Hello, my name is Parker. And my name is Ned. This is The Force Unlimited, yet another Star Wars Unlimited Podcast. This is episode number 21, we're recording on August 18th, and we've got two juicy segments to get to, so we're just going to jump in. Ned, tell us about the match. We just played. So the match was good versus evil in a certain point of view. It was GI blue against Han-toe blue.
And I think that the summary of this is best conveyed by the Space Balls quote, evil will win because good is dumb. So Parker- I mean, shots fired, but you're not wrong. Yeah, Parker really misplayed. Game 1 was kind of back and forth. Parker had me a little bit on the back foot, and then there was a misplay around the seventh sister and cards lining up. And then I just, with GI, once you get that little fingernail in, you can really pry your way into a lot of damage by the midgame.
And so the game ends fast. Yeah, yeah. I should have, you know, I mean, we play fast because every minute we're playing is an episode minute, we're not recording. I don't have a lot of practice with the lines into GI. I think I should figure that out in anticipation of our appearance on late night gaming, since I think you're going to be expected to rep that. And I think it's just going to be a me versus Grand Inquisitor party that night. But I just don't have a lot of practice with those lines.
And yeah, they weren't obvious misplays, but they were one of those. Nope, that was the wrong line. That was the good line in the abstract, the wrong line staring down the cards. And I, you know, I just didn't have answers for the seventh sister Grand Inquisitor combo, even in the sideboard. Even in the sideboard, there's just doing the three damage to my already damaged units is unpleasant.
And you know, I, I, I, I, Han-Tu Blue is, is taking down like, showdowns, taking down a, it's a good deck. It's a good deck. I, I don't know if it's a great deck, and we're two weeks out from Pax West, and I have to kind of commit to a main deck. And Han-Tu is on the very, very short list.
But the, the line that I share at locals, I am increasingly sold on the idea that Han-Tu Blue is an amazingly fun deck to lose with, which again, like is a great mantra for locals, like it's a great deck to bring to locals. A deck that's fun losing, one deck green Vader ramp, not fun to lose with. That is an unpleasant deck to lose with. Like, that's definitely a deck that I just, yeah, but no, fatties, it's fun. The, it's the stop, the stop or be stop kind of deck.
But it's like, yeah, if you're, if you're losing with that deck, it's just like, well, I wasn't, it's because I'm not playing anything. My ramp's not coming up fast enough. Han-Tu is a great deck to lose with, but I, I aspire to more than having fun while losing and then being called dumb. Damn. Oh, that's fire. Sorry, sorry Parker, sorry, sorry. But speaking of Pax West, we're, we, you and I are going. And I'm going to go pretty hard on Star Wars on the minute.
Zulu, sport game cafe out of Northern Seattle is running a plethora of events there. Just as booked, you can be just as booked at Pax West as you can at Jankon. No exciting play map, but, but plenty of exciting game. And I'm going to be going heavy on the draft. That's increasingly my preferred way to play this game. And I, my problem, I played two shadows drafts. And the first one I played, Ray. And I, I wasn't just fed. I feasted. I was just handled. I opened the marrow and was.
Yeah, yeah, yeah. Yeah, like, nobody was even really playing a heroism. And I was just being like, pack after pack, like first ten cards are, like some people competing for blue. But I mean, it was just being handed to power bombs over and over and over again. And blue green. And I ended up running like a nice, like my top end was four gentle giants and a tariff hole. And I mean, the tree tenders and everything like, there was no mid game.
I just had a lockdown, twos and threes, space and ground. And then, you know, I could ray bump those up to get me. And to resource six. I play gentle giant. They either panic or calmly remove it. But one way or the other, like the first gentle giant never lands. But when you drop a second gentle giant and limited and in one of my games, my third was also there. Like that's, I mean, you know, right? Like, yeah, like that, that's, you know, so that was great. That just happened that way.
But I played another one where I was in villainy and I was just like, I had Cad Bane. I was really excited to be Cad Bane. I like it. Sorry, it's been a bit. Can you, can you remind me of Cad Bane? So Cad Bane is the villainy leader, yellow villainy. And he, whenever you play a underworld unit, he pings your opponent, one of your opponent's units of their choice for one damage. Okay. Their choice. All right. Yep. And then at six resources, he flips. He's a two eight raid two.
And that ping damage goes up to two. If you're not familiar with Cad Bane. So, and the other one was Cad Bane, I just wanted to make that happen because he's super fun and limited. And I kind of pushed it. And the signals were clear. And I was like, talk to the hand. I'm playing Cad Bane. And, you know, not optimal. Right. As noted, I aspire to more. I aspire to better.
So in anticipation of Pax West and drafting there against strangers, not my locals, people who paid good money to get to place who at Seattle, which I assume means they take this seriously. How do I, I think the phrase is stay open. Like, how do I not commit, as soon as I know my leaders to what I'm playing? I mean, yeah, that's a fake leading question in that, you know, we've gone over our script. That is a genuine question that I would like to understand better by the end of the episode.
It's hard. So we're going to be basing this segment a lot on two articles. So the first article is Ben Stark's drafting the hard way. And then a recently released response article to drafting the hard way. Ben Stark's article is back in the dawn ages of 2012, which, you know, I was, I was reading about it. And then I'm like, oh, yeah, that's when I got back into magic. And I felt so old. So I'm of the opinion that you feel, I mean, 2000, like the started net runners.
I mean, it's not actually that old. We have, we've referenced channel fireball articles from like the 90s. I think it's timed out. You feel like it's a dawn ages because of time dilation from, you know, magic dropping, like 20 sets a year. Like it feels to you like it's, oh, man, we grew up with like, what? Racist, yeah, yeah, standard and one block each year was the, you know, and standard was 90% re not standard. What was it called?
Yeah. The format was called standard, but also the set, the, yeah, there was the block. You also had block. Oh, you had the corset, the corset. The corset. Yeah. So like the course was like 90% reprints. And then you had block, which was just, yeah, like, yeah, like now that's most of August. You get that many. So I don't think it's the dawn ages. I think that's just time dilation. So anyways, okay.
