Teclis/Book of Hoeth IF chance

[Curu Olannon]

Well, although I probably should've been able to calculate this by means of simple math, I'm not - so I decided to write a program that would numerically calculate the chance of Teclis / an archmage with BoH casting with IF (running each cast a million times should yield pretty accurate results). EDIT: Put up the entire table:

.::. 5+ .:-:. 6+ .::. 7+ .:-:. 8+ .::. 9+ .:-:. 10+ .::. 11+ .:-:. 12+
2- 11.0% : 11.1% : 8.27% : 8.32% : 5.52% : 5.53% : 2.75% : 2.79%
3- 36.9% : 35.6% : 35.1% : 32.4% : 29.6% : 26.3% : 22.1% : 18.0%
4- 59.0% : 58.9% : 59.0% : 58.7% : 57.7% : 55.6% : 53.5% : 49.2%
5- 71.1% : 71.1% : 71.1% : 71.1% : 71.0% : 70.9% : 70.7% : 70.1%

Just some interesting values to bear in mind Don't know if this fits in tactics but I suppose this is the (only?) place someone cares (?) about this

[Orpheus]

http://www.remanlegions.com/Math-Hammer.pdf

this may be of use to you

[Curu Olannon]

Thanks, but this does not take into account that you actually have to beat a value as well, for example casting 2-3 dice on hard spells. Still a very nice sheet though, I'll be sure to save it This may not be quite that relevant but as you can see, in certain cases it can make a difference. It's all about optimizing those power dice

[madelmo]

an IF is a success, regardless of what the casting value of the spell is.

IE: 2 2's and a a 1 will make a pit of shades succeed with IF.


plz correct me if i'm wrong

[Foxbat]

Actually, in both locations within the HE Army Book (pg 66 & pg 102) that refer to the any doubles IF (except for a Miscast of course) use the same language and indicate that an IF applies on a “successful casting roll”. Traditionally, this has been taken to mean that the result must be greater than the spell’s casting value.

[Alathenar]

For me, the casting value has to be equal to or higher with a double, except miscast. Foxbat has got it correct.

[Silver]

an IF is a success, regardless of what the casting value of the spell is.

IE: 2 2's and a a 1 will make a pit of shades succeed with IF.


plz correct me if i'm wrong

For normal IF yes, for Book of Hoeth & Teclis IF, you need the min. cast value and a double.

[Curu Olannon]

Spells cast IF with Teclis/Hoeth need to beat the value as well as getting doubles, this is pretty clear form the armybook as Foxbat pointed out. I've edited my original post now to include the entire table.

I've been thinking a bit about how this can be applied in practice but I haven't reached any gamebreaking ideas yet. I suppose one way to look at the power granted by BoH/Teclis is to calculate each die's "IF value" - meaning how much each die contributes to IF. I'll give you an example:
Casting "Fury of Khaine" on 3 dice gives you a 32,4% chance to get it off at IF. This means each die provides roughly 10,5 "IF %". With 4 dice on the other hand, not only do you increase to 59%, dividing the dice you get that each die provides roughly 15 "IF %". With 5 dice you increase to 71,1%, however per dice this yields roughly 14. As such, from a pure effectiveness point of view, 4 dice is the optimal for casting fury of khaine if you want to capitalize on the effectiveness granted from trom teclis/BoH.

Naturally, if you want a spell to go off with the highest % possible, you of course roll 5 dice. However, often Teclis will be supported by other mages, which do not have "Free IF" (not to mention you might want to save PD for several spells) in which case capitalizing on effectiveness should be a priority. Even with 71% it's a good chance you won't IF anyways.

[SpellArcher]

Plus the chance of miscasting goes up substantially from 4 to 5 dice I suspect. Less of a problem for Teclis than the Book guy I guess.

Thanks for this though. It's another reason to cast on four dice I reckon, which was a decent rule of thumb.

[Stormie]

Yeah, I gotta ask, did you factor in the chance of miscasts into the results?

E.g. if you roll 4 dice needing 8+ to cast, does your system count 5-5-1-1 as IF or fail?

It's very good even so

[Curu Olannon]

Yeah, I gotta ask, did you factor in the chance of miscasts into the results?

E.g. if you roll 4 dice needing 8+ to cast, does your system count 5-5-1-1 as IF or fail?

It's very good even so

This would yield miscast. The system first checks miscast, then proceeds to other doubles I think I've taken care of all the angles, the numbers should be pretty accurate ^^

I'm currently working on other HE related special mathhammer issues (for example, chance of a noble scoring a killing blow with loec/white sword). Is there any situation you can think of where you'd like to know the statistical outcome?

I think I'll write some methods (java, for those of you who're into geekstuff), getting the most important stuff together and then make a nice little HE sheet. What are people's thoughts on this?

[Bel-Hathe]

. What are people's thoughts on this?

Would love it.

[SpikeyElf]

Is there any situation you can think of where you'd like to know the statistical outcome?

Perhaps what's better a 2+ standard AS or a 3+ re-roll (our offensive dragon guy comes into mind).

Would love it too.

[Siegfried VII]

The statistics about the killing blow chances of a Prince/Noble interest me too.

[tiekwando2]

Chance to make a save
2+ 3+(reroll)
s3 .833 .889
s4 .667 .75
s5 .5 .556
s6 .333 .306*
s7 .167 .0

*also chance to get a 6 with re-roll so killing blow after hits, if against opponent with equal ws .75*.306=.229*4=.917
against opponent with lower ws .889*.306=.272*4=1.09

so prince with loec +whitesword should normally get a killing blow

[Curu Olannon]

Chance to make a save
2+ 3+(reroll)
s3 .833 .889
s4 .667 .75
s5 .5 .556
s6 .333 .306*
s7 .167 .0

*also chance to get a 6 with re-roll so killing blow after hits, if against opponent with equal ws .75*.306=.229*4=.917
against opponent with lower ws .889*.306=.272*4=1.09

so prince with loec +whitesword should normally get a killing blow

This is wrong. You've calculated the expected value, not the chance of it happening. Let me give you an example:

If you toss a coin 3 times, your expected value for getting tails is 1,5. However, you are not guaranteed to get tails, in fact your chance of getting tails is 1-(0,5^3) which is roughly 87,5%. Same applies for Loec.

I've written up a decent amount now and I'll post it up soonish

[tiekwando2]

Ahh too true, back to stats class for me