Jump to content
SaltMaster 5000

Carolina Krayts is the best X-Wing podcast

Recommended Posts

31 minutes ago, GreenDragoon said:

The cumulative distribution is great when asking how deep you have to bid to win against initiative X lists.

But TBH, I wonder how useful average initiative is anyway. E.g. the 5A versions of 5/5/4/3/1 and 5/3/3/3/3 play and perform extremely differently and largely so because of their initiative, but their average initiative are 3.6 and 3.4, respectively. I always calculate the average initiative myself, and plotting it vs bid is trivial, but I've stopped mentioning it because I don't see the value.

I think interesting are the distribution of initiatives over faction or total field (histogram or in %), how frequent high initiative ships are per list (i5+i6 or i6 only), and how much these lists bid. I don't really care about low i bids.

I mean, I just think I'd like both regular and cumulative distributions.

As to average, yeah, that seems useless.  The list I'm likely flying tomorrow at a small kit tournament is Anakin, Jag, and Sinker, so average Init 4, but my bidding is based on 6.

As to high-init ship count, that also might be misleading.  5A with 5/5/4/3/1 and Boba/Guri are going to show up in the same category.  A potentially interesting (and maybe not practical to calculate) breakdown might be by total points invested in high-init (5-6) ships.  You'd probably want to have buckets, something like 0-66, 67-100, 101-134, and 135-200.  I suppose it could be divided into thirds, but breaking the middle third into sixths seems handy, to differentiate between a single expensive ace (a Delta7 Anakin, for example) which might be in the 70s somewhere, from a list with two moderately priced aces.

Share this post


Link to post
Share on other sites
Posted (edited)
10 hours ago, theBitterFig said:

A potentially interesting (and maybe not practical to calculate) breakdown might be by total points invested in high-init (5-6) ships.  You'd probably want to have buckets, something like 0-66, 67-100, 101-134, and 135-200. 

That would be great, but (at least for me) not possible as I don't get the information of points per ship and I don't want to rebuild it.

Actually, I just checked and I should be able to get it. Hey @Brunas, I have a feature request xD

e: I should be able to do it, but I don't quite get what the json looks like to pick the correct points per ship instead of the points per list.

edit2: Ok I got it, it's just a small change @Brunas: add the extra header on line 19, add an extra line on line 48 (points_ship = pilot['points']) and 183 ('points_ship' : points_ship,)

That opens some possibilities. For example back in 1.0 at the very beginning of my data digging, I made a graph like this to show how different ships were usually built

zem9Flj.png

Edited by GreenDragoon

Share this post


Link to post
Share on other sites
4 hours ago, GreenDragoon said:

That would be great, but (at least for me) not possible as I don't get the information of points per ship and I don't want to rebuild it.

Actually, I just checked and I should be able to get it. Hey @Brunas, I have a feature request xD

e: I should be able to do it, but I don't quite get what the json looks like to pick the correct points per ship instead of the points per list.

edit2: Ok I got it, it's just a small change @Brunas: add the extra header on line 19, add an extra line on line 48 (points_ship = pilot['points']) and 183 ('points_ship' : points_ship,)

That opens some possibilities. For example back in 1.0 at the very beginning of my data digging, I made a graph like this to show how different ships were usually built

zem9Flj.png

Haha, I'll do this later if someone reminds me!

Share this post


Link to post
Share on other sites

Found some extremely well performing lists. Now looking at 5509 ships in 1402 squads from 34 HS trials where 300 squads made cut at a 21.4% conversion rate.

Besides the usual with TIE swarm (20/60, 33%), there are different versions of rebel beef that have different performance. XXYUhas 39% at 13/33, XYBU is at 35% (8/23).
But the better version is a Leia-free BBYX at 80% (4/5), or YYYU with Leia at 58% (7/12) or XXXU at 56% (5/9). Those are not all, but the ones mentioned already add up to 37/82 with an overall 45% conversion.
The rest are XBBU (6/22, 27%), XXBU (5/17, 29%), XABU (3/11, 27%), YBBU (3/12, 25%), YYAAX (2/2, 100%), XXXYY (2/3, 67%), XXAU (2/5, 40%). I didn't look beyond the 50th list. That all adds up to 60/154 at 39% overall conversion.
These rebel beef lists made up 1 in 5 lists that made cut, and just picking this list gives you double the chance of making cut in the first place!