So we got a 2012 Ben Stark article that is recently addressed by Palo Vito or Palo Vito Dama de Rosa. Okay. And, and yeah. Two things. One, dear listener, if you're in the car, as I hope you are, as our audio gets increasingly better, but our visuals still stay pallid and sad. We do not expect you to read this. We are linking these. We're just a current with magic.
And then on top of that, as Ned likes to say, we are providing the value ad of visiting these articles from a swoo context because a lot of the cutting edge game theory, am I allowed to say game theory when it's theory about gameplay? I think that's right. Yeah. It's a loaded term in math, but the game theory of the game trickles down from magic and is applicable to our new green field game. So all right, drafting the same thing way. Draftically correct way is Palo's articles.
So yeah, I was saying, yeah, yeah. But to start off, we're going to start off with an example because I feel like it is helpful to learn from examples rather than from like a theoretical perspective, unless you are a horrible math grad student in which case, you can, they were never willing to account that I'm a civil man. So let's pretend that you are drafting shadows and you have drafted the following leaders. You have drafted Catbane, Bosch and Han too.
So you're cutting yellow villainy, red villainy and red heroism. So you're probably either in red or villainy. Yeah. Okay. So Pac-1, you open it up. You see Django fat. Great card in Catbane, great card in Bosch, great card just period, slam it. Great. Forelom pops to your mind, you know, forecast, ambush, it's not like an insane bomb, but it's a darn good card. You're happy to play that in constructed formats.
And then your third card, mock-on-clay, plays great with Catbane, plays reasonably well with Bosch. So you're feeling pretty well established in, I'm probably Catbane or Bosch leaning much more towards Catbane, but like Bosch is a live option, right? Next pick comes around. So this is Pac-1, pick four. Yes. And you two cards jumped to your view. One is Chewy Pikesbane, which I think is a bomb by almost anybody's definition and chain code collector, which is a worse forelom.
That's right. Yeah. Yeah. So we're going to ask the question, which card do you pick, and we're going to also stipulate that we're ignoring cash value, sentimental value, et cetera. This is just, we are trying to win as many games of this swoo draft as possible. Okay. Well, then in my case, the way I've been playing is that I'm taking the chain code collector because I'm committed to Catbane, Pac-1, and I'm just kind of grinding that enforcement now. But I suspect this is a loaded question.
This is a loaded question. So we're going to, this is kind of like a thesis and tith, this is kind of thing. So we're going to come around to an interesting perspective on it. But let's start off by talking about Ben Stark's article, which is chronologically first. So that is drafting the hard way. And Ben Stark's contention is you are naive if you are taking chain code collector. You should take. So chain code collector is the easy way. That's the easy way, right?
Okay. So chain code collector is the easy way. And the hard way, which is taking chewy pike spank. And the argument that he puts forward is number one, a signaling argument. So when you are picking your first couple of cards, you have the absolute minimum amount of information, right? Like all you have seen is the cards that you have picked and very few other cards that other people have picked.
But as more and more packs are, as the picks go further and further down, the power or rather the amount of information that you are getting goes up. Because you have seen more cards being picked. So the fact that none of the three players to your right wanted chewy pike spank is a substantial amount of information. Okay. So. Orro. Yeah. But, but. So there's the signaling question. And you can just say pause. You're going to come back to this. But Star Wars Unlimited, we immediately draft.
I mean, there's a see, there's a, you know, like sevens wonder leaders drafts, right? Like, there's this. So you're already tell, you're telegraphing your colors before your pack one pick one, your pack one pick one is actually your fourth pick. Yes. So you do have a lot more information, right? And so you can look to the people to your right and you can see like what's going on here. You can see like, okay, maybe I'm just lucky and they have like monovil and the leaders to my right.
And so they wouldn't take this. And so, but I mean, that is also information. That is something that you need to consider. But the other thing to note is that from Ben starts point of view, just because you are taking chewy does not mean that you are now suddenly flipping a switch and I am now committed to haunt you, right? And so, what Ben starts perspective is this is a gamble. You are taking an option that might pay off.
And the reason why you are doing this is because if there, if we imagine when everybody, if you imagine all the cards and all the packs that are being opened for the draft, there are a certain number of cards that fit into each deck, right? So if I'm playing CAD Bane out of the entire table, there is some fraction of the cards of this table that will fit into my CAD Bane deck.
And the harder that I am pushing to go into CAD Bane, the more people that are all fighting for these same cards, the worse everybody who is trying to play a CAD Bane style deck is going to be. And naturally some cards fit into multiple decks like Django Fett fits into a bunch of different decks because it's a good card, right?
But the more people that are scrambling for the same colors, the same kind of archetypes, the worse off each person is going to be, and your job is to stay open, is to you want to take, especially in the first pack, the best possible cards because you don't know what archetypes are going to be available for you to take.
And if you start fighting immediately, you may end up because the most limited amount of information has been shared in those early picks, you are going to end up having to fight against more people than if you just kind of stayed open. And as like in the end of the day, open as long as possible, right? Is as long as practical or like, you can only put, you know, 30 cards in your deck, right? So your objective should be to have the best possible 30 cards in your deck.
And a lot of the time, you are going to be picking cards that are just going to drop on the floor, right? Like even if you got like every single card that was earmarked for like a CAD Bane or a BOSC style deck, right? There are probably more than 30 cards that you could make an argument for playing in a CAD Bane or a BOSC style deck. And so therefore, you want to take the best 30 cards.
So wasting picks on something like a chain code collector, which there's going to be more chain code collectors. And if there's not, that's also information, right? Like, if the chain code collector is the best possible card that you could be taking going forward, then you need to find a different plan because you are not going to win, right? Right. So, so his.
So for pack four and four, it's just a risk reward payoff of that there's the chewy Pikes Bane is, has a very reasonable chance of just being a deck card in your red villainy deck.
