Effect of roster strength on contract spawned enemy sizes

[Drathnar1]

This topic follows from the previous discussion here: http://steamcommunity.com/app/365360/discussions/0/1486613649674664524/

As that wasn't the original topic for the above thread, I felt it would be good to make a separate thread for it seeing as there appears to be a sizable number of people interested in it.

Here I'm only looking at one particular contract: a 3-skull contract to clear out brigands on day 68, worth 2110G.

My original setup: 9 L10 men, 3 L9 men
Reloading and retaking the contract several times gives the following encounters (number in brackets denotes total number)

(21) 13 raiders, 7 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders
(25) 18 raiders, 5 marksmen, 2 leaders
(21) 12 raiders, 7 marksmen, 1 leader, 1 swordmaster
(20) 12 raiders, 6 marksmen, 1 leader, 1 master archer
(22) 17 raiders, 4 marksmen, 1 leader
(25) 17 raiders, 6 marksmen, 2 leaders
(25) 18 raiders, 5 marksmen, 2 leaders

So there is obviously some degree of variation even when just reloading and retaking the contract. Assuming we lump raiders and marksmen together as tier2 and leaders/ swordmasters/ master archers as tier3, that give a range of roughly 16-23 tier2 and on average nearly 2 tier3.


Next, we dismiss all the men and recruit a whole new bunch: 11 are L1 and 1 L2. As expected the encounter results are quite different:

(10) 7 raiders, 3 marksmen
(14) 9 raiders, 5 poachers
(11) 5 raiders, 6 marksmen
(9) 7 raiders, 1 marksman, 1 leader
(9) 6 raiders, 3 marksmen
(10) 7 raiders, 2 marksmen, 1 leader
(14) 9 raiders, 5 poachers
(12) 9 raiders, 3 marksmen

Evidently loss of levels has a marked effect on enemy strength even though the party is still at 12 men. Now tier1 enemy units spawn and tier2 spans 8-12 in numbers. Tier3s are mostly not seen. But this scenario is not interesting as no one would actually dismiss levelled up men.


In this next set of tests, we keep all original 12 men and recruit another 8 men (7 L1s and 1L2), bringing the total number to 20. The new men are all in reserve of course. I expected the encounters to be about the same as the 1st set of tests or maybe very slightly tougher. I was wrong. Results were:

(15) 11 raiders, 3 marksmen, 1 leader
(16) 11 raiders, 4 marksmen, 1 leader
(19) 4 thugs, 10 raiders, 4 marksmen, 1 leader
(16) 11 raiders, 4 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders
(17) 12 raiders, 4 marksmen, 1 leader
(15) 10 raiders, 4 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders

As is evident, this 3rd set has substantially weaker encounters compared to the 1st.
It appears to me that men in reserves DO have an effect on encounter tuning. It also appears that it may be likely that the average level of the entire party (including reserves) is used to adjust enemy troop sizes.

I find it gamey to tack on a few level 1s to bring down average level if this is indeed the case.


We need more tests and I'd encourage others to chip in with their findings too.

[Tephros83]

Originally posted by Drathnar1:

This topic follows from the previous discussion here: https://steamcommunity.com/app/365360/discussions/0/3183345176712640101/

As that wasn't the original topic for the above thread, I felt it would be good to make a separate thread for it seeing as there appears to be a sizable number of people interested in it.

Here I'm only looking at one particular contract: a 3-skull contract to clear out brigands on day 68, worth 2110G.

My original setup: 9 L10 men, 3 L9 men
Reloading and retaking the contract several times gives the following encounters (number in brackets denotes total number)

(21) 13 raiders, 7 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders
(25) 18 raiders, 5 marksmen, 2 leaders
(21) 12 raiders, 7 marksmen, 1 leader, 1 swordmaster
(20) 12 raiders, 6 marksmen, 1 leader, 1 master archer
(22) 17 raiders, 4 marksmen, 1 leader
(25) 17 raiders, 6 marksmen, 2 leaders
(25) 18 raiders, 5 marksmen, 2 leaders

So there is obviously some degree of variation even when just reloading and retaking the contract. Assuming we lump raiders and marksmen together as tier2 and leaders/ swordmasters/ master archers as tier3, that give a range of roughly 16-23 tier2 and on average nearly 2 tier3.