Empire does extremely well with Vader/Soontir and 1 or 2 bombers. The version with 3 ships has 71% conversion (5/7), the one with 4 ships is at 46% (6/13).

Honorable mention goes to 2xDelta 2xTorrent at 28% (8/29), sad are XXAA (8/48, 17%) and 5A (6/30, 20%).

 

Lists that are simply bad:

  • VaderSoontirReaper (12%, 3/24)
  • Belbullab + 6 Vultures (9%, 2/22)
  • SF+VN+Ups (14%, 3/21)
  • XXY (16%, 3/19)
  • 2 Firesprays (5%, 1/19)
  • 3 Aethersprites (6%, 1/18)
  • XXU (12%, 2/17)
  • and funnily enough YYYYY at 14% with 2/14

there are several at around 17%-20%, but the ones above are both quite frequent and bad

Share this post


Link to post
Share on other sites
Posted (edited)
5 minutes ago, GreenDragoon said:

Lists that are simply bad:

  • VaderSoontirReaper (12%, 3/24)
  • Belbullab + 6 Vultures (9%, 2/22)
  • SF+VN+Ups (14%, 3/21)
  • XXY (16%, 3/19)
  • 2 Firesprays (5%, 1/19)
  • 3 Aethersprites (6%, 1/18)
  • XXU (12%, 2/17)
  • and funnily enough YYYYY at 14% with 2/14

Thanks for sharing the #s. Nothing unexpected. 

One thing though, I dunno if id call something with 12%+ conversion rate bad.  Maybe normal?

XXU, Vader Soontir Vermiel, Kylo QD Tav, there all solid just not leia beef or Imperial Ace jouster tier.

Edited by Boom Owl

Share this post


Link to post
Share on other sites
1 minute ago, Boom Owl said:

One thing though, I dunno if id call something with 12%+ conversion rate bad.  Maybe normal?

They are compared to others that make cut. But of course they at least made cut, so you have a point.

Share this post


Link to post
Share on other sites
Posted (edited)
10 minutes ago, GreenDragoon said:
  • Belbullab + 6 Vultures (9%, 2/22)

X2I2XZA.png

And then to face the reality that even if Vultures dropped by 25% 20% to 16 points, you would still only be able to fit... Belbullab + 7.

Edited by svelok
math hard

Share this post


Link to post
Share on other sites
Posted (edited)
44 minutes ago, GreenDragoon said:

They are compared to others that make cut. But of course they at least made cut, so you have a point.

Sure “bad” in comparison. Decent lists that mostly any logical player can make cut with.

Defining logical is more difficult.

Edited by Boom Owl

Share this post


Link to post
Share on other sites
Posted (edited)

With the usual disclaimer that I look at ship combinations here and not the actual list. Based on the mentioned dataset.

There were 295 different combinations being played. 67 of them (23%) made up 71% of all lists that were played (995/1402). Out of those 67, 58 made cut. So 9 lists are popular but don't make cut:

  • Infiltrator + 5 Vultures (14 times!)
  • 3 T65s (9 times)
  • XYAU (?!, 8 times)
  • ScumFalcon + 2Fangs (8 times)
  • Falcon+Xwing (8 times)
  • 3 T70s (8 times)
  • Belbullab + 5 vultures (7 times)
  • 5 TIE bombers (6 times)
  • Falcon+Xwing+Uwing (6 times)

The remaining 929 lists forming 58 combinations made up 232/300 lists in the cut (77%). Filtering for lists that made cut at least twice goes down to 51 lists. Splitting them up means that the forum is not the ideal place anymore, but whatever: Here is the relevant Wave 3 Meta per faction. All percentages are conversion rates! If you have a plan for those, you are golden. Alternatively you can look at the subset in bold, as they are with 25% a bit above the 21% conversion rate, and prepare for those 25 lists that perform well

 

Galactic Empire (56/178, 31%)

  • 6 TIE Fighters (20/60, 33%)
  • Vader + Miniswarm (5/21, 24%), where the miniswarm is
    • 4 TIEs (3/15, 20%)
    • 3 TIEs + a striker (2/6, 33%)
  • TIE Salad (3/12, 25%)
  • Vader, Soontir, X (28/85, 32%), where X is
    • a reaper (3/24, 12%)
    • a striker (7/21, 33%)
    • 2 bombers (6/13, 46%)
    • another advanced (5/11, 45%)
    • 3 TIEs (2/9, 22%)
    • 1 bomber (5/7, 71%) (!!!)

