With the US Regionals season being done I wrote a blog post on the current state of the Regionals data that I have been compiling.

Good article. Having tracked your data I am not really surprised in the overall.

Truly, an epic job of data analysis.

There are a lot of things I would love to poke at if we had a complete dataset of the lists, but I'm not sufficiently obsessive to go through all the forum posts to try to sort it out. I am contemplating going in & flagging the clean lists (regionals that have fully detailed lists) and figure out a way to get all the data into Sheets. (trying to preserve each ship & its upgrades, for example)

What I would poke if I had time/data:

ship upgrades - how prevalent are various types of upgrades (Dial control, accuracy, movement, etc - would need to define upgrade types)

Fighter types - Fighters/Bombers/Multirole, Rogue & Intel, Speed, Aces (but with 34 separate fighter types, that would be a huge amount of data entry)

I did some number crunching on my lunch breaks at work on the vassal Galactic Civil War tournament lists. I should compare/contrast where I can.

Again, thanks for all your hard work!

Thanks for putting in the effort to gather all of this data and publicize it, Shmitty. Huge service to the community.

Truly, an epic job of data analysis.

There are a lot of things I would love to poke at if we had a complete dataset of the lists, but I'm not sufficiently obsessive to go through all the forum posts to try to sort it out. I am contemplating going in & flagging the clean lists (regionals that have fully detailed lists) and figure out a way to get all the data into Sheets. (trying to preserve each ship & its upgrades, for example)

What I would poke if I had time/data:

ship upgrades - how prevalent are various types of upgrades (Dial control, accuracy, movement, etc - would need to define upgrade types)

Fighter types - Fighters/Bombers/Multirole, Rogue & Intel, Speed, Aces (but with 34 separate fighter types, that would be a huge amount of data entry)

I did some number crunching on my lunch breaks at work on the vassal Galactic Civil War tournament lists. I should compare/contrast where I can.

Again, thanks for all your hard work!

Yeah, I'd like to get in more details, but the data entry becomes pretty crazy and you almost end up with information overload. I do really like your idea on upgrade categories. That would definitely be something to work on for the next time.

Thanks for the work. Often cited on IFF, it's great and terrible all at once to be able to point at data when making statements and decisions.

I'd be tempted to write some text-manipulation formulas in excel to scrape the data into the spreadsheet. But if people didn't use a standard notation, that would fail.

Still, maybe a worthwhile project for my lunchbreaks, assuming nothings on fire that day. I can test the logic on the new Vassal tournament lists now appearing. (If only people would all use the same fleet builder)

Good analysis and really good work with the statistics. I really appreciated all you've done for the entire season.

I've also been reading the material closely for the entire season. I'm a math person who likes to crunch the numbers and see what they tell me.

Here are a few thoughts section by section:

Data quality and usage: I really appreciate that you've opened with this section and cautioned everyone on how robust you think the data is.

Rebels versus Imperials: I personally enjoy Rebels, and I'm glad that the data drifted back toward people taking them more closely. A 55-45 split is perfectly fine. If it gets much more than that, then there might be something to it in the game balance.

Admirals and Archetypes, o my! My sense is that there is a very strong correlation between Admrial and the most prevalent archetype. I know from how I design fleets that the first question I'm starting from is "How am I going to handle a DeMSU?" One ship defines the meta. You're either in the 20% of the people taking it, or you're thinking about how you're going to play against it. The Rhymerball partly does this by attempting to extend the hull and by having squadrons that counter Demolisher before it flies away. For the Rebels, Rieekan emerged as a clear counter and Ackbar had popularity as soon as Wave-4 hit the stores in the pre-Clonisher days.

What I will say here as someone who has worked in game design: balance is choice. When the choices are lopsided, that is a fairly clear sign of underlying balance issues since people don't just decide not to take those commanders for "no reasons." I've wanted to get Mon Mothma to work for quite a long time. I just like the idea of evading all the time. Certainly those rerolls got to help a little against DeMSU black hit/crits. The problem is that they just don't help enough. As a good player, can I sit down and get some decent performance out of a fleet with her? Absolutely. Do I think I'd do as well as the Rieekan fleet? Absolutely not. Flotillas might help.

