In case you didn't see the thread: https://community.fantasyflightgames.com/topic/218901-stele-open-x-wing-broadcasts-are-up-on-youtube/#entry2196685

https://www.youtube.com/playlist?list=PLCYXyCs3ctMy5_dVhHwAkYQfaj3CDdRhI2016 Brother's Grim (Selden, NY) X-Wing Regional.

Here you go friend.

If you need a volunteer proof-reader, I'm willing to take a look. My background is sciences, software algorithms, and statistics, but I suspect the latter two coupled with the actual game knowledge might prove enough to be able to follow along, once you reach that point.

No, I don't think this is a 'sneak peak'.

Nor do I think it's going to be 'fun'.

Yes, I do know just what I just signed up for. I've helped with two dissertations and a thesis in my time. It's drudge work, but I should manage at least once through.

Just being helpful.

The lit survey for the new topic has been skinny in one direction and fat in another. There's a great deal of information about product evaluation in general, as well as its application in particular fields. There's a fair amount of evaluation of user interfaces in computer game design, as well. But there's no evidence at all that anyone is even thinking of evaluation of games, let alone using repeatable, reasonably objective quantitative methods to improve mechanisms within board games.

Stochastic combat models have been around for decades. The corresponding literature survey should be relatively fat, with quite a bit of material coming from the Naval Postgraduate School.

I once did a project based on warhammer for a stats class but doing it for your whole dissertation is cool! I'm sure this sort of thing is potentially applicable to a lot of things. I'm interested in your approach - I would tend to approach it like MJ based on probability and working up.

Well, both approaches definitely use probability. Any evaluation needs to use probability to determine how much damage ships output and receive, including a stochastic combat model that Vorpal's approach is taking.

The divergence is that I'm using those probabilities to calculate ships' attrition rate coefficients, and then using those coefficients to evaluate what a balanced ship cost is. The evaluation criteria for balanced costs, in my case, is derived from the analytical solution of an adaptation of Lanchester's original differential equations, modified to account for a small discrete number of combat units.

Vorpal is using probability as a tool to generate a stochastic combat model, in effect simulating an entire battle of X-wing (I think). This approach is decades old, but hasn't been applied specifically to X-wing yet. In general, applying combat models and evaluation criteria to game theory and design is a particular application of pre-existing theory and models. However, as Vorpal pointed out, the literature covering such applications is extremely sparse.

Vorpal's evaluation criteria will be based off an entirely different field of study rather than the traditional combat approaches such as Operations Theory coming out of the Naval Postgraduate School. Such an approach could be a novel contribution to the field of study and is why the topic could be worthy of PhD consideration. I still think that the literature survey is incomplete without referencing the other tools that have been previously been published to solve the problem he is attacking, but ultimately that is up to his committee.

I was in the good graces of all my committee members for my PhD, and I can't imagine having to start over several years in, so I wish the best of luck to you Vorpal!

I discovered after 1.5 years into my PhD research that the approach I had developed worked fine on synthetic, but not on real data. Luckily I could re-use most of it with minor modifications, so it was nothing like starting over, but I can imagine what you're going through. I started losing interest in my subject in my last (fifth) year, but managed to complete my dissertation, and then left the academic world.

Like Bob suggested, I would ensure that your work will be sufficiently novel. It sounds to me like you will be using Monte Carlo techniques? My doubts are two-fold at this point:

-a limitation to "jousting" is of limited use in my opinion, but the positioning aspect is much more difficult to model

-is X-Wing alone sufficient to cover a field as wide as "evaluation"? I could see it as one case study for testing a methodology, but would probably add other cases that the committee members can more easily relate to.

Anyways, I wish you the best of luck!

While I'm a biologist, I did a chunk of my Ph.D. on probability theory and modeling of biological systems. You and MJ are both welcome to PM me if you'd like some independent eyes on any given problem, I have a soft spot for mathhammer..

Signal transduction, protein folding, thermodynamics, ecosystems, or something I haven't thought of?

@MJ: I was thinking of your literature question in terms of the need for quantitative approaches in product evaluation. The *methods* section of my lit review covers various testing and IRT models after looking at the history of development of combat modeling and simulation (as well as your model, specifically, addressing it mostly in the same terms I did upthread, although at somewhat greater length).

The novelty is satisfied because I'm not saying this kind of evaluation is new, or that the methods I'm using are new--but that they can and should be applied in the relatively novel context of building board games and other tabletop games.

@Other PhDs expressing sympathy and/or offering help: I'm actually looking for an outside committee member with a thorough understanding of X-wing. So be careful what you volunteer for...!

While I'm a biologist, I did a chunk of my Ph.D. on probability theory and modeling of biological systems. You and MJ are both welcome to PM me if you'd like some independent eyes on any given problem, I have a soft spot for mathhammer..

Signal transduction, protein folding, thermodynamics, ecosystems, or something I haven't thought of?

It was simulations of mutation accumulation (hence the probabilistic element) in spermatogonial stem cells. Using cellular models to make predictions of new incidence of disease at population levels.

Cool! I'm very much not familiar with that aspect of biology, but I will cease derailing Vorpal's thread further.