Yes. But there's, you know, there's, if it pays off, it's, you know, by lottery ticket with 20% odds is a good deal, even though 80% of the time it's wasted money because the payoff of, oh wait, nobody is, I mean, so, so Chubaka Pikes Bane went from somebody's pack one pick one to somebody's pack one pick two, somebody's pack one pick three. Yeah. In your hands, like people have already opted past Pikes Bane for three cards. Yes. And a very decent chance that that's going to pay off.
Yes. And, you know, you're going to get a whole bunch of blue heroism beef. Yeah. And then you can fit that into Tonto. And, you know, there are arguments that won't happen, right? Like maybe ahead of you, somebody had like a podamer end or like a crap dragon or something else. And it's going to end up that your Han-2 deck doesn't pay off.
But his contention in drafting the hard way is that if you draft the easy way, there are times that you are just going to get fed and you are going to win, right? Like you, your first couple of picks are almost anything, right? Right. Your first couple of picks are going to line up with exactly what you want. And the rest of the draft is a breeze. You just take all the cards that match and, and bam, pops your Echel.
Whereas drafting the hard way, you will sometimes have a little bit of more of a struggle moment in your deck building. But the odds of you having a complete garbage draft where you end up with like a pile of cards that doesn't really work together goes way down because you kept yourself open and because you read the signals. Okay. So, so the argument is that I'll avoid the situation where you just have a deck full of fine, yes, borderline playable and draft cards of your color.
Yeah. Because that is what is going to happen if three people are all trying to draft headband at like an eight person pod odds are high that all three of those people are going to end up with fine decks or, you know, again, depending on the situation. Okay. Okay. It sounds like a compelling argument. I think that it is. But there's been a article responding to that. Yes. How those drafts in the correct way. So what is draft? I mean, I would rather draft the easy way.
So I'm not going to be the palos argument is draft the easy way. Palos argument is essentially that drafting the easy way gets a bad rap, right? Like it is a tendency of people who are sophisticated limited players to try and stay open as long as possible. And the longer that you are staying open, you're trying to keep yourself open for like the powerful cards that are going to come. But if most of the cards in the draft are basically fine, all you're doing is you are weakening.
So like, let's imagine that two sets, like one set, there are a couple of cards that are amazing. And then the rest of the cards are kind of Garbo. And another set where it's kind of flatter, right? And his contention is that in a steeper situation, you have to be open because otherwise you're playing a bunch of cards down here. But in a flatter situation, it's okay. Optimizing for the low end early is better than getting the keeping open to a marginal top end.
If you just immediately start optimizing, picking the best comments in your color. Yeah. And yeah, like, you know, you pass along a bomb in a color that's way off, but you've immediately optimized for, then the gains can be there as well as palatism. Yeah. Yeah. Okay. And there's like a second order argument. And that is when you are taking Chui, there is a certain amount of bad signal that you are sending, right?
Like if I am taking Chui, I am discouraging everybody from my left from going into blue heroism. Because otherwise they might see the Chui and they might be confronted with the same decision that I was confronted with, right? And if they're not going into blue heroism, then there's a higher likelihood they're going into Villanileville, Villanile Red. Right.
And then when pack two comes around, they're passing to you and they've already taken the Villanileville, Villanile Red cards because you speculated. Right. Okay. So, but Palo is not making an argument that you we should always draft the easy way. Palo's argument is essentially considered drafting the easy way. And he provides a series of tests to talk about when one should consider drafting the easy way versus when one should consider drafting the hard way.
So the first question that he raises is how deep is the format where how deep is the format means if you look at like an average pack of cards, how far down do you have to go in that pack of cards before you get a card that you are not willing to play in like even a limited deck? And if you do a deeper format, if you miss out on some cards, it's not as bad.
Right. Like if I'm taking Pike Spain and you know, I'm having instead of playing because I took Pike's Bang Pick 4, I'm instead having to play my pick six card, my six pick card in my deck instead of Chaincode Collector. Like how much worse is the card than Chaincode Collector? If on average most of the cards are about the same power level, it's fine to speculate. Right. Yeah. And there's plenty of like great common yellow cards in shadows and swoo.
Right. Okay. So, I mean, so how, and I'm not as familiar as magic, but I mean, so Star Wars Unlimited is a, you know, very much as a draft as a first class citizen game. Like they've spoken at length about this game needs to be super draftable, which tells me that like there's a design goal for a deep draft pool in swoo in every set. So I mean, still a value, is it relative to other swoo sets of which we currently only have two, but let's pretend we're talking a year from now or is it relative?
I mean, or is our all swoo sets pretty deep, we believe, just because the game is true? I mean, with a sample size of two, both of the sets have been reasonably deep, but I would say that sometimes there are archetypes that don't come together, right? Like I would say that in Spark, heroism yellow didn't quite feel like it had like an archetype together and you were kind of taking cards that were pretty good. And sometimes.
Although I remember you doing a draft and basically feasting because everybody discounted heroism yellow and you were like, right, okay, it's not the best color, but if nobody else takes it and I just get to take all of the heroism yellow cards that I want. Yeah. Like that's better than fighting one other person for whatever. Right. And you've done more shadows drafts than I have. So is there like a color combination that you felt that hasn't quite worked in shadows draft?
I mean, I did a lot of pre-releases, five pre-releases and two drafts. And in my experience for shadows, I think it's more that because each color, the heroism villainy pairing only has one common leader, there are types, for example, I just think and get at me in the comments if I'm wrong, I think Gar Saxon in draft is pretty bad. Like obviously you'd rather have Kira, but like there just aren't enough draftable.
Like good upgrades, good upgrades or ways to get XP or ways to get shields to make his plus one plus a worth it and you know, when his flip turn is all about like protecting that sick upgrade, the dark saber obviously or whatever so that when you lose it, you get the upgrade back with like congratulations, you got, I don't know, a pistol or a ruthless next back. Yeah. Like which is sure that's a card that you get back, but is that worth your leader pick?
So I don't think it's that the color pairings are bad. I think it's that there are leaders who are just kind of out of the gate bad because you can't get what they need. You know, similar to Tahondo in the last episode we kind of knocked on Hondo a little bit and constructed in Mayacolpa. Like sounds like there's some. No, I watched those games. No, there was definitely some juice in those Gen Con games now that I was able to watch the footage.