Vader and Soontir are educated guesses, I didn't actually check that it's them.

First Order (19/72, 26%)

  • TripAces with SF, VN, FO at 4/15 (27%)
  • Silencer + 2 Upsilons (7/20, 35%)
  • Silencer, SF, Upsilon (3/21, 14%)
  • 3TIE + Upsilon (5/16, 31%) where the 3 TIEs are
    • 2 FO + 1SF (2/7, 29%)
    • 1FO + 2 SF (3/9, 33%)

Separatists (9/62, 15%)

  • 2 Infiltrators (2/12, 17%)
  • 2 Infiltrators + Belbullab (2/11, 18%)
  • Command+Swarm (5/39, 13%), where the command ship + swarm are
    • Belbullab + 6 (2/22, 9%)
    • Infiltrator + 4 (3/17, 18%)

Scum&Villainy (10/38, 26%)

  • 4 Fangs (2/11, 18%)
  • 2 Fangs + Starviper (4/9, 44%)
  • Firespray + Fang + Escapecraft (2/12, 17%)
  • Firespray + Starviper (2/6, 33%)

Rebel Alliance (78/270)

  • 4 Ship Rebel Beef (55/163, 34%) with the combination
    • XXXU (5/9, 56%)
    • XXYU (13/33, 39%)
    • XYYU (5/24, 21%)
    • XXBU (5/17, 29%)
    • YYYU (7/12, 58%)
    • YBBU (3/12, 25%)
    • XABU (3/11, 27%)
  • Falcon + 2 at 12/44 (27%) with the 2:
    • XY (6/18, 33%)
    • XX (3/14, 21%)
    • XA (3/12, 25%)
  • Trench Run (3 Rebels) at 5/36 (14%)
    • XXY (3/19, 16%)
    • XXU (2/17, 12%)
  • XXXX (4/13, 31%)
  • YYYYY (2/14, 14%)

Resistance (25/118, 21%)

  • 5 A (6/30, 20%) (sadface)
  • FourShip Resistance is at 12/63 (19%) with 3 combinations of X and A:
    • XXAA (8/48, 17%)
    • XAAA (2/9, 22%)
    • XXXA (2/6, 33%) (I'm happy about this one as I've long argued that it's a good combination)
  • XA Starfortress (2/12, 17%)
  • XXA Reistance Aces (5/13, 38%)

Galactic Republic (26/106, 26%)

  • 2 Jedi + X (15/61, 25%)
    • 1 ARC (7/32, 22%)
    • 2 Torrents (8/29, 28%)
  • 1 Delta + 2 ARCs (2/13, 15%)
  • Obi + 3 ARCs (7/18, 39%)
  • Delta + 2 ARCs + Torrent (2/8, 25%)
  • 2 ARCs + 4 Torrents (2/6, 33%)
Edited by GreenDragoon

Share this post


Link to post
Share on other sites
Posted (edited)
6 hours ago, GreenDragoon said:

Sorry for the triple post, but one more thought:

It looks to me like a good list can be reduced to one of two things:

  1. High initiative
  2. High hitpoints

Both neutralize the relevancy of maneuver choice in their own way.

I dunno about neutralize maneuver choice, those things still require correct choices. But 100% agree if your taking low init without high hp your almost definitely trying harder than you have to. Iden + i5 tactics counts as hp.

A big part of this seems related to Soontir and Bwings being randomly added to the format for no reason.

Edited by Boom Owl

Share this post


Link to post
Share on other sites
3 hours ago, Boom Owl said:

I dunno about neutralize maneuver choice, those things still require correct choices. But 100% agree if your taking low init without high hp your almost definitely trying harder than you have to. Iden + i5 tactics counts as hp.