Next, we dismiss all the men and recruit a whole new bunch: 11 are L1 and 1 L2. As expected the encounter results are quite different:

(10) 7 raiders, 3 marksmen
(14) 9 raiders, 5 poachers
(11) 5 raiders, 6 marksmen
(9) 7 raiders, 1 marksman, 1 leader
(9) 6 raiders, 3 marksmen
(10) 7 raiders, 2 marksmen, 1 leader
(14) 9 raiders, 5 poachers
(12) 9 raiders, 3 marksmen

Evidently loss of levels has a marked effect on enemy strength even though the party is still at 12 men. Now tier1 enemy units spawn and tier2 spans 8-12 in numbers. Tier3s are mostly not seen. But this scenario is not interesting as no one would actually dismiss levelled up men.


In this next set of tests, we keep all original 12 men and recruit another 8 men (7 L1s and 1L2), bringing the total number to 20. The new men are all in reserve of course. I expected the encounters to be about the same as the 1st set of tests or maybe very slightly tougher. I was wrong. Results were:

(15) 11 raiders, 3 marksmen, 1 leader
(16) 11 raiders, 4 marksmen, 1 leader
(19) 4 thugs, 10 raiders, 4 marksmen, 1 leader
(16) 11 raiders, 4 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders
(17) 12 raiders, 4 marksmen, 1 leader
(15) 10 raiders, 4 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders

As is evident, this 3rd set has substantially weaker encounters compared to the 1st.
It appears to me that men in reserves DO have an effect on encounter tuning. It also appears that it may be likely that the average level of the entire party (including reserves) is used to adjust enemy troop sizes.

I find it gamey to tack on a few level 1s to bring down average level if this is indeed the case.


We need more tests and I'd encourage others to chip in with their findings too.

Thank you for looking into this. If that's how it works, it seems fine. It would be harder to train new recruits if the 20th guy is level 1 and it made the enemies harder.

[Azov]

Originally posted by Tephros83:

Originally posted by Drathnar1:

This topic follows from the previous discussion here: https://steamcommunity.com/app/365360/discussions/0/3183345176712640101/

As that wasn't the original topic for the above thread, I felt it would be good to make a separate thread for it seeing as there appears to be a sizable number of people interested in it.

Here I'm only looking at one particular contract: a 3-skull contract to clear out brigands on day 68, worth 2110G.

My original setup: 9 L10 men, 3 L9 men
Reloading and retaking the contract several times gives the following encounters (number in brackets denotes total number)

(21) 13 raiders, 7 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders
(25) 18 raiders, 5 marksmen, 2 leaders
(21) 12 raiders, 7 marksmen, 1 leader, 1 swordmaster
(20) 12 raiders, 6 marksmen, 1 leader, 1 master archer
(22) 17 raiders, 4 marksmen, 1 leader
(25) 17 raiders, 6 marksmen, 2 leaders
(25) 18 raiders, 5 marksmen, 2 leaders

So there is obviously some degree of variation even when just reloading and retaking the contract. Assuming we lump raiders and marksmen together as tier2 and leaders/ swordmasters/ master archers as tier3, that give a range of roughly 16-23 tier2 and on average nearly 2 tier3.


Next, we dismiss all the men and recruit a whole new bunch: 11 are L1 and 1 L2. As expected the encounter results are quite different:

(10) 7 raiders, 3 marksmen
(14) 9 raiders, 5 poachers
(11) 5 raiders, 6 marksmen
(9) 7 raiders, 1 marksman, 1 leader
(9) 6 raiders, 3 marksmen
(10) 7 raiders, 2 marksmen, 1 leader
(14) 9 raiders, 5 poachers
(12) 9 raiders, 3 marksmen

Evidently loss of levels has a marked effect on enemy strength even though the party is still at 12 men. Now tier1 enemy units spawn and tier2 spans 8-12 in numbers. Tier3s are mostly not seen. But this scenario is not interesting as no one would actually dismiss levelled up men.


