Breaking from Steady

[Cunningcairn]

I have just loaded the new update and for the first time had a unit break from steady after a single combat. Is that something that can happen?

[rbodleyscott]

I have just loaded the new update and for the first time had a unit break from steady after a single combat. Is that something that can happen?

Not in a straight up fight.

It could happen

a) if they autodropped after a flank charge, and then double dropped from the post-combat cohesion test.
b) If a general "fell" and so they took 2 cohesion tests, and double-dropped on one of them.
c) If they autobroke from losses.

[Cunningcairn]

I have just loaded the new update and for the first time had a unit break from steady after a single combat. Is that something that can happen?

Not in a straight up fight.

It could happen

a) if they autodropped after a flank charge, and then double dropped from the post-combat cohesion test.
b) If a general "fell" and so they took 2 cohesion tests, and double-dropped on one of them.
c) If they autobroke from losses.

I suppose a) covers it. It was made to flee by an enemy cav pursuing another of my cav. It had already moved 3 squares when it contacted and then moved another 4 squares to catch my cav in the rear and break it. In the same game a cav of mine pursuing a routed enemy cav caught an enemy disrupted mf bow in the rear. Bw just stood as if it was no problem and my cav retired. There have been a number of these extreme events happening in a few of my games over the last few days. Some in my favour. I am not imagining this. Every so often I am experiencing sequences of unbelievable results one after another. This hasn't happened for a few months but it's definitely back.

[rbodleyscott]

I have just loaded the new update and for the first time had a unit break from steady after a single combat. Is that something that can happen?

Not in a straight up fight.

It could happen

a) if they autodropped after a flank charge, and then double dropped from the post-combat cohesion test.
b) If a general "fell" and so they took 2 cohesion tests, and double-dropped on one of them.
c) If they autobroke from losses.

I suppose a) covers it. It was made to flee by an enemy cav pursuing another of my cav. It had already moved 3 squares when it contacted and then moved another 4 squares to catch my cav in the rear and break it. In the same game a cav of mine pursuing a routed enemy cav caught an enemy disrupted mf bow in the rear. Bw just stood as if it was no problem and my cav retired. There have been a number of these extreme events happening in a few of my games over the last few days. Some in my favour. I am not imagining this. Every so often I am experiencing sequences of unbelievable results one after another. This hasn't happened for a few months but it's definitely back.

If you are relating this to a possible cause in the latest update, nothing has changed (in this or previous updates) that would alter the frequency of such events. Runs of uncommon events are themselves going to be rare, so of course you will only see them intermittently. Temporal proximity to an update (or any other change) does not necessarily imply causality, especially for rare sequences of events.

Every so often I am experiencing sequences of unbelievable results one after another.

That is the nature of randomness. Similar events come in clumps, not evenly spread. If they were evenly spread, they would not be random.

It also begs the question of what constitutes "unbelievable". In what way is a unit routing immediately on being charged in the flank unbelievable?

As has been said before, the more unlikely (and infrequent) results are there to represent the effects of unexpected battlefield events that cannot be modelled any other way. They are specifically intended to cause things to "not go according to plan". It is an intentional part of the game design. ("No plan survives first contact with the enemy" - to paraphrase Helmuth van Moltke).

[Cunningcairn]

Not in a straight up fight.

It could happen

a) if they autodropped after a flank charge, and then double dropped from the post-combat cohesion test.
b) If a general "fell" and so they took 2 cohesion tests, and double-dropped on one of them.
c) If they autobroke from losses.

I suppose a) covers it. It was made to flee by an enemy cav pursuing another of my cav. It had already moved 3 squares when it contacted and then moved another 4 squares to catch my cav in the rear and break it. In the same game a cav of mine pursuing a routed enemy cav caught an enemy disrupted mf bow in the rear. Bw just stood as if it was no problem and my cav retired. There have been a number of these extreme events happening in a few of my games over the last few days. Some in my favour. I am not imagining this. Every so often I am experiencing sequences of unbelievable results one after another. This hasn't happened for a few months but it's definitely back.

