I am not sure, but I think FFG does mostly in house playtesting
Those NDAs tend to suck the fun out of it tho
This is not true. We have several local area players who have been play testers who do not work for FFG.
Formula for ships and squadrons
I came up with this about 6 months ago. It might help:
https://community.fantasyflightgames.com/topic/184846-squadrons-reverse-engineered/
Warning: I do biostatistics as part of my job. Without further ado, I present formulae that replicate all generic fighters perfectly:
The REG Procedure
Model: MODEL1
Dependent Variable: POINTS
Number of Observations Read 16 Number of Observations Used 16
Analysis of Variance Source DF Sum of
Squares Mean
Square F Value Pr > F Model 14 115.52528 8.25181 3.71 0.3883 Error 1 2.22472 2.22472 Corrected Total 15 117.75000
Squares Mean
Square F Value Pr > F Model 14 115.52528 8.25181 3.71 0.3883 Error 1 2.22472 2.22472 Corrected Total 15 117.75000
Root MSE 1.49155 R-Square 0.9811 Dependent Mean 12.87500 Adj R-Sq 0.7166 Coeff Var 11.58484
Note: Model is not full rank. Least-squares solutions for the parameters are not unique. Some statistics will be misleading. A reported DF of 0 or B means that the estimate is biased.
Note: The following parameters have been set to 0, since the variables are a linear combination of other variables as shown.
SQD_R = 0 HEAVY = Intercept - 2 * SHIP_R - SHIP_B - SHIP_K + BOMB + GRIT BRACE = 0 SCATTER = 0
Parameter Estimates Variable DF Parameter
Estimate Standard
Error t Value Pr > |t| Intercept B 4.40009 8.60416 0.51 0.6991 HIT 1 0.38792 1.35397 0.29 0.8224 SPD 1 -0.17130 0.77938 -0.22 0.8623 SQD_R 0 0 . . . SQD_B 1 -0.17640 1.72423 -0.10 0.9351 SQD_K 1 -1.10445 1.59336 -0.69 0.6141 SHIP_R B 6.75929 11.53744 0.59 0.6626 SHIP_B B 4.05299 3.74133 1.08 0.4746 SHIP_K B 4.92237 5.81316 0.85 0.5527 BOMB B -0.62782 4.07053 -0.15 0.9026 INTEL 1 2.92080 3.60085 0.81 0.5661 ROGUE 1 5.09965 2.60633 1.96 0.3008 GRIT B 1.71386 2.98422 0.57 0.6681 COUNTER 1 1.15896 1.18055 0.98 0.5059 ESCORT 1 1.74835 2.55205 0.69 0.6176 HEAVY 0 0 . . . BRACE 0 0 . . . SCATTER 0 0 . . . SWARM 1 0.11246 5.24712 0.02 0.9864
Estimate Standard
Error t Value Pr > |t| Intercept B 4.40009 8.60416 0.51 0.6991 HIT 1 0.38792 1.35397 0.29 0.8224 SPD 1 -0.17130 0.77938 -0.22 0.8623 SQD_R 0 0 . . . SQD_B 1 -0.17640 1.72423 -0.10 0.9351 SQD_K 1 -1.10445 1.59336 -0.69 0.6141 SHIP_R B 6.75929 11.53744 0.59 0.6626 SHIP_B B 4.05299 3.74133 1.08 0.4746 SHIP_K B 4.92237 5.81316 0.85 0.5527 BOMB B -0.62782 4.07053 -0.15 0.9026 INTEL 1 2.92080 3.60085 0.81 0.5661 ROGUE 1 5.09965 2.60633 1.96 0.3008 GRIT B 1.71386 2.98422 0.57 0.6681 COUNTER 1 1.15896 1.18055 0.98 0.5059 ESCORT 1 1.74835 2.55205 0.69 0.6176 HEAVY 0 0 . . . BRACE 0 0 . . . SCATTER 0 0 . . . SWARM 1 0.11246 5.24712 0.02 0.9864
FACTION NAME POINTS predicted resid IMP LN 8 8 -7.11E-15 IMP IN 11 11 2.13E-14 IMP SA 9 9 2.13E-14 IMP ADV 12 12 -7.11E-15 IMP YV-666 15 15 1.42E-14 IMP Aggress 16 16 2.84E-14 IMP Firespr 18 18 1.42E-14 IMP Jumpmas 12 12 2.84E-14 REBEL x-wing 13 13 -3.55E-15 REBEL a-wing 11 11 -3.55E-15 REBEL b-wing 14 14 -3.55E-15 REBEL y-wing 10 10 -3.55E-15 REBEL YT-1300 13 13 -1.78E-15 REBEL YT-2400 16 16 0 REBEL HWK-290 12 12 -3.55E-15 REBEL H6 16 16 -3.55E-15
Go regression! If you anyone wants to tabulate ship dice for me in a similar fashion, I'll do a similar analysis for ships 
.... my brain hurts
I would love to make use of this, but I'm afraid I have no idea how to parse what's going on >.<
Edited by FoaS.... my brain hurts
I would love to make use of this, but I'm afraid I have no idea how to parse what's going on >.<
Me neither! SHHHHH! It's a secret!
Don't toy with my emotions, man!
Sorry, to clarify, I added too many degrees of freedom for too few data points, resulting in an exact fit equation that is rather meaningless (sorry, bad statistics joke here... I guess the statistical sarcasm didn't get through!). If I reran the analysis without the keywords, it wouldn't take much sleuthing to determine their worth based on the predicted costs after that. In fact, here it is!
Points = -2.77 + 1.27[HP] + .36 [sPD] + 1.39[AS blue] + .87 [AS black] + 2.76[sHIP RED] + 3.58[sHIP BLUE] + 3.17[sHIP BLACK].
Of course, this completely ignores all key words, so take this with many grains of salt.
That's totally fair - My last message was meant to be entirely in jest.
I take it in your first data-dump the keywords contributed to the result?
That's totally fair - My last message was meant to be entirely in jest.
I take it in your first data-dump the keywords contributed to the result?
They sure did: that many DF also results in each speed and blue die increase contributing negative points. Oops.
I think I've seen in some threads that FFG had access to the specs for ships that the movies that were used in the movies. If someone wants to spend a crapload of time studying the movies, we could probably extrapolate from the known relative value. Nice idea, but I'm not going to do it.