In this next set of tests, we keep all original 12 men and recruit another 8 men (7 L1s and 1L2), bringing the total number to 20. The new men are all in reserve of course. I expected the encounters to be about the same as the 1st set of tests or maybe very slightly tougher. I was wrong. Results were:

(15) 11 raiders, 3 marksmen, 1 leader
(16) 11 raiders, 4 marksmen, 1 leader
(19) 4 thugs, 10 raiders, 4 marksmen, 1 leader
(16) 11 raiders, 4 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders
(17) 12 raiders, 4 marksmen, 1 leader
(15) 10 raiders, 4 marksmen, 1 leader
(18) 12 raiders, 4 marksmen, 2 leaders

As is evident, this 3rd set has substantially weaker encounters compared to the 1st.
It appears to me that men in reserves DO have an effect on encounter tuning. It also appears that it may be likely that the average level of the entire party (including reserves) is used to adjust enemy troop sizes.

I find it gamey to tack on a few level 1s to bring down average level if this is indeed the case.


We need more tests and I'd encourage others to chip in with their findings too.

Thank you for looking into this. If that's how it works, it seems fine. It would be harder to train new recruits if the 20th guy is level 1 and it made the enemies harder.

it wouldn't, you would just put nice armor on the new guy and make him stand behind all the time, plus, veterans are more than capable to defend flanking

[Estieukua]

On the other hand, softer and less numerous opponents to fight mean a lesser gain of XPts and make it harder to speed-level newcomers, I guess.

But I do think it would be better if opponents were simply scaled to the 12 highest-level members of your company, regardless of who's currently in/out of reserve.

[nightworg]

Awesome work. Very interesting. I think it is fine if the average level is used. Sure you get lesser xp, but I rather have an easy fight when adding level one brothers.

[Drathnar1]

Thanks all. I'll try to write up something at some point. I'd really like more data at this stage though. I usually like to wait a little on topics that I feel require work from more than 1 person. Establishing enemy melee skill range is one such topic where numbers can vary quite a bit and are influenced by quite a few factors. Thankfully Laiseurjika has stepped in with some results :)

Similarly, the size and composition of enemies spawned from contracts are determined by many factors.
Quoting Rap:
Contract difficulty scales with time, the strength of your roster, the difficulty rating (skulls), the type of contract, and some factors specific to the type of contract

I try to keep the other factors constant but there is always the likelihood that I've overlooked something. That's why rechecking and retesting from an independent, separate source(s) is vital.

[千仞万渊]

Although the result is totally contrary to my playing through and tests, I'll try to make a similar test once I have spare yime

[Estieukua]

Originally posted by 千仞万渊:

Although the result is totally contrary to my playing through and tests, I'll try to make a similar test once I have spare yime

Maybe the day makes a difference. The above data was from day 68. I think (but don't remember exactly) that some of the contrary things you mentioned before were from an earlier point in the game.

[Drathnar1]

Originally posted by Laiseurjika:

Originally posted by 千仞万渊:

Although the result is totally contrary to my playing through and tests, I'll try to make a similar test once I have spare yime

Maybe the day makes a difference. The above data was from day 68. I think (but don't remember exactly) that some of the contrary things you mentioned before were from an earlier point in the game.

I postulate the difference is due to the balance between additional men/levels and average level.

Note that in an earlier test in the original thread this discussion was branched from, I did get harder encounters with more men. But that was when I had like a mix of L3 and L1 men. So I'd say at low levels, adding more men doesn't decrease the average level by very much, and the effect of more men takes precedence.
At high levels, with a party of mostly L10 men, taking up a bunch of L1 men depresses average level pretty significantly, resulting in a reduction in encounter difficulty.

Of course I could be wrong. It's just a postulation at present to fit the data. Which is why I'm waiting on others to do them too. If we all get roughly the same pattern, there's probably some truth to be found there. On the other hand, if we get different results, then it falls on us to alter our hypothesis to be able explain why it is so (own findings and that of others).