A Reader Responds

I got a great email from reader Simon Wright about my article on smoothing out activations in the Song of engine (from March 2011 … wow, was it that long ago?). Rather than keep the conversation in email, I thought I would post his original email, and my response. Simon wrote:

Hi Dale,

I've been reading your blog for the past few months, which I greatly enjoy. I've got some ideas I'd like to share. A year or so back you made a few posts about altering some of the Song of Blades and Heroes mechanics to allow for 'smoother' gameplay. In other words, tweaking the game slightly to avoid frustrating situations where the majority of a player's Warband (especially poor Quality models) do absolutely nothing for several turns (this can happen even when rolling only one die for activation, as an average Quality model still has a 50% chance of not getting an Action). While I (like many players) personally love the risk/reward concept behind the SoBH engine, like you I feel that it needs to somewhat less random so as not to lead to frustration due to randomization that's out of a player's control.

So here's my idea (well two, actually).

1) When a model activates, it can always perform one action. However, before this 'default' action is performed, the player can choose to 'gamble' for up to two more actions. To gamble, the player takes a Quality check on three dice. On two successes, the model gains a second action. On three successes, the model gains a third action. On two or more failures, the model gains no bonus actions, and after performing its default action, the player's turn ends.

2) Now I remember you were talking about how the turnover mechanic punishes the rest of a player's Warband, and not the model that caused the turnover. Well I was thinking, what about if the turnover just ends the model's activation, and not the player's? So when gambling for more actions as per above, if two failures are rolled, the model cannot perform any actions on that activation (not even the default action). Bear in mind that if the player wants to gamble for more actions, they would need to do so before they perform the model's default action.

I haven't done a whole lot of playtesting with the second idea. From what I have seen so far of the first idea though, it does the job of allowing low Quality models (like that Q5+ Troll that only ever seems to stand around and then suddenly bolt off the battlefield during a morale check) to actually do something without requiring constant Group Moves/Orders from a nearby Leader. This may make large hordes of low quality models too powerful though, and I'm working on coming up with something to fix this if it becomes a real issue. In the end this might change the game completely and require limitations on the amount of models in a Warband.

But that's it. I'm interested in hearing your opinion and criticism. Let me know what you think Dale, and keep blogging!

Kind regards from a new reader,

Simon

So, Simon brings up a really good point, that something really needs to be done to make lower Quality warbands more viable. It has been discussed a little bit on the Song of Blades and Heroes forum on Yahoo, but it seems like most people poo-poo the idea, or simply say "yeah, that is why I don't use low Quality warbands".

Let's Look at the Math

I can imagine some people has just skipped to the next blog …

Suppose you decide to always roll two dice, without fail, for every activation attempt with every figure in your warband. The chance of a turnover, by Quality, is as follows:

Quality	Chance to Turn Over
2	3%
3	11%
4	25%
5	44%
6	69%

As you can see, the odds do not go up linearly. It gets worse when you consider the number of times you roll for each quality level.

Quality	Chance 1 Can Act	Chance 2 Can Act	Chance 4 Can Act	Chance 6 Can Act
2	97%	95%	89%	84%
3	89%	79%	62%	49%
4	75%	56%	32%	18%
5	56%	31%	10%	3%
6	31%	9%	1%	0%

Hopefully my math is not off that badly … just a lot of rounding in there.

You can see that larger, low quality warbands do not stand much of a chance of getting all of their models activated (at least with two or more actions) and smaller, high quality warbands really are not gambling all that much. I strongly suspect that the points system does not take these percentages into account, but in the end would it really matter? If you could buy 23 Quality 6 figures for every one Quality 2 figure (representing the ratio of turnover), you likely could not activate them enough to make those numbers count.

Option #1

Looking at Simon's option number one he is basically granting one action for free, but making it statistically harder to get a second or third action. I actually like that. With one free action Quality 6 archers will pin cushion a Quality 2 force. Hmmm … maybe that swings the pendulum too far the other way against Quality.

The problem is that the mechanism gets away from a core design concept of Andrea's that not everyone can act, even if they get to roll for activation. Even if you roll only one die, failure may not mean a turnover, but it does mean you cannot act. It seems like we need to keep that concept in.

What you need to do is describe the effect you would like to see, then develop the formula that creates those effects. So, if I were doing this, I would say that I want at least a 33% chance of performing one action, and the rest goes down from there. In order to accomplish this, I could roll one die of one color (let's say red for this example) and up to two dice of another color. If you add +1 to the red die's score you have increased the chance of the first success, but if the person gambles with the other dice, then so be it; they are at the normal Quality roll. Don't think it should be only 33% chance for one success? Fine, add +2. In all cases, a natural '1' always fails, even with the modifiers.

Option #2

The concept of applying the failure to the one that failed, rather than the rest of the warband, definitely changes the odds of the game because more people will mostly gamble with three dice for everything. Think about how many games where you automatically choose to roll three dice with the last figure in the warband for the turn. Why not? Even though you have increased the change of a turnover, who cares, because everyone else has moved.

The idea I had was to mark all units that "turned over" and they would forfeit their chance of activation the following turn. I think it is simple and elegant, but rather marker heavy, for those that don't like junk on the table. (I use a circle punched out of grass green colored foam sheet, so it tends to blend in pretty well, but not too much that you miss them.)

Well, I may have to try all of this out with a game. Give me a break from working on Mesoamerican Saga or playing BattleLore online.

Thanks again to Simon Wright for writing and the kind words. This is the kind of thought-provoking feedback that I enjoy.