So like Mac Alba, but you know, in a 30 card deck, even if you can get enough good smuggle cards, like smuggling out like five, six, seven cards and limited, like the decks are already thin at 30 and every time you smuggle, you mill yourself one.
So like, you know, I think, you know, there's some, I think there's just a couple of leaders who it's like, they're a bad pick if you can avoid it because you, not because you won't see good cards, but because you in their color, but like there's a card type or component type or a keyword type that they're meant to synergize with, which you're not seeing enough with. But I don't think in shadows, maybe Redville, no, they're not all the, I can't think of a color.
It's more, I mean, it's such a like upgrades matter. You know, like a bounty hunter's matter, there's a lot of keywords. And if you're too dependent on that keyword and you aren't able to pull it off, then you end up in that mediocre playable, right? Like playing CADBane with a lot of good yellow, villainy cards that aren't underworld. Like that. That's just like a bad leader if you're not getting the value out of them.
Yeah. So it's hard to get enough, you know, a boss is good just because he's such a beast and he doesn't have to fire much to really fire. Yeah. But again, getting enough bounties to feel like you really made the right choice if there's another villainy red option for you, right? Yeah. Anyway, so before we get too far, okay. So yeah. So there's the format depth and I think that the leader depth relates to that, right?
If my villainy leader is Gar Saxons, let's imagine I have two heroism leaders and a villainy leader and my villainy leader is Gar Saxons, let's say. How deep is the format for me in villainy blue? Probably not as deep as I'd like and so I might want to speculate less if I'm leading in a villainy blue direction. So that's sort of generic set wise considerations.
But then there's also like card specific considerations because I kind of stacked the deck by making it chewy pike span against chain code collector. But you know, like how good does the card need to be that you need to consider on speculating on it versus chain code collector? And so Palo proposes a couple of things that you should consider. So one of them is how good is the card compared to what you have now?
So like for example, how good is chewy pike span compared to the things that you have? Would you rather play a deck with Django for one and Makonkli? Or would you rather just have like if you were white room situation, would you rather have pike span or would you rather have those three cards in your hand assuming that you had the resources to pay them? Or just I mean, right, like which one do you want to be your dead pile?
Like assuming the rest of the draft goes the way you wanted to, you're getting 40 something more cards. Yes. So happy if your dead pile is Makonkli and for a lot of them and so forth. Yeah. The second is how good is the card relative to an average card? Well, better the card is relative to an average card than where you want to speculate on it. I mean, that really just feels like how good is the card, right?
Right. But I think as a consideration on that, there's this idea of synergy power versus base power. So centered base power is like how good is the card if you have complete white room scenario or just playing this one card, how good is the card, how good is to use the pike span? I'd say white room. Pretty good. Like it's good stats, it smuggles, it kills something with five or less remaining hit points. Like unless the format is very odd, it has a large number of units with lots of hit points.
You're generally happy to be drawing. Uh, uh, chewy, but with chain code, right? Like chain codes are four to with ambush. So it's effectively like a removal effect that costs four. But if there's a lot of bounties in the format, then the fact that it's got that synergy power going for it. So if you're already, uh, if the card you're speculating on has a lot of power, but it requires a lot of synergy to work.
Um, so like, I don't know if the dark saber might be, you know, the dark saber is pretty powerful on its own, but I think that it really shines in certain kinds of decks. Is there another card that you'd recommend? No, I mean, so mcclunky, for example, is it? Yeah, it's a dark saber in every, I, I, most decks that are running dark saber are just running because it's a lot of beef by itself. Right. If you have other Mandalorians out great, but that's not why you're running it.
Yeah. Um, no, but I mean, like mcclunky, right? Like mcclunky is a fantastic card. If you have a critical mass of, uh, cheap underworld units, right? Like if you're, yeah, if you're bouncing Xanadu blood to deal three damage to something, I don't think you're feeling real good. No, you want. Like, yeah, so it's a great card. Yeah. If you have the synergy. So, um, and I think, but I think that goes back to, you know, that, uh, the leader synergy again, that there's this built in pool of synergy.
Raise synergy is easy to hit, right? Like, you know, you, you need two power cost cards in your deck. A lot of them because you need turn one, turn two plays, many of which are two power cards. So it's not a huge sacrifice to take a card that really synergizes with Ray because you kind of need that card. If you're in that color at all, and you know, it's going to synergize, whereas like taking in a garbage upgrade just because you're in Gar Saxon, like you only have 30 cards.
You don't want eight garbage upgrades just to trigger. Gar. Yeah. So, I mean, I mean, how many brutal tactics do you need? Right. Yeah. Exactly. And, and, and so, yeah, that's, yeah, there's a component of like, I think the synergy that you're asking is just a hypothetical synergy. You already know some, even if it's pack one pick four, yeah, one pick three, you know some synergy that you're aiming for. You're aiming potentially one of your three leaders is looking for Mandalorans.
You know that, right? Right. Like, you're, you got Boca-Ton. You know you're looking for Mandalorans. You're looking for whatever smuggles. You're looking for bounties. You're looking for whatever it is they need to fire. You're looking for that pack one pick one. So, okay. So, how are we going to actually zoom? I'm going to zoom in on that. So, yes, for the synergy argument. So, like, let's say that you have Boca-Ton as one of your leaders for, for sake of argument.
And you're being given, I don't know, let's say like a, like a house cast soldier, right? Like how much do you value it? So, that's a two cost red, two three with the Mandalorian trading saboteur. And I'd say like that's, that's kind of a marginal card, right? Like, I'm not, I like it because it's a two, but there are a lot of things that I would rather play over it.
How aggressive do I want to be versus that versus, let's say like, I don't know, a four or five cost red, sorry, let me look up a four cost red card that I would like more than that. Like a hunting next sue, right? Where, yeah. Where pillage. Yeah, pillage. Yeah, pillage is great. Yeah. Yeah, I mean, so, okay, so, so, yeah.