Average fleet size: The average I think reflects that Squadron-heavy lists tend to run 2-3 (usually 3 activations). If you weren't taking heavy squadrons, then you wanted 4 activations minimum.

Average number of squadrons: Lots of Rhymerballs means lots of 8-10 squadron totals in the meta. What would be nice in the future will be to track specific squadron totals, or perhaps point values. For example, how did 4 squadrons perform against those Rhymerballs?

Average number of deployments: Correlates with the average number of squadrons. Squadrons represent deployments. The fact that some people can pick Superior Positions and still win shows that deployment isn't always all that it is cracked up to be. In certain fleets, it is absolutely and critically meaningful. MC30s and Demolisher are very much helped by advantages in deployment.

Ship prevalence: I'm also curious about the MC30. I agree its a great ship. It may just need us to discover a list that will help it thrive.

Upgrades: We definitely want to look at this more comprehensively next year. Flotillas and the Wave-4 ships are about to unload a lot of new upgrade cards that will undoubtedly affect all of the ships and upgrade cards we currently have.

Here's where it gets interesting: Flotillas come out in a month. We get two new commanders, some new concepts entirely. All of those will affect the current meta considerably. As with any new release, we'll probably see them in numbers for a while. Most fleets can benefit from at least one. Whether we settle at a point where fleets can go Flotilla-less is a good question. The placement is probably going to be negligible, since you could just take a couple of squadrons for the same cost and accomplish the same thing. The activation advantage could be considerable.

So i was totally planning on turning in info from the tacoma regional to you but then you were there. I really hope you do this again for regionals this november-(whenever they end).

Ill gladly do anything i can. I understand this takes up some time but store championships would be fun to see as well.

Cant thank you enough for starting this project.

What does the archetype "imperial wedges" encompass? Two ISD-IIs?

What does the archetype "imperial wedges" encompass? Two ISD-IIs?

2X ISDs, 3x ISDs, I-V-V, 3x VSD. Without a significant bomber force

Fleets trying to use the big Imperial front arcs.

What does the archetype "imperial wedges" encompass? Two ISD-IIs?

2X ISDs, 3x ISDs, I-V-V, 3x VSD. Without a significant bomber force

Fleets trying to use the big Imperial front arcs.

Interesting. If 7 (5 percent) out of the 143 fleets used some variation of those compositions, then there is a strong chance that I was the only one included in this sample size to field 2 ISD-IIs.

What do you (or anyone else) suspect accounts for this data? Is it the low number of activations?

What do you (or anyone else) suspect accounts for this data? Is it the low number of activations?

I'm pretty sure it's about activations. I think that the only way 2ISD fllet can counter DeMSU without going to 5 activations is to bring a bomber force (Maybe there is somesing with 4-activations 2ISD builds though)

Very interesting. Excellent work, Shmitty.

I did some number crunching and it looks like the numbers are so skewed towards DeMSU/Rhymerball due to their overwhelming advantage in the early Regionals (April/early May). Once people started to bring counters, the number dropped significantly and the numer of Rebel Swarms rose.

Here is Top-4 data per archetype per tournament week (I only included tournaments with Top-4 (or Top-3) being available.

04/09 4/23 4/30 05/07 5/14 5/28 06/04 06/11

Rhymerball 50.00% 43.75% 40.00% 31.25% 20.83% 23.21% 21.67% 20.90%

DeMSU 25.00% 31.25% 30.00% 25.00% 20.83% 17.86% 16.67% 19.40%

Rebel Gunline 0.00% 0.00% 0.00% 3.13% 6.25% 8.93% 8.33% 8.96%

Rebel Carriers 12.50% 6.25% 5.00% 9.38% 14.58% 16.07% 18.33% 16.42%

Generic Rebels 0.00% 0.00% 0.00% 0.00% 2.08% 1.79% 1.67% 1.49%

Generic Imps 0.00% 6.25% 10.00% 12.50% 8.33% 7.14% 6.67% 5.97%

Rebel Swarm 12.50% 6.25% 10.00% 12.50% 18.75% 17.86% 18.33% 19.40%

Imperials Wedges 0.00% 6.25% 5.00% 3.13% 4.17% 3.57% 3.33% 2.99%

Imperial Swarm 0.00% 0.00% 0.00% 3.13% 4.17% 3.57% 5.00% 4.48%

PT109: Good observation!