If you are relating this to a possible cause in the latest update, nothing has changed (in this or previous updates) that would alter the frequency of such events. Runs of uncommon events are themselves going to be rare, so of course you will only see them intermittently. Temporal proximity to an update (or any other change) does not necessarily imply causality, especially for rare sequences of events.

Every so often I am experiencing sequences of unbelievable results one after another.

That is the nature of randomness. Similar events come in clumps, not evenly spread. If they were evenly spread, they would not be random.

It also begs the question of what constitutes "unbelievable". In what way is a unit routing immediately on being charged in the flank unbelievable?

As has been said before, the more unlikely (and infrequent) results are there to represent the effects of unexpected battlefield events that cannot be modelled any other way. They are specifically intended to cause things to "not go according to plan". It is an intentional part of the game design. ("No plan survives first contact with the enemy" - to paraphrase Helmuth van Moltke).

Richard thank you for the explanation. I understand fully. I really don't want to sound like a stuck record but the point is that these results are not random. If no changes were made then there is something wrong with the randomness of results. In my games I can now predict when there will be sequence of "stretching the odds" results as they happen in batches. There is a long term cycle to them occurring in batches. What is unbelievable is not the fact that one "stretching the odds" result occurs but that so many occur within a few moves of each other in the same games. This is then followed by a series of many games where results are as would be expected or slightly against the odds.

[rbodleyscott]

As I say, random events do occur in clumps, they are not evenly spread. If they were evenly spread, with no clumps, they would not be random.

https://www.oreilly.com/ideas/the-inher ... randomness

https://en.wikipedia.org/wiki/Poisson_clumping

https://core.ac.uk/download/pdf/82644986.pdf

[MarkShot]

Isn't one of the basic laws of probability such that the 1000th flip of a coin has no relationship with flips 1-999?

Psychology teaches us that primate brains look for patterns even where there are none (and humans have been told). This has been shown experimentally, and rats do far better than humans in truly random situations.

But, of course, your mechanics do induce clumping:

* The nature of play is to find the weak spot and hit it hard

* You generate adjacent cohesion checks upon a break of a unit

Both of these result in clump-like behavior.

[MarkShot]

But dynamic morale to some extent negates randomness in human activities. In sports, this is known as being "hot". This is very well modeled in the game Virtual Pool by Celeris. An AI opponent will from time to time get hot, and sink a run of 4-10 balls; maybe even run the table.

You have modeled the downside of this when a general falls.

[Cunningcairn]

As I say, random events do occur in clumps, they are not evenly spread. If they were evenly spread, with no clumps, they would not be random.

https://www.oreilly.com/ideas/the-inher ... randomness

https://en.wikipedia.org/wiki/Poisson_clumping

https://core.ac.uk/download/pdf/82644986.pdf

Thanks for that Richard I was beginning to worry that I had required psychic abilities Poisson clumping does explain it although I never realised how predictable it can appear.

[Cunningcairn]

Isn't one of the basic laws of probability such that the 1000th flip of a coin has no relationship with flips 1-999?

Psychology teaches us that primate brains look for patterns even where there are none (and humans have been told). This has been shown experimentally, and rats do far better than humans in truly random situations.

But, of course, your mechanics do induce clumping:

* The nature of play is to find the weak spot and hit it hard

* You generate adjacent cohesion checks upon a break of a unit

Both of these result in clump-like behavior.

Mark I think you have hit the nail on the head. I was aware of clumping back in the day when I occasionally wasted my hard earned cash in the casinos. As you say the game mechanics themselves can be enhancing the clump like behaviour. I think another factor is that certain 6 to 1 results are quite dramatic and I tend to notice them more. The combination of the Poisson affect, the game mechanics, and dramatic 6 to 1 results have made me a bit testy and heading down the dark and lonely road to conspiracy theory. I do believe I am now cured In saying that can we ever be sure Cambridge Analytica aren't involved

[Waffenamt]

All-in-all that has been quite an edifying read. Thanks so much