Thanks!

That gives me enough data to use New York in the Top 8 as well.

The only thing - my list is currently characterized as "Imperial swarm" in NY tournament (although I prefer to call it "ISD swarm") and "Generic Imperials" in Boston (9th place).

Thanks!

That gives me enough data to use New York in the Top 8 as well.

The only thing - my list is currently characterized as "Imperial swarm" in NY tournament (although I prefer to call it "ISD swarm") and "Generic Imperials" in Boston (9th place).

Oops, I will get them classified the same. I just cut down on the number of categories as it was getting a little busy, so not sure I will add in an ISD Swarm.

Thanks!

That gives me enough data to use New York in the Top 8 as well.

The only thing - my list is currently characterized as "Imperial swarm" in NY tournament (although I prefer to call it "ISD swarm") and "Generic Imperials" in Boston (9th place).

Oops, I will get them classified the same. I just cut down on the number of categories as it was getting a little busy, so not sure I will add in an ISD Swarm.

I agree, currently it's kinda of a one-off fleet (that uses activation advantage to go second and position heavy-hitters to threaten multiple ships next round), so "Imperial swarm" may be a better category. (Or even "Imperial Wedges" as Raiders are still wedge-shaped )

Thanks!

That gives me enough data to use New York in the Top 8 as well.

The only thing - my list is currently characterized as "Imperial swarm" in NY tournament (although I prefer to call it "ISD swarm") and "Generic Imperials" in Boston (9th place).

Oops, I will get them classified the same. I just cut down on the number of categories as it was getting a little busy, so not sure I will add in an ISD Swarm.

I agree, currently it's kinda of a one-off fleet (that uses activation advantage to go second and position heavy-hitters to threaten multiple ships next round), so "Imperial swarm" may be a better category. (Or even "Imperial Wedges" as Raiders are still wedge-shaped )

Death by Pizza.

Death by Pizza.

I finished 11th at Sheffield with this fleet:

Author: ungentlemanly

Faction: Rebel Alliance
Points: 392/400

Commander: Mon Mothma

Assault Objective: Most Wanted
Defense Objective: Fire Lanes
Navigation Objective: Intel Sweep

[ flagship ] CR90 Corvette A (44 points)
- Mon Mothma ( 30 points)
- Turbolaser Reroute Circuits ( 7 points)

CR90 Corvette A (44 points)
- Turbolaser Reroute Circuits ( 7 points)

CR90 Corvette A (44 points)
- Turbolaser Reroute Circuits ( 7 points)

MC30c Torpedo Frigate (63 points)
- Admonition ( 8 points)
- Ordnance Experts ( 4 points)
- Assault Proton Torpedoes ( 5 points)

MC30c Torpedo Frigate (63 points)
- Foresight ( 8 points)
- Ordnance Experts ( 4 points)
- Assault Proton Torpedoes ( 5 points)

3 A-Wing Squadrons ( 33 points)
1 YT-2400 ( 16 points)

The Store owner said he was going to publish the top 8 list, when he does I will share the link with you.

Some additional MI regional info. Was 16th, so not good but had fun, even my blow out loss to Ben #4, can I take an assist for propelling him up the rankings . Lists are all mid to low place, so probably not the most desired but here ya go.

16th

Imperial 400pts

ISD2 w/ Motti, Support Officer, Gunnery Team, Tractor Beam, Leading Shots, XI7 Turbolasers - 171

ISD1 w/ Relentless, Wulf Yularen, Sensor Teams, Tractor Beams - 131

Raider2 w/ Isard, Tractor Beam, SW7 Ion Cannons - 62

FIGHTERS - 36

TIE Advanced x1

TIE Fighter x3

Game 1 vs Austin (placed middle of the pack I believe)

AFb w/ Riekien (ECM, XI7, Gun Teams, Eng Officer)????

MC30 Torpedo w/ Foresight, Ordinance Experts, XI-7 Turbolasers, Assault Proton Torpedos

Neb-b Support w/ Salvation, Slaved Turrets

Neb-b Support w/ Slaved Turrets

Jan Ors, Xwingx3

Game 2 vs Jeff (placed in the lower 1/2)

ISD2 w/ Motti, Relentless, Gunnery teams, ECMs, Leading Shots?, XI7 Turbolasers

VSD1 w/ Ordinance Experts, Assault Proton Torpedoes x2

Aggressor x2, Firespray x2

-------

Commentary:

Not the best environment for playing around with ideas, but excellent in identifying flaws in build and play. Basic goal was to use the 2 Star Destroyers as a battering ram w/ the ISD1 tanking and tractor beams to mess with opposition speed. Raider there for activation, additional tractor beam.