A big part of this seems related to Soontir and Bwings being randomly added to the format for no reason.

What we need is a b-wing with soontir’s 1.0 pilot ability. 

Share this post


Link to post
Share on other sites
Posted (edited)
3 hours ago, Boom Owl said:

I dunno about neutralize maneuver choice, those things still require correct choices

Yes, but to a lower degree than ships without either of these characteristics.

My reason is that high HP forgive mistakes more, and high initiative is 'super high skill' full information arc dodging - again very forgiving. And forgiving means to me that maneuvers matter less, as the consequences are smaller.

Both characteristics have a relatively low skill floor.

 

Ah, I forgot this:

Bwings seem to be surprisingly inconsequential, whereas Soontir and Starvipers basically changed the faction overnight.

Edited by GreenDragoon

Share this post


Link to post
Share on other sites
2 hours ago, GreenDragoon said:

Yes, but to a lower degree than ships without either of these characteristics.

My reason is that high HP forgive mistakes more, and high initiative is 'super high skill' full information arc dodging - again very forgiving. And forgiving means to me that maneuvers matter less, as the consequences are smaller

And both avoid getting initiative killed.

Is initiative killing the enemy?

Share this post


Link to post
Share on other sites

So having played my last HS Trial of the season yesterday, I realised on the way home that despite being someone who generally prefers Hyperspace, I'm really looking forward to pivoting back to Extended again for a while. Say what you will about the current state of either meta, it's still pretty refreshing to have two different formats to play around in.

Share this post


Link to post
Share on other sites
Posted (edited)
6 hours ago, svelok said:

That tournament Jesus Christ. Ben was so disappointed in me and my mov...

 

Ah you think round timer is your ally? You merely adopted the clock. I was born in it, molded by it. I didn't see a complete game until I was already a man, by then it was nothing to me but regen!

Edited by Brunas

Share this post


Link to post
Share on other sites
Posted (edited)
On 5/11/2019 at 1:09 PM, GreenDragoon said:

They are compared to others that make cut. But of course they at least made cut, so you have a point.

So, are there 1400 players in 30 Hyperspace trials or 1400 lists entered (and then a lot more players)?   I'm trying to nail down what 300 players in the cut means in terms of percent of players that make the cut period.  That should give a kind of baseline for what a player's chance is of making the cut generally (assuming everything else is equal --which I know it isn't), and that then would give you a number that is a break point for expected number of times any list should make the cut to be "average."

Edited by AlexW

Share this post


Link to post
Share on other sites
Posted (edited)
3 hours ago, AlexW said:

So, are there 1400 players in 30 Hyperspace trials or 1400 lists entered (and then a lot more players)? 

1402 lists entered, but not a lot more players. But yes, the actual conversion rate is lower than the given 21%. I can get you the real number

e: that was quick. 34 trials. Total number of lists is 1460, entered lists are 1402. Cut is 300. So actual conversion rate is 20.5%. If skill and list did not matter, you had a 1/5 chance to make cut.

 

As for cut numbers and if anyone thinks 20% sounds high: every top8 cut in a tournament with 40 or fewer, and every top16 in a trial with 80 or fewer pulls the number above 20%.
Now, FFG recommends a top16 for 77 or more players, and there were just 3 out of the 34 trials. One had a top4 cut. All the others were top8. Basic structure says that only more than 40 players should get a top8, but the (*for trials mandatory) advanced structure dictates top8 for 13-76 players - that's a pretty large range and gives you conversion between 10-60%!

Edited by GreenDragoon

Share this post


Link to post
Share on other sites
Posted (edited)
2 hours ago, Biophysical said:

https://starfightermafia.blogspot.com/2019/05/article-11-starfighter-mafia-goes-to.html

This is what happens when a guy who mostly plays Imperials makes a Rebel squad.

Final table only to get kind of diced in the last game?  Confirmed and sad.  :(

What happens when a guy who plays mostly Resistance makes a Republic squad is he get dumpstered all day.  Also sad.

Edited by gennataos

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...