So, again, if you're not in Bocca-Ton, that you don't want, I mean, that's the house cast is something that you're taking because it's the last card in the pack of the colors you're looking at. And you need some twos, right? And you need some twos. Right. But like Bocca-Ton is, makes that card suddenly, mmmmmm, worthy of consideration a little bit earlier, I mean, it's still not a great card. But yeah. Okay, okay. So, I mean, like the synthesis position and to kind of try and put a cap on this.
The core idea is one ought to draft. When one drafts, it is important to not fall into the easy pit of, you know, the, the trench on the right side, the ditch on the right side of, I have picked three yellow cards, and therefore I will take every subsequent yellow card and ignore the other cards of which might be better. And one also ought not to just cherry pick the quote, quote, good stuff, and then hope that it comes together at the end, right? But there is.
So I kind of, I feel like Palo is, I mean, it's a well written article, and I cast zero shade here. But if the question is, should I draft the easy path of go for the synergy, or should I draft for the hard path of, leave myself open. Palo has said, taking the, like, you know, the my favorite answer is that, yeah, it depends. Like, no, exactly.
And I mean, there are criteria there, but you have to evaluate each card of all these questions of variables of how far a field, you know, and, you know, it's risk versus payoff, right? Like how big are, right? Like how far a field is this card? And then how big a payoff is it if I'm actually able to make this work? And so, okay. Then the other question I have, and this is a question we haven't been able to ask in a while, Ned, is what is your name? My name is Monday. My name is Monday.
I'm a mathematician. So, it's been a hot minute, but I finally had time together because I got most of my prep done for Paxbests. And so I was able to run this regression with the shadows cards. Okay, so the math for the segment is the classic math, because we've done a couple different episodes. The most recent math we've done has been on like tournament performance allies. This goes back to what I have been evangelizing as Ned Math, the, Yes.
The towards a perfect custom homebrew card, how are the fantasy flight developers valuing things you might find on a card? Yes, that is exactly it. Is the goal of this regression. So I'm using a tool called linear regression, where essentially you plunk everything and then you fill a line to it. And there are more complicated and better ways to do it, but linear regression. Well, let's not dive into the weeks, right?
I was gonna say, the episode has been growing, and we've got probably a lot of listeners who haven't gone back into our catalog. We've explained this, and I want to reward our loyal followers by not explaining how this works again, but the truth is math is math to quote the Incredibles. And Ned has been able to, regardless of what is true in a universal context, or what the developers think is true, we don't know what either of those things are. No, I don't know any of these things.
Ned doesn't know what either of those things are, but Ned can predict, or not even predict, but Ned can describe accurately what we have seen across two sets that, based on the 504 cards, so that's right, that's right. 504 cards that we have seen, this is a truth, that these are how the developers in these two sets so far are valuing power, how they're valuing hit points. And we, yeah. We'll, Ned boils the budget down to adjusted stat points.
Yes. Our original pitch or Ned's original look at just Sparker Rebellion was that power was worth about two adjusted stat points, and hit points were worth about one. Yes. Pretty high confidence. We got more information in shadows, and Ned was able to revise that, but believe power is now 1.3 to hit point one. Yes. There's a slight preference.
So we can say given AA cards cost, yes, we're able to figure out some number of adjusted stat points, yes, that we expect, that math expects will be there. And then we can say, here's what you can buy with those adjusted stat points. Yes. And then up till now, Ned has been limited to power and hit points. That is what we have discovered. What do we have quantified? But there's a lot more cards. Go with the cards.
And Arena. Which we, Arena is one hit point, and hot dang does black sun from cell block guard be like, yeah, no, there it is. It's up. One hit point is one stat point by Zua hit point or the space arena, and there it is, like, down to the keywords, perfect. Love it. So or traits down to the trait. Yeah. So do we have new things you can quantify or was it just a set point? Yeah, we have a lot more variables.
So something that wasn't mentioned is previously also estimated what I call damage and card draw, where that's like cards that draw cards when they enter the battlefield, sorry, cards that draw cards when they are played, and cards that deal damage when they are played. And sometimes, and one of the things about regression is that, so card draw is obvious. Like, yeah, you have the Yoda gives you a card, you factor that in. Right, but you should have a card draw, right?
Because Yoda isn't guaranteed to draw you a card, right? Like the game might end before you draw. You a card, right? The game might end before you to draw to your card. You would have might get wailed and never get played. There's a bunch of these things, and so up on the GitHub, I have all of my estimations, and I had to revise a lot of them because I came to a bit like, having played a lot more swoo, I now have a much better feeling about how to evaluate things.
So like, for example, as you're always clear, the, without going into all of it, the estimations, like Yoda is worth more than zero card draw, less than one, and how much card draw Yoda is worth, is a number that Ned has made his best assessment on. Which is one. It turns out that it's, I have asked what you would have at watch. Sure, but we've talked like, taking the class examin'. Right, right. And, you know, so we've gone over all of that.
If you really care, dive into the GitHub and get at Ned in the comments, but for 95% of our viewers, and if you don't care, assume that we're Ned's massaging in good faith. So what else have we learned there? So I have added the following variables, where ready, which is measured between zero and four with zero being fixed, and four being legendary. I have added the presence or absence of it bounty. I have added grit.
I have added restore, which is numerical ambush, shielded, and raid, which is also numerical. So, okay, I'll start go back on the numeric ones. So awesome, okay. Wow, okay, so I'm excited to hear about rarity. But, unless you have a thought, I say we do the, let's do the keywords first. Sure, yeah. So, bounty is worth about 1.74 adjusted set points. And this is, it is hard for me to disambiguate bounties as in like escape wookie. What is the 333 yellow? That's a card.
Yeah, no, that's an exhaust unit, right? And high low bid enforcer, which drives the card. How ought you to evaluate these? I can't really. Compared to Dr. Amazon, right? Ready 12 resources. Like these are all. They're all just wrapped up into the category bounty and the regression found it significant. So that's the way. So the existence of a bounty, obviously is significant. How much? Right. Because every bounty is right.
