MikeAP wrote:As a former officer I'm sure you're aware that a commander is responsible for everything a unit does, or fails to do...From the Civil War to World War Two, to modern day Afghanistan commanders (at all levels) have been fired, relieved, and rotated out for poor performance or lack of results.
For repeatedly and consistently bad results yes. But that is not the case in PzC. If you have a positive prediction, then on average you WILL have positive results. You might have a bad streak of luck yes, but that eventually the odds will even that out and you'll end up with a streak of luck. Eventually the average result will equal exactly the prediction. That's just how math works, even if it initially doesn't "feel" that way.
As for the commanders again... yes commanders are relieved every now and then, but usually because they are responsible for their bad results. not because they had bad luck and ran into circumstances that were impossible to predict. Unless they're relieved by politicians which not always do that because of performance reasons. if a commander makes a good plan based on the right information and then simply has bad luck, nobody can blame him and if he is a good commander such stuff will not happen consistently, but on average he will have a very positive result. if he continually has negative results, his plans are bad, because luck is then no longer an explanation.
Same with PzC, the probability of continually, over a long period of time, having your odds vary in the negative direction form the prediction, is very very low. And getting lower the longer that "streak" of bad luck holds. If you watch videos or AAR from some of the best players like deducter or kerensky or whoever you'll see that they tend to push the odds quite far in their favor before risking an attack because they want to push even the worst case result to the best possible outcome and make it nearly impossible to have a bad result. A good commander does exactly the same, he takes every possible precaution to avoid things that can go wrong. Sometimes that is not possible but you should always try to achieve it. Just because on average my result will be 3:1, I will not attack, because I know with a bit more preparation I can achieve a 3:0 prediction.
I'm not arguing against people who prefer the chess mode. That's just a matter of taste, and it seems a lot of people like that mode more. But I wouldn't say that it is "more realistic", since real battles simply have a variance in outcome, otherwise battles could simply be predicted easily and they cannot. D-Day worked, but were the casualties higher or lower than "predicted" by the responsible commanders? I'm guessing Omaha beach was a lot worse, while some of the others might have been better than predicted or equal to the prediction. But that just happens. Sometimes things go your way, sometimes they don't. And to prevent unnecessary losses all a commander can do is take every single precaution and optimization that he can beforehand, and then hope for the best and trust in his men. Not every bad outcome is human error or bad judgement. Sometimes the other guys are simply hitting better or are just at the right spot at the right time out of pure luck. But if you planned really well, you can avoid that most of the time.
Deducter e.g. rarely ever attacks if he has a 3:1 result prediction, simply because he thinks the 1 is already too much. he doesn't want to lose ANYONE. And he knows a 1 usually means, I might lose nobody, I might lose 3. If I opt for 0 I might lose worst case 1 or 2. That's better, so I'll opt for that. And since I started adopting that as well and stopped attacking when I simply didn't like the odds (and at one point you start disliking odds like 4:1 because you don
t like the 1 at the end) my results have vastly improved. With random chance enabled I think you have to always see the range of possible values and the prediction as the middle ground. That way you'll think like "Ok when I have 1 loss on average I might end up with 0 or 3 losses, let's more that more to the left.
Where I do agree is that maybe it would make it easier for the player if he didn't just see the average result but also the best case and worst case results. That might make it more in-your-face obvious what you have to expect. but even then you wouldn't exactly know the spread, the variance... I don't even know exactly what the exact outcome spead looks like, weather they are evenly distrubuted or have a normal distribution... I'm guessing an even distribution since they are using simple dice rolls. MAYBE it would make sense to offer an option where instead a normal distribution is chosen, making extreme results off the expected one, more unlikely. But I guess with dice-chess and chess mode, most people are already happy with how it works now.
And I also seem to remember devs commenting that giving the player soo many numbers in the prediction would destroy the game's simple intuitive UI that doesn't overload newbies with lots of statistics but simply gives a normal prediction.
I personally, as a developer, find it very intersting everytime I see such a discussion, how some players percieve random chance very differently from others. You wouldn't believe how many discussions I had with game designers about whether or not to "tweak" a dice roll, so it ends up being "more like I would expect" instead of simply being a mathematically correct dice.
Really difficult topic
