Seeking Arrow Test (Updated 1st June 2009 to V1.1 with Death Bless tests)

[llamabeast]

I think the death bless afflictions have already been confirmed with other overland spells, so probably no need to test it again.

Also Zeldor, seeking arrow is AN, so armour is useless.

[Calahan]

What surprised me most about the Death Bless was that out of the units that were tagged as 'Survived+Wounded', almost all of the time they only had one wound. This was consistent across all the tests.

Not once did any survivor have multiple wounds directly from the 'Seeking Arrow' attack. The only time I did see more than one wound, and accounts for a very small percentage (about 1 in 50), was when a unit got diseased through the attack, and (I'm guessing) picked up a second wound through the normal result of being diseased (which happens after the magic phase). I never once saw a unit with two wounds without 'Diseased' being one of them. And only once did this occur with one of the Death Bless tests. (since I was looking out for more than one wound, and always checked the details if the icon had more than one red heart)

I will be running some more tests with the Death Bless against higher HP units when I get chance (maybe the weekend). But my current predictions are that because the Death Bless made no difference to the 82 Scouts in the 'Survived+Wounded v Death Bless' category, I'd be surprised if all 82 examples were anomalous. If the Death Bless worked with 'Seeking Arrow', then to me at least, some should have had multiple afflictions (and not brought about via disease).

Only further testing will prove things either way.

[Agema]

That's just probability.

The chance of suffering an affliction relates to the amount of damage the unit suffered. Thus the units most likely to be afflicted from a first arrow will have the fewest remaining HP, and thus stand the least chance of surviving a second arrow, and if they do survive the second arrow, will have taken such little damage there is a very low chance of taking an affliction from it.

If you'll pardon me for extreme geekiness, but...

The optimal chance of seeing two afflictions from one 10HP target will be when it takes 4 or 5 damage from the first arrow. Let's say 5 HP damage: 50% will be afflicted. The second arrow needs to do 1-4 damage to get a second affliction (any more damage kills), which will occur in (estimated) around 10-15% of cases. These 1-4 damage hits then have 10-40% of inflicting another affliction (leaning towards the 40%). Without going through exactly and taking an overestimate, 50% x 15% x 40% is 3%.

Now factor in how many take 5 damage from the first arrow, you're probably looking at something like 0.2-0.3% of the total trial n value having two afflictions. Expand to include everyone who took damage off and survived two arrows (i.e. the same principles for 1-8HP from the first arrow), the chance of two afflictions is probably <1%. Then also consider how many of your n=300 per trial will not have been hit by 2 arrows anyway: the majority either not hit or hit by just one (killed or otherwise).

Consequently, dual afflicted scouts should be expected to be incredibly rare in this experiment.

[Calahan]

Thanks for the views and ideas everyone. It is certainly all appreciated.

I hope to find some time today to sort out the Death Bless part of this test, since that seems to be the main uncertainty at hand.

Originally Posted by Agema

If you'll pardon me for extreme geekiness, but...

The optimal chance of seeing two afflictions from one 10HP target will be when it takes 4 or 5 damage from the first arrow. Let's say 5 HP damage: 50% will be afflicted. The second arrow needs to do 1-4 damage to get a second affliction (any more damage kills), which will occur in (estimated) around 10-15% of cases. These 1-4 damage hits then have 10-40% of inflicting another affliction (leaning towards the 40%). Without going through exactly and taking an overestimate, 50% x 15% x 40% is 3%.

Now factor in how many take 5 damage from the first arrow, you're probably looking at something like 0.2-0.3% of the total trial n value having two afflictions. Expand to include everyone who took damage off and survived two arrows (i.e. the same principles for 1-8HP from the first arrow), the chance of two afflictions is probably <1%. Then also consider how many of your n=300 per trial will not have been hit by 2 arrows anyway: the majority either not hit or hit by just one (killed or otherwise).

Consequently, dual afflicted scouts should be expected to be incredibly rare in this experiment.