And when you're lumping everything together, you know, like the, the high, the things that you're paying a lot like Dr. Amazon, for example, I think is probably dragging the value of bounty up. And, you know, escape wookie is, is fugitive wookie is dragging the value of bounty down. And there's in theory, and set nine, that you could be quantifying each bounty by the value it has just in that.
Yeah, draw a car, negative draw a card, you know, a ponant draws a card is not, so it's no longer bounty. It's that this card does the thing. Yes. Okay, but for now, all the bounties are treated equal. And they should give you an extra, almost two stat points. Okay. Almost two adjusted stat points. Which is, you know, a lot, right? Yeah. So next up, we have grit. The cost of grit is approximately negative 1.23 stat points.
So if your unit has grit, it will have about a point of power less than you would expect from another comparable unit. That feels like that tracks. Yeah. Okay. Restore is negative 0.73. So like, restore one is less than a hit point. So restore one probably isn't going to be costing you any stat points. Is the report, restore getting 0.7, costing 0.73 adjusted stat points per, is that per point of restore? So in theory, restore. So restore three on Luke.
This, the math is finding that's about two in adjusted stat points. Yes. Yeah. So, yeah, that is the estimate. Naturally, like not everything fits. Like some cards are going to be above the line. Some cards are going to be below the line. But this is, this is the best estimate. The line. Am bush. Yeah. Am bush is about the same as grit negative 1.22. So you're about a point of power shy or slightly more than a hit point shy if you are an ambush in it.
Shielded is also in the grit space of negative 1.24. So again, it's like a point of power. Not to put to find a point on it. But if you compare like probe droid versus crafty smuggler, you can see that we're like crafty smuggler versus the wide variety of two, three's with different keywords. You can see that crafty, there's a reason why crafty smuggler is included versus the two, three's with keywords.
And that is two, two shielded is slightly is worth slightly more than a two, three with another keyword ish. OK. OK. And raid is about the same as restore its negative 0.69 per point of raid. So each point of raid is like two points of raid is effective to a point of power. So based on the ones you've been able to quantify so far, we're finding about a point of power for a keyword. Yeah. A keyword, then you're looking at about a value of three for two power, two hit points from the two points.
OK. Two hit points for about three. OK. So those are the keywords. And again, there's constant intervals on these. So that's the regressor does its best estimate. But this isn't. But these are all notable. These are all within some bound of you would be comfortable standing in front of a group of people or say podcast in front of thousands. Yes. And say, yeah, this is probably true. Yeah, this is probably true. Yeah. I would absolutely stand by these numbers.
These are the absolute best estimate of the numbers as has been fed in. I am completely willing to stand behind a defend. OK. But the one that I asked you to calculate, the one that I was excited that if we had enough is rarity. It is. We do have enough. And it's pretty significant. So rarity is about 0.256 per point of rarity. So like a standard card is 0. And rarity has a little bit of an annoyance because there are some cards that appear more than once in the rarity.
So for example, if you consider taggy, taggy appears once on the fixed sheet in Moffgitian stack and also as a common in Spark. So. Yeah, I think he's a common. I mean, yeah, I mean, that's really good. So I have estimated that as two different cards. OK. OK. Yeah. But that's not entirely fair. And there needs to be some deduplication going through. But it was funny to watch taggy popping up as two different cards. No, there are the 10 re-points in the shadows starter.
But otherwise, yeah, the numbers that you're running aren't including like the varying rarities in organics play. Right. Right. OK. So you're saying that at 0.25, then being... So a legendary is worth about a point, a hit point more than being a common. As an at cause a point to be a legendary. No, no, no, you get an extra stat point. Oh, you get an extra stat point. You get an extra stat point. Which is actually like one hit point for a legendary's that tracks.
Like, I mean, that feels like considering you know, yeah. Like, I mean, Yoda's not a legendary. But so that's a question. So you did it run uniqueness. I did run uniqueness, but uniqueness was not significant. So we haven't been able to do that. Yeah. So to this point, I also had to remove draw because the number of units that had draw as a when played effect was not large enough for me to be able to have a good estimator on it.
Again, the larger your population is, you do need a certain density thing to appear. Yeah. I'm just wondering and revisit this. But so uniqueness can only occur at uncommon. Right? So there are no common unique units. Yes. Not counting. That is a selection effect. That goes into why I don't have overwhelm is because the regression is very confident that overwhelm, being overwhelm buys you some stat points. And the reason for that is that we do not have very many overwhelm units that are small.
Ah, OK. I yeah. So there are selection effects that you want to have to consider. Well, that's actually like, and I'm seeing these numbers for the first time, dear listener. That's actually really exciting for me that, I mean, one point, I mean, we wanted there to be good comments. We wanted to be good on comments, said every TCG dev ever about every TCG ever. So like, great. But we'd like to look at the pudding, please.
And I mean, I feel like one point in, prefer a legendary compared to a starter card. But that does not boggle the mind. That doesn't feel like awful. No, the difference between a common and a rare then is half a hip-point. Yes. Yeah, that seems fine. I mean, with the caveat that I'm not taking text into account, right? So yeah, because they get more complicated and do weird things. Right, they get more complicated, right?
And so that becomes increasingly difficult for me for like a simple tool to estimate. So another thing that I will talk about is now that we've talked about words. But before we get there next time, you do more regression length of the words in the ability box. I would be very curious to see you. That's a good one. I was going to know if the words are good, but we can look at if there are more of them. So come back with that. I absolutely will.
Another thing that I'm going to add is ramp right now. Super laser technician just looks like a bad card. Because ramp is not modeled because it's a relative, is not a very common feature on cards. But I want to see five ramp cards now. Yeah. And I standard like five clear ramp cards. And of which, yeah, price on your head and the assault for persistence. Oh, well, and then the assault cruiser. Yeah. And then DJ smuggle is also kind of a ramp effect.
OK. Yeah. Well, when we got a pool of enough to inform, because I super laser tech is not a bad card. And then I completely agree. So tell the math. Yeah. I will absolutely tell the math that it's got that wrong. And this is one of the reasons why the math is a guideline. And not like it is a good first place to look at things. And we will get to that in a little bit is that I was surprised.
