Naval movement bug?

[friedrich2]

As Germany, I moved my fleet out from Kiel to Faeroes Gap. I eliminated all ships of speed 3 or less, so all I had was 4s, 5s, and 6s.

I doubt this is relevant, but it may be a trigger for the bug: I loaded a unit onto a transport that was sitting in the same hex. I moved out to Faeroes Gap, then picked up another unit, and was able to place my fleet only in box 1. I was expecting to be able to place them in box 2.

Beginning from Kiel, entering North Sea I should be down to 3 moves. In Faeroes Gap I should be down to 2 remaining: box 1, then box 2.

We're not doing Presence of the Enemy, and I don't know of any other rules that might be relevant. We're playing 1.4.5.

[Larry Smith]

I'm not to sure, but if you loaded the second unit onto the transport, that might be where the extra MP was used, otherwise, it shouldn't matter. Are you certain you eliminated the two speed-3 BB's Germany gets? I always stick them in another port just to keep them out of the way.

[BrianJH]

I think you are loosing 1 extra movement point because you have Italian Naval units also based in Kiel.

Check out rule 11.4.2

How far can units move?

A unit must stop moving when you have spent its entire movement
allowance or it has reached the limit of its range, whichever happens
first.
You spend 1 point of a unit’s range:
- for each sea area and port it moves into.
You spend 1 point of a unit’s movement allowance:
- for each sea area and port it moves into;
- if it starts the movement out of supply;
- if it starts the impulse in a port with naval units controlled by
another major power; and
- for each point of the (unmodified) search number of the section
you put the unit into.


But I did did discover a bug here too. Switching to the Italian Player and moving out the all of the Italian Navy to some other port(s), then, switching back to the German Player it will then allow you to move German naval units out from Kiel as if the Italians were never there, and then allows you to move to the 2 box in Faeroes Gap. I don't think, given reading of the rule above that should be allowed. As it still started the impulse with units from another major power in the same port.

[Larry Smith]

I've encountered that quirk myself. Anytime I had mixed nationality naval units in a port, I made a point of moving one of them out first - usually to cover convoys - before moving my strike force/blockade force out.
With so many rules to keep track of, like many others, I let the program remind me what was right and what was not, and thus, got used to that bug as being right.

[friedrich2]

That must be it, BrianJH. Thanks. I had no idea that rule existed.

On a side note, because I'm trying to play the game with better strategy, I have developed a calculator that aids me. I have a list of inputs to establish total available attacking resources versus total available defending resources to better gauge my likely outcome, to avoid all the ridiculous attacks I've had to double back on after realizing how catastrophic the outcome will certainly be.

Anyway, I am putting together one for invasions, so I finally established a calculation for notional units. When loading that file, skipping to a German invasion from Faeroes Gap to take Liverpool, for example, the notional unit is 5 strong. Attacking with the 7-factor militia on Scapa Flow, I get a result of 3.5:5. This is agreed upon by my notional calculator and MWiF. So 3.5:5 = 0.7:1.

Using the 2d10 table, I would expect the resultant die modifier (based solely on the combat ratio) to be 1.4 when using the formula 2*(ratio) or 1.143 when using the formula 4-2/(ratio). (I mention both because this is borderland.) But proceeding to the land combat selection and resolution screen, it is reporting that the base combat ratio modifier is +1.5, which implies that I have a combat ratio of 0.8:1 or 0.75:1. This seems like another bug. Or am I missing something?

[Courtenay]

ORIGINAL: friedrich2

Anyway, I am putting together one for invasions, so I finally established a calculation for notional units. When loading that file, skipping to a German invasion from Faeroes Gap to take Liverpool, for example, the notional unit is 5 strong. Attacking with the 7-factor militia on Scapa Flow, I get a result of 3.5:5. This is agreed upon by my notional calculator and MWiF. So 3.5:5 = 0.7:1.

Using the 2d10 table, I would expect the resultant die modifier (based solely on the combat ratio) to be 1.4 when using the formula 2*(ratio) or 1.143 when using the formula 4-2/(ratio). (I mention both because this is borderland.) But proceeding to the land combat selection and resolution screen, it is reporting that the base combat ratio modifier is +1.5, which implies that I have a combat ratio of 0.8:1 or 0.75:1. This seems like another bug. Or am I missing something?