The tractor beams worked great, messed with all my opponents plans. Raider should have been bare bones 1 w/ tractor beam... Isard didn't result in anything useful and I tried being to aggressive with it. My main downfall was not engineering quick enough and being to aggressive, Motti ISDs can take a beating but are not invincible.

Two games I wish there was a 7th turn a the game ended with a star destroyer about to run down an enemy ship thanks to continues tractor beams. The last I got tabled, but Ben's remaining 2 ships were burning wrecks.

Great fun was had though, and 16 still got me the acrylics, which was the goal.

Thanks for the additional lists!

I was also sent some information for a Regional in Vienna, Austria and am working to get that added as well.

We now know the winners of 17 different Regional tournaments. If someone wants to help with a bit of analysis, you could split the tournaments in to small, medium, and large categories and see if that impacts who is winning them overall.

Rebels are winning with high activation fleets.

Imperials are winning with lots of squadrons.

The latter is absolute fact. The Rhymerball dominance is very telling. I wonder how well that dominance will hold in wave 3. Rebels are getting some very nice anti-Rhymer synergy from Bright Hope (which flips the bird to single die bombers) and Toryn Farr (essentially flight controllers for everyone). I'm not sure the data is really showing the former. While Rebel swarm lists make up an equal portion of winners as a combination of Rebel carriers and Rebel gunlines, they also make up the highest portion of the bottom quarter of all lists. I think it's more fair to say that Rebels are winning with Rieekan, who is far and away the most common thing about top performing Rebels.

Rebels are winning with high activation fleets.

Imperials are winning with lots of squadrons.

The latter is absolute fact. The Rhymerball dominance is very telling. I wonder how well that dominance will hold in wave 3. Rebels are getting some very nice anti-Rhymer synergy from Bright Hope (which flips the bird to single die bombers) and Toryn Farr (essentially flight controllers for everyone). I'm not sure the data is really showing the former. While Rebel swarm lists make up an equal portion of winners as a combination of Rebel carriers and Rebel gunlines, they also make up the highest portion of the bottom quarter of all lists. I think it's more fair to say that Rebels are winning with Rieekan, who is far and away the most common thing about top performing Rebels.

This.

I think a closer look at the actual ships shows that the Rebel archetypes with Rieekan have been all over the board. I'll probably comb through the data at a later date as we get more information from additional regionals.

We do want to keep player psychology in mind when interpreting the data. Players love to win, and they also gravitate toward builds/units that are easier. Not everyone, since there is always the guy who wants to be counter-cultural or who really wants to challenge himself to break the molds. We need more of these people, no doubt. The majority, however, will slowly gravitate toward what seems to be working, and what seems to be the easiest. The overall trendlines run in this direction. Demolisher and Rhymerball are both pretty userfriendly, and they both cause quite a few headaches for an opponent. The more problems you can present to your opponent, generally, the more likely you are to induce mistakes in his play. Playing against them requires lists that are well thought out.

As for swarm versus non-swarm, the average activation count has held pretty steady around 3.6 for all players. Overall, it pushes 3.6 for the winners. The Rebels are higher at 3.8. There's only one winning Rebel list that has 5 activation. My sense is that some of the better Rebel ships are cheaper and work better in lower quantities. Or for that matter, the Rebels have three small base ships to choose from, so it isn't surprising that their lists would naturally have smaller cheaper ships and therefore more activations.

Yeah, I wrote those statements after looking at the graphs a bit. The Rebels seem to show a preference towards a higher activation count, but I was wrong to state that's how they are winning.

I just added a scatter plot of tournament size vs the points per round it takes to win. Nothing really remarkable about it, other than showing once there are about 16 players it pretty much takes 9 points per round to win.

I'd love additional suggestions on graphs or number crunching that anyone would like to see. I have just been adding new bits as it occurs to me.

Colorado 4th place was a triple vic list w/ 8x YV-666 I believe.

