The first beta of the next release cycle for vasl-templates is now available.
The main new feature doesn't have anything to do with setting up VASL scenarios, but is so insanely cool, I couldn't
not do it after I thought of it. My opponent had
finished his regular moan about the dicebot pointed out a few somewhat idiosyncratic rolls, so I decided to add something to take a closer look at what was going on. You can now feed VASL log files into the program, and after much churning and grinding of gears, get something like this:
Across the top are distribution graphs of dr's and DR's, along with the expected results (in green) for easy comparison. The pie charts show the split of rolls between the players e.g. I rolled 59% of the DR's (283 of 483), while my opponent had the other 41%.
At the bottom is a plot of die rolls as the game progressed. You can change the window size for the moving average to smooth things out, or get the raw rolls e.g.
You can also filter by the type of roll e.g. IFT, MC, TH, etc., so if you're a lazy bugger like me and always press F1 or F2 to get a generic dr/DR, this might encourage you to generate the "proper" rolls.
If you analyzed multiple log files, they will be combined and reported on as a single group, but you can also drill down and look at each session individually:
The final graph is an experimental feature, and tries to answer the question: "how hot were my dice?" Below is the analysis for the second session:
As you can see from the averages (my 7.5 to the Other Guy's 6.4) and the time-plot, my dice sucked pretty hard. Hotness takes a closer look at the distribution. This metric doesn't work so well with small datasets, so if there are less than 100 rolls, the bars are shown in grey, but my 89 rolls is close, and for the purpose of this explanation, it doesn't really matter.
While I rolled slightly more 2's than expected (4 instead of the expected 2.5), I was a bit light on the 4's and 5's, and didn't roll any 3's at all. This makes my hotness score go down. I also rolled more 8's and 11's than expected, which also makes my score down, ending up at -1.8.
OTOH, the Other Guy rolled more 3's than expected, and a whole lot more 4's and 5's. He was also very light on the 9's and 10's, and didn't roll any 11's at all. This would normally score very well (about +2.8), but the fact that he rolled no 2's and a lot of 12's (4, instead of the expected 1.6) cancelled all that out, for a final score of +0.1. 2's and 12's are weighted quite heavily, because of the special rules that kick in when you roll these.
Again, this is an experimental feature, and will need some tuning, but that's the idea.
Finally, if all this wasn't enough, you can also download the raw data, bung it in a spreadsheet and perform whatever arcane incantations you want on it:
There is also a new field in the Settings dialog where you can enter your VASL user name. It's not essential, but if you do this, you will always be shown first in the analysis (which also means you will always be assigned the same color).
Please PM me if you'd like to take this version for a spin (Windows only), but even if you don't want to run it, if you have any log files that I can use for testing and tuning, that would be very much appreciated. Turns out, there aren't that many online.