I was both surprised and heartened in the cards that popped up as the things that have come up as the most under-costed relative to their stats. Are you ready to talk about that? I want to talk about a couple more things first. OK. I've got a tease. I'm going to, I mean, it's the stick around to the deer listener. And yeah. Yeah. These are five most over-static cards. So yeah. So there's a constant term.
So basically, if you were a zero cost card, I would estimate that you would have about two and a half points of adjusted stats. So like a 1 1 for 0 is about what I would estimate you at. So that makes the battle droids. Yeah. Battle droids, which means clones are, I mean, we knew they're better. But yes, clones are above curve, not battle droids, or below curve. OK. OK. That is the estimation.
Each point of cost will buy you by itself about 1 point of power, 1.2 worth of stats, worth of adjusted stats. OK. But there is a squared cost term. So like, for example, and the reason for that I've discussed before, but basically, if you look at the units, if you were to plot like the adjusted stats of all the units, it's not R2's like. R2's the best card in the game if you don't take into account that a couple extra points at 1 is different than more later. Yes. Yeah. OK.
So you have this like squared cost term. So more expensive units tend to just have more stats than a pure line would predict. It's not a lot. It's 0.07, but it's times cost squared. So it would like a six cost unit is going to be 36 times 0.07. So that's like another three-ish points of stats that you're going to get on a six cost unit versus a 1 cost unit. OK. OK. Yeah. And I've tried a bunch of other functions. Cost squared seems like it works pretty well. It gives good significance.
I'm happy to make arguments for more complex forcing functions. But another note about regression is that the more variables you add, the less stable the regression becomes because you get essentially there are a bunch of different ways to express things and you get confusion. So before you're running each of these, you're not adding all these variables. You're adding each of these is like their own run of like. No, no, no. I'm not aware of this.
People run with all of these variables because if you do each one individually, you will get it bad. It's that this is the number of variables that I've been able to get through regression to sustainably support. Oh. And the more cards we get, the more cards we get, the more variables it can sustainably support. To use an analogy, if you're talking about baseball pitchers, if you say, OK, well, the pitchers left handed and the batter is right handed. And it's in the summer on a cloudy day.
And it rained two days ago. And there is a home game in a northern hemisphere, like a northern US field. If you add enough caveats, you can get down to a very, very small sample size. And that's as you're adding more and more like. And you want a big sample size. So how many variables can you add and still have a sufficient sample size? Exactly. Right. Because otherwise, like each circumstance is unique. And then you don't get a good regression. You get garbage.
OK. The last thing that we have garbage. I do the fun stuff. We do not have garbage. So this hopefully can appear. But this is my error histogram. So QQ plots. I got some feedback from some colleagues of mine that QQ plots are lovely. But the layperson does not appreciate a solid QQ plot. So the error histogram is a more intuitive way of expressing this information. So essentially, what it is. Speaking of layperson, yeah, I like bokers. And just bar graphs are made by Excel.
And so I can track them. But yes. So this is the x-axis is the error. So how far off we were in our predictions? Where positive. Yeah, a positive number is a we were off by we underguest. And a negative number is we overguest. The regression overguest. And then you ideally want this error histogram to peak at about 0, meaning that most of the time you are on. It has shifted to one side or another. That means that you're consistently undershooting or overshooting.
You'll notice that this peaks at about 0. We may be overshooting a little bit. I could see an argument for that. But it looks like it peaks at about 0. And then you also want it not to be skewed. So you want about the same mass of the bell curve to follow on both sides. And the thing that I forgot to mention, but ought to be said, is that the height of each box is the number bar. Thank you. Is the number of times that we had that particular miss? Oh, OK. Right. So you're tracking.
And so you can't track a. Yeah. OK. Wow. I mean, that looks speaking on behalf of all the lay people, that looks fantastic, Ned. So that looks accurate. So can you tell me? So there is, it looks like there's one or two occurrences that are like way on. Is that super laser technician that we weigh? I think that the like minus four, or the minus three to minus four is super laser technician. Oh, no, sorry. What's the one that's almost six? That would be on the list.
Oh, you know, is that the list of top five? Oh, yeah. Let's see that then. Yeah, OK. OK. So top five over cost. So this is the math says one thing. This is under cost it. These are under cost it. How do I put your cost it? Yeah. Don't say under cost it. Overestadded. It sells a bet. Overestadded, yeah. So this is the math says they should cost based on 500 cards. Says they should have four cards. They should cost X. And they cost a little bit less than X. Yes. Or a lot less than X. Yeah.
OK. So top five, what is number five? Number five is crate dragon. Crate dragon is almost entirely on the basis of the estimate, or not entirely, but a significant amount of it is on the basis of the damage that I've assessed it, where I've assessed it at about five points of damage on play. Oh, because it's going to take. Yeah. If they rivals fall immediately. Exactly. They're taking six points of damage to something. Yeah. OK. Oh, OK.
So it doesn't take into the, because that blows blew my mind that it was on the list at five, because it doesn't take into account overwhelm. And I was like, there's no way you could codify its ability. But you've codified its ability as five points of damage on play. Yes. Because sometimes it gets rival falls for six. Sometimes your point runs into it. Sometimes things happen. But I've codified it at about five. How much damage does your model count when they scoop? The model is not a problem.
In fact, that is an interesting situation, because you might estimate it. That's a lot of crates that they scoop. And it inflicted it inflicted no damage. But worth 30. Yeah. One or the other. OK. All right. So five is great. Right. It has 22 points of total stats. And we predicted it at 19.34 for an error of 2.66 points. So about a hit point and a point of power worth of underestimation.
Next up, number four, we have wild rank or with, again, like another, I estimated it at four points of damage coming in, because I'm estimating like it's going to deal two damage to other units. There will be two other units in the battlefield, which it will deal two damage to each of them. So here's a question. Does your model for dealing damage include dealing damage to your own units as a good thing? It does not, currently.