I just added a scatter plot of tournament size vs the points per round it takes to win. Nothing really remarkable about it, other than showing once there are about 16 players it pretty much takes 9 points per round to win.

I don't believe that's true. You need to analyze the points difference between the winner and the second place for this statement, i think. The correct one would be "the winners are the lists that tend to do 9/1 or better against other lists"

I just added a scatter plot of tournament size vs the points per round it takes to win. Nothing really remarkable about it, other than showing once there are about 16 players it pretty much takes 9 points per round to win.

I don't believe that's true. You need to analyze the points difference between the winner and the second place for this statement, i think. The correct one would be "the winners are the lists that tend to do 9/1 or better against other lists"

Very true. I think there is plenty of data to say that I need to choose my words better when posting on here.

I just added a scatter plot of tournament size vs the points per round it takes to win. Nothing really remarkable about it, other than showing once there are about 16 players it pretty much takes 9 points per round to win.

I don't believe that's true. You need to analyze the points difference between the winner and the second place for this statement, i think. The correct one would be "the winners are the lists that tend to do 9/1 or better against other lists"

Very true. I think there is plenty of data to say that I need to choose my words better when posting on here.

The same applies to me as well. I didn't mean to sound harsh, sorry for that.

Not harsh at all. I'm always getting on my student's cases about drawing conclusions that the data doesn't support. It's good to get that reminder myself.\

I'd like to get a good discussion going about the trends that are showing in the data and how they might be useful. Not sure if that will be here or something that I put together for my blog. Something like a round-table blog post. A moderated podcast discussion would also be interesting.

I'd like some opinions on this: So far I've included in the Top 8 set of data 2 tournaments where I only have the top 7 fleet lists as it seemed useful. I now have 3 tournaments where I know 3 of the Top 4. Should include those in the Top 4 data set?

It's already imperfect. Why not?

One this hat is hard to plot is he amount of skilled players to new players and such there are.

How good are these players who are winning with outlier lists compared to their opponents? How are their match ups? Are they clubbing baby seals in 2 of 3 games? Are they just completely dominating the field?

Colorado 4th place was a triple vic list w/ 8x YV-666 I believe.

That's awesome. That's about as against the grain as you can go.

One this hat is hard to plot is he amount of skilled players to new players and such there are.

How good are these players who are winning with outlier lists compared to their opponents? How are their match ups? Are they clubbing baby seals in 2 of 3 games? Are they just completely dominating the field?

Realistically, there will be a lot of luck in the current format. If we are playing 3 swiss rounds, ignoring the issue of byes, we would have even with just 16 players:

Round 1: 8 winners @ 1-0

Round 2: 4 winners @ 2-0

Round 3: 2 winners @ 3-0

So even then, the winner should be whomever of those two had the easier matchups to get there. The most common way to win a three round tournament (openly copping to having done this myself) is play someone totally unprepared or new in game one and obliterate them. In game 2, you get an average skill player who also blew up a new person, and you blow them up. Then in round three, you already have a massive lead on points and just need to hold on.

On the other hand, if the other undefeated guy played strong games against two very good opponents and won them narrowly, you will win the tournament. The assumption that all players are of equal skill is the problem... even though you may have never played the other guy and he could have beaten you head to head!

Likewise, FFG's belief in points as a sorting mechanism rather than record produces weird outcomes. Consider: 9-1, 1-9, 9-1 vs. 6-4, 6-4, 6-4. The former player is rated more highly in a points system, despite having a worse record. Maybe they are a better player, or maybe the 2nd player just had much tougher matchups.

Realistically we need many more swiss rounds before can accurately declare winners. I would suggest at regionals/store champs/etc. your best players cluster in the top, but that it doesn't tell you much more than "the top x were probably the x best" unless they all played each other, where x depends on just how big the tournament is.

One this hat is hard to plot is he amount of skilled players to new players and such there are.

How good are these players who are winning with outlier lists compared to their opponents?

When we have full lists, we can examine the variance in scores, so you could maybe tell which regionals had more uneven matches. Otherwise, we'd need match by match results. Even then, experienced players lose 0-10 sometimes.

I'm considering crunching the Galactic civil war results. They should be a more uniform skill group, but also they have other wierdness- uniques & faction balance for example.

