Die rolls

[Dancing Bear]

Hi all
is anyone experiencing a high number of leader casualities since going to 1.07? I'm wondering if the security measures are causing the game to look up the same results repeatedly, so we are getting lot of repeat rolls for some actions? Or maybe, we have just been unlucky.

[Mardonius]

I do know that when the first edition of this game was put out the leader casualty role was 1 in 12 (random linear) vice 1 in 36 (bell curve, 2 rolls of 1 six sided die). Perhaps we have regressed to the original pattern.

[NeverMan]

Isn't this like the 5th thread on this topic?

[Dancing Bear]

Marshall can you comment on this, is leader casualty a 1 in 12 event or a 1 in 36?

ORIGINAL: Mardonius

I do know that when the first edition of this game was put out the leader casualty role was 1 in 12 (random linear) vice 1 in 36 (bell curve, 2 rolls of 1 six sided die). Perhaps we have regressed to the original pattern.

[ndrose]

There was a thread quite a while back on the 1-in-12 vs. 1-in-36 thing. That was fixed in an early version, but as Mardonius says, it could always get broken again....

More recently there was a thread about the casualties always landing on the senior leader; that's supposed to be fixed in 1.08, I think.

[Dancing Bear]

Just had another leader in one of our games get wounded today. This one game has had 3 leaders killed or wounded in a year (game time. i.e. 1805). This is the seocnd leader to be wounded or killed this month in this game. In another game we have had one leader killed/wounded in 3 months of game time. Does this seem right? I can't imagine we have had that many battles where leaders were involved during this time frame, even considered numerous seige battles.

[ndrose]

That does sound like a lot. The siege battles shouldn't count, either. If any leader casualties occur during a siege, that would itself be a bug, I think. (Or is it a deviation?)

And if the game's (correctly) not doing leader casualties in sieges, then the denominator is just the number of field combats, so the rate's higher.

Might be worth setting up a test game and running it numerous times to find the real rate.

[Dancing Bear]

That's a good point. In this case, the commander was wounded when the beseiged forces attacked the besiegers, which is still not a battle (a field, limited or trival) where a leader should be wounded.

One, maybe two leader casualities related to seiges, while the others have been chit exchange combats.

If there is a change, it might be related to 1.07, since all of the casualities have happened since Christmas (at least 3 and maybe all of them in February).

Anybody else having this problem?

[ndrose]

I set up a test case vs. the AI, with Napoleon vs. Mack, and replayed the battle a number of times. Each battle should count as 2 trials, since each side has a leader. First I tried it 20 times, and had no casualties. Then I tried a further 20, and had 1 casualty. So that's 1 in 80 trials.

Assuming a 1/12 probability of a casualty, there is less than a 1% chance of getting 1 casualty or fewer in 80 trials. But assuming a 1/36 probability, there's a 34% chance of 1 casualty or fewer. So in that case, it looks to me as if it's operating correctly.

Of course, that's not to say a bug might creep in under other circumstances.

The fact that you're getting casualties in sieges is a bug, certainly. There can be lots of sieges in a game, and if it's possible that 2 of your 3 casualties came that way it may indicate that that's the only bug. That is, it may be applying the right probability, but over too large a denominator.

[Dancing Bear]

That's excellent research, Ndrose.

The log was checked for the third battle and it was an open field combat, but we have had 2 leaders killed/wounded in open field and one is a seige battle since mid February 2010 (two months in game time).

Almost of all our seige battles have not involved leaders, so should not have increased the odds. I would say more than 50% of the battles (seige and non-seige) that involved leaders have resulted in a casualty. No casualities when the leader was in a seiging force that attacked a garrison, so it might be close to 100% of field battles with leaders or when a beseiged force attacked a stack with a leader.

[bresh]

Remember witch leaders are casulties if from a group is still not random, but should be fixed in 1.08.


Regards
Bresh

[ereiser]

I am also involved in the game that DB is in.  After checking the log (whew), there have been 16 open field combats with leaders, 7 of which involved leaders on both sides (and I think 8 were PBEM battles).  This means 23 checks for casualties.  In 23 checks, there have been 2 casualties, both KIA (1 against computer, 1 PBEM).
 
While definitely higher than average for both casualties and effect, the sampling size is insufficient to draw any conclusions, but it certainly bears further scrutiny.
 
The 3rd case, where the besieged attacked the besiegers, is definitely a bug.  There should not have been a casualty check (fortunately in this case the leader was only wounded for a month so no long-term harm).  That is the first time in the game that this set of circumstances occurred.
 
ER

[Marshall Ellis]

I just looked at this code and I am still doing a 1 in 36 chance of a casualty. There was a 2 in 12 chance at first release but this has been fixed BUT I will continue to monitor...

BTW: Bresh is correct! leader selection for casualty was not random until 1.08! It always chose the current leader assigned to the force!

[ndrose]

In 23 trials, some 50-odd percent of the time there would be no casualties, 34% of the time 1 casualty, and the remaining 13% of the time 2 or more casualties. So it is higher than average, but not especially weird.

2 KIA is definitely an outlier. It would be very time-consuming to try to check statistically whether the game is doing this right (you'd have to do hundreds, perhaps thousands, of trials). But maybe Marshall could double-check the code on this.

The siege combat casualties need to be fixed. Did anyone submit this to Mantis?

[Dancing Bear]

Well, maybe it is bad luck to have 3 leaders wounded/killed in 2 months of play. I was wondering if the pre-roll security feature for PBEM was properly clearing itself after each month? Could this be a possibility?

The leader wounded ina seige battle is a bug, and I have reported it to Mantis (althuogh I accidently locked it, so let me know if you can not read this, Marshal).

[Grognot]

Heh.  In a long-distance hotseat game with a friend of mine (trading saved-game files using Dropbox), the AI's Napoleon brought the pain to a sacrificial corps of mine (an Austrian corps that contained only militia, being used to cut supply lines). 

His forces wiped out my corps to a man... but Napoleon himself was killed.  In Feb 1806.   Doh!

[Marshall Ellis]

FYI, I'm running a 2 AI games and I haven't seen a killed leader yet in about 2 years of game play???

[Dancing Bear]

I was wondering if the pre-rolling security coding was adding modifiers to the leader casualty die rolls (same as a land combat), which might explain the inferred increase in incidence of leader casualties in our PBEM game.

[Marshall Ellis]

Modifiers? Such as what? I didn't think there were mods for casualties?

[Dancing Bear]

Your right there should not be any. If there were, then it would be a bug. I was looking for an explanation for why we had three leaders wounded/killed in 4 months, which seemed very unlucky.

One turned out to be because he was wounded in a seige battle (he was in the besieging force that was attacked by the beseigers). Leaders are not meant to be wounded in siege battes, and so is a bug (reported on Mantis, interestingly, the same battle caused the besiegers and beseiged to swith places, so it was unusual for two reasons).