting them with is bank.

You may have noticed that your experience has shown the same results as Mathhammering. This is a good thing! This provides key point indicators that our mathhammer works, giving us encouragement that further items that we can gleam from mathhammering will be valid.

Not to be difficult, but I don't think that is the same thing at all. ClockworkZion is using basic probability - "In order for them to roll more 1's I should make them roll more dice" ==> shoot more shots -the chance of them actually wounding is irrelevant as long as you get a hit. Here again - basic probability - shoot more shots - more hits.

From what you are saying, mathhammering allows more precise allocation -> "I'm not going to shoot weapon x because weapon Y has a higher chance". That isn't the same as saying "Well, I can get a wound with this weapon, I'm going to shoot at this until it is dead"