Thank you very much for your input Agema, and I've never thought geekiness needed to be excused It's all good in my books.

I think there are a few key mechanics in this test which are not 100% known (at least not to my knowledge)

1)What is the exact damage from the 'Seeking Arrow' spell? Is it.....
a) 1-8
b) 1-8 + DRN
c) 8
d) 8 + DRN

I'm assuming it's probably b) but may well be d), but maybe that's the wrong assumption.

2)If the target survives a 'Seeking Arrow' attack, does it have a separate, and independent chance of receiving an affliction linked to the spell itself? And if so, does it follow the usual affliction mechanics (manual page 81) or does it not matter if you have 10, 100, or 1000 HP?

3)Is it possible for a single 'Seeking Arrow' to cause more than one affliction?

I must confess to not having that much experience of using or receiving 'Seeking Arrow' spam, so some of the above may be unknown to me, but common knowledge to others. But hope to do tests that will at least attempt to clear-up some of the above (if just for my own curiosity).

Not even sure why I'm picking on the 'Seeking Arrow' spell. Since there are probably dozens of spells that need testing.

Originally Posted by Micah

Will death bless allow more than 1 affliction per hit? If not it makes sense that it wouldn't raise the chances as much as advertised since the multiplier would stop at 100%. So if you do 5 damage to an 10 HP unit that's a 50% chance of an affliction, the most the D bless could do at that point is doubling it (+100%) which would put it up to a 100% affliction chance, any additional chance would be wasted. (And of course any testing for this should not include the D9 bless, since I assume that would allow a second affliction due to being 2 different damage checks.)

Thanks for your views Micah. I feel it's always appreciated by the community when one of the very top players shares their thoughts on a subject. So thank you for that.

Yes, whether or not a single 'Seeking Arrow' can cause more than one affliction with a Death Bless is a key point and needs to be tested for. I will also test to see if the D9/D10 bless causes more than one affliction. Since it's obvious that if the tests show that a D4-D8 bless can only cause one affliction, but a D9/D10 can cause 2+, then your assumption that there are two separate attacks, and two damage checks taking place, should be correct.

[Redeyes]

Originally Posted by Calahan

a) 1-8
b) 1-8 + DRN
c) 8
d) 8 + DRN

I'm assuming it's probably b) but may well be d), but maybe that's the wrong assumption.

Supposedly 8 + DRN - DRN, if AN works like an attack against 0 protection

[Zeldor]

Maybe you need to wait one turn before putting shroud on the caster before death bless sticks, there are some problems with bless from shrouds. And try a volley of arrows against tartarian. 3-5 arrows should do enough afflictions, if it works.

[Illuminated One]

If you'd want to test the afflictions only you could mod the Seeking Arrow Spell to do say 50 damage and test it on 100 hp units.

That way you'd have 50% affliction chance without Death Bless and only a little risk of killing the units.

[MaxWilson]

I tested death bless (using either my prophet or a shroud, I think a prophet) and no death bless. I don't remember the exact figures, but it was less than +350% boost and more than 0%. I believe affliction rate approximately doubled. The post should be somewhere in the forum archives for 2007, around the time we first discovered that the affliction chance affects spells at all. (I remember KO being surprised.)

I'm interested to see if Calahan gets the same results when he tests. It isn't science unless it's repeatable.

-Max

[Micah]

Will death bless allow more than 1 affliction per hit? If not it makes sense that it wouldn't raise the chances as much as advertised since the multiplier would stop at 100%. So if you do 5 damage to an 10 HP unit that's a 50% chance of an affliction, the most the D bless could do at that point is doubling it (+100%) which would put it up to a 100% affliction chance, any additional chance would be wasted. (And of course any testing for this should not include the D9 bless, since I assume that would allow a second affliction due to being 2 different damage checks.)

[chrispedersen]

yeah the +400% affliction is amusing when you combine it with that spell Blood spell that does 1AN to the whole battlefield.
Or roughly 40% chance of an affliction... to a normal human.