Realistically, there will be a lot of luck in the current format. If we are playing 3 swiss rounds, ignoring the issue of byes, we would have even with just 16 players:

Round 1: 8 winners @ 1-0

Round 2: 4 winners @ 2-0

Round 3: 2 winners @ 3-0

So even then, the winner should be whomever of those two had the easier matchups to get there. The most common way to win a three round tournament (openly copping to having done this myself) is play someone totally unprepared or new in game one and obliterate them. In game 2, you get an average skill player who also blew up a new person, and you blow them up. Then in round three, you already have a massive lead on points and just need to hold on.

On the other hand, if the other undefeated guy played strong games against two very good opponents and won them narrowly, you will win the tournament. The assumption that all players are of equal skill is the problem... even though you may have never played the other guy and he could have beaten you head to head!

This is a good analysis, however it misses one variable: the rankings after the second round. If you have two matchups like this, you'll be at the first table and will be playing the second strongest person _at that time_. If you have 4 2-0 players with similar scores, then not doing a decisive win puts you at risk of being overtaken by winner on table 2 that does 8-2 or 9-1. So while there is some amount of luck involved (and I personally agree that another round (or a cut to a top 2, which would be a better and more realistic option) would be nice), it is less luck then your analysis tend to indicate.

Which is partly why I find the Top 4 data more interesting than the data about the winners.

If there is a single question I am trying to answer in gathering this data it would be "If we compare the fleets at the top tables to the total fleets that show up what trends appear?"

This data won't tell a person what will win a tournament, but might instruct them as to the types of fleets that they should be prepared to face.

I definitely second this idea about results being hard to verify a true winner. There should be some kind of Strength of Schedule modifier in there or additional rounds. Especially as this game grows and the Regionals get even larger, you're going to see a lot more ties for first place. Then it'll come down to who battered their earlier opponents the hardest, which is where Strength of Schedule will make a huge difference. Right now it's Head-to-head (which I support, but with a three-way or more tie, it doesn't matter) then MoV. So if I got two newer opponents rounds one and two and club them into adorable little baby seal ties (I actually had a high school teacher with a baby seal skin bowtie), then have a tough round three game but still tie for first place, I'm going to beat someone who played more experienced opponents.

I mean, consider the Cambridge regional. I went 460 MoV round one, lost round two by about thirty or forty points which dropped me all the way to table four, then won round three 9-1 with another high MoV. The person who beat me round two ended up at table one with a tougher matchup, so I managed to tie for first place because I had an easier matchup that final round. Now he won with head-to-head, which is the way it should be structured, but if someone had also tied us, I would have won because HtH doesn't matter with a three-way tie, unless you've beaten everyone tied with you. It's also logical to imagine in a larger tournament, I could have lost round two and then tied someone for first who I hadn't played who had a tougher schedule, but now I win because I had two easier opponents and a higher MoV.

But more than three rounds is tough. You either have to start very early and go very late, or break it up over two days, and that will cause a huge drop in attendance for Regionals and Store Champs. I mean, I'm always a fan of starting early (11:00 is too late in my opinion, when you have 6:45 of gameplay with no breaks factored in). If you started at nine, you would be done with round three around five, and that's with an hour for lunch. Then you can do a top four/eight/sixteen cut and play a final, fourth round and wrap the whole thing up by eight. You can even announce pairings for the cut, start the fourth round and hand out prizes to anyone below the cut so that they can leave if they want. Hell, you could even trim a good fifteen minutes from the round time, start at ten and finish by seven with four rounds and a thirty minute lunch.

So if I got two newer opponents rounds one and two and club them into adorable little baby seal ties (I actually had a high school teacher with a baby seal skin bowtie), then have a tough round three game but still tie for first place, I'm going to beat someone who played more experienced opponents.

I believe this scenario is very unlikely (For that your inexperienced opponent in round 2 should also somehow get a baby seal tie from his round 1 match up). Getting less experienced opponent in round 3 after a close game or a narrow loss in round 2 is a more realistic scenario (and it does create complications that you mentioned). I personally think that the best option is to do 3 rounds of swiss with a top 2 cut afterwards (This way It's only one game round after the 3 rounds of swiss and it is only for the top 2 players). Doing one more swiss round for top 4 or 8 will not solve a problem as you may again have people tied and would need to decide on MOV)