You are missing something. The 2d10 die roll modifiers are not linear in the combat ratio for odds of less than 1:1.

First, if one does not use fractional odds, one has the following table:

1-6 or less: -10
1-5: -6
1-4: -4
1-3: -2
1-2: -0
2-3: +1
1-1: +2

Above 1:1, the 2d10 die roll modifier is simple: just 2 times the combat odds. At less than one-to-one, as you can see, it is a very strange function.

If one is using fractional odds, it becomes stranger.

I have never understood exactly how the fractional odds rule works for low odds attacks; I have have seen several different contradictory formulae. It would be nice if ADG gave an explicit algorithm, but to the best of my knowledge, they have not.

If anyone knows an offical algorithm, I would love to see it.

[friedrich2]

I'm telling you, that between 1 / 5 odds and 1 / 1 odds, the formula 4 - 2 / odds = the modifier.

4 - 2 / 0.2 = 4 - 10, matches -6 in the table
4 - 2 / 0.25 = 4 - 8, matches -4 in the table
4 - 2 / 0.33 = 4 - 6, matches -2 in the table
4 - 2 / 0.5 = 4 - 4, matches 0 in the table
4 - 2 / 0.67 = 4 - 3, matches +1 in the table
4 - 2 / 1, matches +2 (does match 2 * 1 = +2, along with the rest of the table)

You did mention specifically fractional odds. I'm not in the habit of making low-odds attacks so I haven't fully explored this but I would expect that if the odds are anywhere between 1/5 and 1, you should be able to use this formula to establish exactly how big the fractional would be.

But certainly, in this borderland, I have no explanation for why 0.7 is coming up as +1.5.

(FYI, this forum software is reading what I'm typing as a link, phone number, or email address. It isn't. It's a line of calculations. That's an algorithm I'd like to see. And now I have to retype everything? Instructions should include a warning to copy before posting.)

[paulderynck]

That's fine if there's no fractional portion.

With fractional odds the die roll modifier no longer has discrete levels of -2, -4, -6 etc. It is a continuous function. So we need to know when the die roll modifier will be -1 or -3 as well. This is simple.
1-6 or less: -10
4-23: -9
2-11: -8
4-21: -7
1-5: -6
2-9: -5
1-4: -4
2-7: -3
1-3: -2
2-5: -1
1-2: -0
2-3: +1
1-1: +2

The mathematically correct way to determine fractional odds for attacks lower than a DRM of zero is this: Take the amount needed for the next lowest odds ratio. Whatever portion over that is the numerator of the fractional amount. The denominator is the delta between the amount needed for the next higher odds level minus the amount needed for the next lowest odds ratio.

For example: 4 attacking 9. This is better than 2-5 odds so the die roll modifier is -1 with a fractional chance of being a zero. 2 to 5 odds against a defender of 9 would require 3.6 attack factors. The next level up is a 1-2 which requires 4.5 attack factors. The fractional is: (4-3.6)/(4.5-3.6) = .4/.9 = .444. (In other words, the attacker is 44.4% of the way between 2-5 odds and 1-2 odds.)

So the attack is a -1 with a 44.4% chance of being a zero.

Now the attack of 3.5 to 5. First off in RAW7 all point 5s and higher round up. So you really have 4 to 5. It is better than a 2-3. So a DRM of +1 at least with a chance for a +2. Fractional portion is (4-3.333)/(5-3.333) = .667/1.667 = .4. Should be a fractional of 40%. But it appears MWiF is doing something other than what is mathematically correct.

By rounding up the point 5s you get an extra "free" benefit on all attacks using fractional odds where the attacker has an extra point 5 or more (maybe due to river or fort hexsides). Really the point 5 should be retained and used to calculate the fractional part. (It will be this way in the next version of the rules.) So it should be: (3.5-3.333)/(5-3.333) = .167/1.667 = only a 10% chance for a +2.

[friedrich2]

Wow. That algorithm is ridiculously convoluted, but I shouldn't have much difficulty modeling it in my calculator. Thanks for the input.

So it should be ... only a 10% chance for a +2.

I guess the next question is why in MWiF 3.5:5 results in a +1.5 which implies a 50% chance for a +2.

[paulderynck]

The round-up is most of it. Perhaps MWiF doesn't have the fraction portion quite right.

[Courtenay]

Thank you. Yours is the best explanation I have seen.