Cost efficient blood hunting

[JimMorrison]

Originally Posted by Redeyes

Couldn't it be, like conventional wisdom says, that rods have no effect for 0 blood blood hunters?

This is true, and it's funny to note the massive disparity between the 2 test groups.

Originally Posted by Redeyes

Randomness would account for the entire difference in the test, then.
100 tests is ultimately a fairly small sample size.

Well somewhat, but bear in mind, in a way it's really 600 tests, compiled in batches of 100, 200, and 300 at a time. The fact that the tests with B1/B2/B3 all show fairly small levels of deviation, and a clear trend simply implies that using B0's is in fact just sporadic and unreliable at best, whereas we can come to expect a certain level of performance from real blood hunters, assuming Pop>5000 and Unrest<1.


And I missed the part where this was LA. Anyway my point was that if it is default SiteFreq, then we can extrapolate from there, and as long as we estimate cautiously, we shouldn't be too disappointed.

[archaeolept]

Originally Posted by ano

It was LA Mictlan, not EA.
Also, does anybody have a real proof of the fact that effectiveness of blood hunt depends on magic site frequency?

Yes

Originally Posted by chrispedersen
Gut tells me its statistically significant. Haven't had to do such calculations in 20 years.. but it seems so.

No

edit: also, nice tests there Tmoe!

[JimMorrison]

This is really an excellent resource! Always knew Pans were pricey for Blood, but wow, poor Pangaea.

It's funny, with Lanka, I always use Kala-Mukhas for my hunters, as there are so many uses for the 75% that don't get their Blood pick.

It's a fun list, shows why EA+LA Mictlan are the "kings of blood", and there are several others who get around 1/1 sufficiency in their hunting, and most others with Blood access are really paying for it.

(Oh and one note for beginners who read this list - the Bogarus Five-Fold Angel looks like a great Blood Hunter, but they cause additional Unrest in any province they are in, so unless you have Patrollers, which cost more upkeep and kill more pop, you will have troubles.)

[chrispedersen]

Prophetize your blood hunter = )

[Tmoe]

Originally Posted by JimMorrison

This is really an excellent resource! Always knew Pans were pricey for Blood, but wow, poor Pangaea.

It's funny, with Lanka, I always use Kala-Mukhas for my hunters, as there are so many uses for the 75% that don't get their Blood pick.

Yeah Pangaea got it pretty bad when you consider that even the independent scout has better blood slave/upkeep cost ratio .

I didn't quite understand what you meant with the term Blood pick and why Kala-mukhas would be better? I would assume Raktapata is the natural blood hunter choice for Lanka since they get the most out of the SDR and have the best slave/upkeep ratio.

And to add some more data, here's the list for the most slaves captured in a single turn by the test groups.

Results for 0-lvl hunters
1 x 0-lvl - Most slaves found in single turn: 13
2 x 0-lvl - Most slaves found in single turn: 9
3 x 0-lvl - Most slaves found in single turn: 16

1 x 0-lvl w rod - Most slaves found in single turn: 9
2 x 0-lvl w rod - Most slaves found in single turn: 9
3 x 0-lvl w rod - Most slaves found in single turn: 20


Results for 1-lvl hunters
1 x 1-lvl - Most slaves found in single turn: 10
2 x 1-lvl - Most slaves found in single turn: 18
3 x 1-lvl - Most slaves found in single turn: 24

1 x 1-lvl w rod - Most slaves found in single turn: 17
2 x 1-lvl w rod - Most slaves found in single turn: 21
3 x 1-lvl w rod - Most slaves found in single turn: 39


Results for 2-lvl hunters
1 x 2-lvl - Most slaves found in single turn: 15
2 x 2-lvl - Most slaves found in single turn: 24
3 x 2-lvl - Most slaves found in single turn: 29

1 x 2-lvl w rod - Most slaves found in single turn: 15
2 x 2-lvl w rod - Most slaves found in single turn: 24
3 x 2-lvl w rod - Most slaves found in single turn: 36


Results for 3-lvl hunters
1 x 3-lvl - Most slaves found in single turn: 17
2 x 3-lvl - Most slaves found in single turn: 29
3 x 3-lvl - Most slaves found in single turn: 31

1 x 3-lvl w rod - Most slaves found in single turn: 13
2 x 3-lvl w rod - Most slaves found in single turn: 32
3 x 3-lvl w rod - Most slaves found in single turn: 37

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

Unfortunately I cant attach the open office calc document where all the raw data and results are but if somebody wants it they can throw me a message and i can email it to you

[JimMorrison]

Originally Posted by Tmoe

I didn't quite understand what you meant with the term Blood pick and why Kala-mukhas would be better? I would assume Raktapata is the natural blood hunter choice for Lanka since they get the most out of the SDR and have the best slave/upkeep ratio.

Because the Kala-Mukha is 1B + 1NDB. I like the 2B as Blood Hunters, because then I can use them freely, without reliance on the SDR. Since they can reanimate, they can remote search Death, they Preach well as H2, and they get useful minor forging options (always need Bags of Wine!), I feel they are worth the investment. And I forgot it's 33% that get B2, so really they're a good option, depending on how you like to use them.

Also, you can shift to 2 B2+SDR per province, or do a little patrolling, and really start to milk your efficiency. At a certain point, Lanka is happy with enough gold to buy 1 Raksharaja and X Palankashas per turn, as long as you can pull in enough slaves to destroy your foe.

[chrispedersen]

Originally Posted by archaeolept

Yes

No

edit: also, nice tests there Tmoe!

Rather than just saying no - you wanna provide a calculation?

[ano]

Once again, where is information that blood hunting depends on site frequency from? I also heard about it and probably it is so but is there anyone who can give a solid proof?

[ano]

This is true, and it's funny to note the massive disparity between the 2 test groups.

Maybe it's another bug and SDR not raises but lowers bloodhunting quality for those who have no skill in Blood magic?

[vfb]

Originally Posted by ano

Maybe it's another bug and SDR not raises but lowers bloodhunting quality for those who have no skill in Blood magic?

Or maybe 100 tests is too small a sample size to get accurate numbers for an event with an expected occurrence of 1 in 10, especially when there is a 1 in 6 chance of the event producing extraordinary results when it does occur.