Twilight of the Sun King is a rules set taking a minimalist approach to game mechanics. The core of the system is the morale check. If a unit fails a morale check, a hit is scored. Most units can take 2 or 3 hits before being removed from the game. Oddly, the game designers chose the sum of 2 average dice for the random element in a morale check. I suspect they wanted a limited range of outcomes yet they wanted the middle results to be most common. Given that there are only 3 results (Pass, Fail or Destroyed), it seems to me that this could have been done in a better way.
Analysis
If a morale check total is 8 or more, the unit passes. If the morale check is 7-5, the unit takes 1 hit. If the morale check is 4 or less, the unit is immediately removed from the game.
2 Average dice graph. Left column is the individual chance of each occurrence while the right is the sum in ascending order.
With the sum of 2 average dice, a 5 or 6 will come up 1/3 of the time. Similarly a 8 or 9 comes up 1/3 of the time. A 7 will come up about 27.78 percent of the time. The outliers (4 and 10) each come up 2.78 percent of the time. Thus, a unit passed a straight up morale check 36.11 percent (slightly ore than 1/3) of the time. They fail 63.89 (slightly less than 2/3) of the time. 2.78 percent is an automatic removal of the unit. Rolling 2D6 would be an obvious solution but there is a problem. The spread from pass to destruction (8 to 4) is only 4. If you roll 2D6 you would either have a spread of 6, which throws off the modifiers to morale checks, or you would have automatic destruction occur many more often that intended. A ‘4’ would occur 16.67 percent of the time which is 6 times more than intended.
The Change
Ignoring the outliers, we can generalize and say that a unit passes 1/3 of the time and fails 1/3 of the time. That neatly works on a single D6. A unit passes on a 5+ and fails on a 4-. We could say that a unit is destroyed on a 1 but that would also occur 6 times more than intended. My solution is to make a result of 0 an automatic destruction. So, using a single D6, a 5+ is a pass, a 1-4 is a fail and a 0 or less is a destruction. The observant will not that a unit will pass 2.78 percent less than before. But you should also note that a unit that has 0 or more positive modifiers cannot be automatically destroyed. That’s probably a fair trade. Also, while the spread for pass to destruction is 5 instead of 4, it will still likely not break the system.
Conclusion
It is clear to me that the designers of the game wanted units to fail morale checks often with only a small chance of a catastrophic failure. However, 2 average dice are not necessary for morale checks in the game. You can yield similar results with a single D6, which I hope most gamers can appreciate.
John's Wargame Page 
Interesting. I wasn’t aware of this rules set. Where can it be found?
Hey Ed,
Pike and Shot Society has the latest version. http://www.pikeandshotsociety.org/paypal-uk-us-stock
Steven’s Balagan has plenty of information on the game including an early iteration of the rules. http://balagan.info/wargaming/twilight-of-the-sun-king
When I get off my behind and actually play a game or three, I’ll give the rules a proper review.
You do stuff like I do in my spare time! Great analysis but are there a lot of dice roll modifiers (or maybe common dice roll modifiers) that might skew the percentages of a pass/fail moving from 2dAv to 1d6?
Hi Shaun,
I guess great minds think alike! 😉 In the rules as written, the modifiers only cover a spread of 4 points. Using the D6 modification, there is actually a spread of 5. Too many modifiers is likely not going to be an issue. That the check is changed to linear from a bell curve might be. Only testing will tell. The extra point in the spread from pass to destruction might alleviate any other issues.
John
Hi John,
A nice analysis which I have just found out about, a friend told me about it. I did the current set by the way so any ‘blame’ is mine. Although in the end I went with the 2 Average dice because the previous authors had used these and I was wary of changing too much in a system that works well and many like.
I think your analysis is good and I did something like this when developing the current version of the rules. But as others have suggested the modifiers should be taken into account. Notably I think the +1 for Rear Support. In normal circumstances we expect unit to have at least 1 unit giving rear support. Not having rear support is risky and players (and indeed real commanders) rarely risk being without it. Usual this is when forced to by perhaps being very outnumbered and not having enough units to spare to form a 2nd line or loses/the flow of the battle mean you can’t.
In effect this means that an average unit (Trained) needs a 7 to pass and can not get a ‘destruction’ role. I am not sure if you have the 1st printed version but in that you needed a score of 7 (2 Average dice) to pass. But this version used a -1 if you did NOT have rear support and I changed it to 8 and +1 as proved easier for players to remember in playtesting – every one is happy to forget a minus but keen to claim a plus.
In ‘normal’ circumstances only units with a -1 will be subject to the percentages you have worked on. This will typically be Raw or Small units. But of course as the battle develops there will be wide variations on this.
This does not of course mean that you analysis is ‘wrong’ and that you couldn’t use a single D6. You would of course just need different ‘target’ numbers and assume that in most cases units will have the +1 for rear support. I am tempted here to suggest my own numbers which I have worked out but I am really hoping that you will look at this and do another post. So I will keep quiet and hope that you will read this and comment again.
I will say though that I, and a friend who is working on an Ancients version, have thought about using different dice. I am working on a version covering the period 1620 – 1680 and that will certainly have different options on dice. My preferred option is 2 D6 but changing the target numbers, specifically the ‘destruction’ number. As you mention to keep it at 4 is far too much. If you use 2 as the target this would keep the same idea as we have now (i.e. Trained with rear support are safe in normal circumstances). But playtesting suggested that this was too low. The greater spread of results you get from using 2 D6 compared to 1 average dice mean that on aveeage you need more ‘minuses’ to get down to reasonable chances of ‘destruction’.
Therefore as a compromise I have gone with (perhaps pending you changing my mind) if you use 2 D6 then you still need 8+ to pass but you need 3 or less for destruction. One of the groups who did playtesting much preferred 2 D6 but the other didn’t think it made much difference. I intend this to be an ‘official’ option in the upcoming 1620-80 rules. I have also been considering using a single D6, my friend who is tinkering with an Ancients version favours using 1 dice.
We have been thinking more of a D6 plus some standard number, probably 3 or 4. So you would roll a D6 and add say a standard 3 then whatever modifiers you might have. While different to your idea it is essentially the same but just done in a slightly different way. We have not done much playtesting on this idea and I am a little wary of it. Using 2 dice, of any kind, averages things more than using a single dice. But of course, assuming you get the numbers right, overall the results should be very similar which ever way you go. My ‘gut feeling’ is that using 2 D6 will result in slightly more random results than 2 average dice. Using a single D6 (with or without a standard modifier) a little bit more random than both. But overall it won’t make that much difference to a game on statistically average dice rolls, if that ever happens in reality.
In summary I am hoping that this message might prompt you to look at the odds/dice issue but bearing in mind that normally a unit will have a +1 for rear support. Also hopefully looking at changing the ‘destruction’ target depending on what dice are used. I would like to offer different dice options and clearly D6 are the obvious choice. I would be very interested in what are the best options on this.
All the best and I am now off to have a good look around what looks like a great blog.
Hi Nick,
First off, the piece was less of a criticism of the system and more of a “I wonder if things could be done differently?” sort of post. 2 average dice are quite viable. The problem lies in the fact that most gamers don’t have them and they are not really readily available. You can’t cannibalize Monopoly to get them! 😀 That said, they are also easy enough to reproduce on an ordinary D6. My favorite way to do that is a ‘1’ counts as a ‘3’ while a ‘6’ counts as a ‘4’.
The single D6 and the average dice methods are similar in several respects. I laid out the D6 to have a spread of 7 outcomes with 0-6. That is the same as the average dice system with 4-10. Of course the extreme numbers are less likely to occur. With 2D6 you have 11 possible outcomes. A ‘3’ IS a nice compromise for 2D6. Again, the outlying numbers are less likely to occur than the middle numbers but a ‘3’ is about 3 times as likely to occur as a ‘4’ int he Average Dice system. A 1 point larger spread from pass to destruction will probably render the higher chance of rolling a unnoticeable.
I do want to give the rules a go with average dice just to see how they work. I’ll probably do a couple of games using these two alternate systems to see how they feel as well.
Ancients using this system is an interesting idea. Check out my very own “Throw me a ‘6’” rules on the “Old School Page” here. Most of the methodology I used for that game would probably convert well to modifiers in the form of grading units. Pikes vs Swordsman get +1 when they are full strength. Swordsmen get +1 on the second and subsequent round of melee. Etc. I only rise above +1 once with the rear attack.
John
Hi John,
I certainly didn’t take it as criticism and I didn’t take it as such. Also it must be said criticism is fine as long as it is constructive.
I know what you mean about the average dice. I used to have loads but could only find 1 when I first started on these rules and so had to use a D6 and the 1=3, 6 = 4 idea until I could track down more average dice. This is exactly why the rules need 1 or more alternatives on dice, preferably D6.
I think the differences between the various ideas are, or at least can be, relatively little with the right numbers. Plus of course what difference there is will be the same for both sides. I hope you do try them all and it will be interesting to see how it goes. Using two D6 should make the ‘combat’ a little more deadly. But when I have used D6 and the group that use D6 all the time this is difficult to notice.
In any case I look forward to hearing how you get on. As mentioned before I will certainly include two D6 with a 3 ‘destruction’ level in the future and would be open to some kind of single D6 set up as well.
On the ancients idea, yes I think the system could work well for it. I am not sure my friend has got far with it but it is certainly something I might have a go at in the future. That is unless your good self or some one else does it first. I think the ideas in ‘Throw me a 6’ could work well. I think the key thing is to give pluses/minuses to reflect specific advantages/disadvantages. So I think my friend and I are thinking of more targeted rules period wise, rather than a general purpose set. He is looking mainly at doing Romans vs Barbarian Warband battles (i.e. Gauls, Germans, Ancient Britons) first. Then going from there to do other sub periods – he loves Romans so I suspect we are talking Romans vs some other army type.
All the best,
Nick
Hi, I’m the author or the original version (written 1995; published 2001). I used 1d6 for morale. A score of less than 4 was a morale failure. The average dice thing was added later by other contributors. Personally I stick with my version and 1d6. You can check out the original version here … https://balagan.info/twilight-of-the-sun-king-2001-version