OK. So you're assuming you're playing wild rank, the math assumes you're playing wild rank or when there's at least two other enemy units on an empty battlefield, yes. On an empty ground arena, yeah. OK. OK. Yeah. I feel like that's not that uncommon of a description of when one plays wild rank or no. You're not playing wild rank or while you have four ground units and they have none. Yeah. I mean, so wild rank or has 15.2 points of total stats.
And we predict it at 12.34 points of total stats for an error of 2.86 of overpowered to its expected cost. OK. Number three, I was actually surprised by, because this isn't even taking the cost reduction into account. And that is Java's rank core. Java's rank core at 19.8 points of total stats with a prediction of 16.76 points of total stats for an error of 3.
Well, that that tracks for me because actually, when you ran the old model on Spark Rebellion, one of the most overstated cards that you found was the Blizzard Assault AT, AT. And then, Java's rank core has the same cost, same stats. And then, if you're valuing it as some number of three or six damage or three damage, yeah. Three damage without costing itself. And yeah, yeah. So I can see how it's like the model treats it as a significantly superior Blizzard.
Number two is a card that I think that, speaking for myself, I am sick of seeing. It is a Darth Vader unit. So this comes in. Are you giving it an extra stats for the unit? I'm valuing it at 3.3 worth of stats and a card draw, because you are drawing a card out of your deck effectively and playing it for an additional 3.3 of stats. And you can now take into account its ambush and the history of stats. Yeah, it's got a keyword that's tracked. Yeah, that makes sense. That makes sense.
Its total stats is 19.6. The prediction we would expect it to have 14.32 points of stats for 5.28. So that's that little bar if you look off there on the right in the air. That's one of the two. Start. Yeah. And then there's one more. Oh, sorry. Before you talk about the one more, I think this goes to show that there is not a little, but a lot of nuance between what the math perceives and reality. And what is played at planetary qualifier. Please do not.
Nobody I think is, but don't use this as a guide. But the only thing that's used for a guide for is creating custom cards, because I think there is a non-trivial gap in power level between number 2, Darth Vader, and number 1, which is. Gideon's light cruiser, which actually doesn't even fit on the error plot. It is like another one notch over to the right. Oh, wow. OK. Yeah. Yeah. It does not fit on the error plot. And if I included it, it makes the error plot look weird. I mean, not weird.
But yeah. So Gideon's light cruiser is in space has a reasonable stat spread. I'm valuing it as if you're playing it in Gideon. Because why would you not? Because why would you not? And so it is that. And it effectively has the Darth Vader come into play where it draws you a card and is three three worth of stats. Because I'm assuming that you have a unit in your discard pile that is three three, and it's effectively drawing it for you out of the discard pile. And which is crazy.
I mean, I'm, yeah. Moff Gideon is not bad. That's the error. Yeah. He's not the starter villainy leader that preceded him. And if he's to be cracked, he hasn't been cracked yet. But that does it crazy to me. When I saw that list, I was like, yeah. All of these see some competitive play except for number one. So stats aren't everything. Like the math isn't everything. There's. Right. No, exactly. But Gideon's light cruiser is estimated or has 23 points of total stats.
And we would predict that given its cost and these total that it would instead have 16.03 points of total stats. For an error of 7. Yeah. Yeah. That's insane. Yeah. Wow. Well, I think that's thank you, Ned, for doing this. I know people absolutely love it. And then also some of you will get incense, which is great. I'm happy happy to have arguments with people. Passion plus iron is how you succeed at social media and content creation, Ned. So you're killing it. So thank you.
Thank you, dear listener, for bearing with us. But that is all she wrote. This was episode 21 of The For some Limited. And we're recording on August 18th. If you have any thoughts about the episode, please email us at The For some Limited Gmail. And I want to really note, dear listener, I see the stats. And we actually hear enough feedback that I know a lot of people are listening.
But nobody's asking us questions, which breaks my heart because deep down secretly, I know that both Ned and I and our heart of hearts wish we were college professors. And we're getting paid. Like 10 year to money to just answer banal, myopic questions. So if you have, we are in a position to answer questions that nobody else can answer between the two of us. If you have questions, please send them to us. We would love to answer them.
And in fact, because I was cleaning out my son's room in honor of Twilight, the Republic, we've got Obi-Wan Kenobi Lego set that I will send to anybody who gives us a question that we answer on air, as opposed to just love the show, which I mean, is nice to hear. That's always great to hear. Never gets old. Don't stop sending those. I love those. But also the answering, if you're questioning, it's asked and answered on air. I'll send somebody a Lego set. I don't know if that's worse, anything.
But I don't need two Obi-Wan Kenobi from Attack of the Clone Lego sets. So I'll give one in honor of Twilight Republic. Do you have any final thoughts, Ned? I do. So it's actually a somewhat question directed at you. Do you mind if I shill for my parents that I'm going to be doing at Pac-Man? No. So I have been actually like soft shilling for you, Ned. I am going to be playing a boat load of Star was eliminated at Pac-Man.
I have very publicly only going to do one of two things, either by the exclusive convention exclusive Dark Side Six Pack of premium Star was eliminated cards, or fuel a lot of sad tweets and content engagement about how the system was buggy. And I couldn't get the convention exclusives. I'm literally only going to Pac-West to procure. But Ned, you were already going long before Pac-West who was announced. What are you doing at Pac-West?
I am going to be running with the UnPub Network play tests of two of my board games that I have designed. They are in re one of them is in a more that finalized state than the other. But if you are going to Pac-West and you would like to play something that I designed, I will be running play tests on Saturday morning and Sunday morning, both at 10 a.m. at the UnPub area.
And one of them is a trick taker with light historical theme and the other which is a game about dwarves going into holes and murdering monsters to get money. I will say, having played a number of Ned's games, if you like the vibe of the show, you will like the vibe of Ned's games and the in versus also true. Yes. Strong and inundation that if you listen to the show when you're at Pac-West, take the time to sit down and play one of Ned's contractions.
And with that dear listener, we will see you post Pac-West in a few weeks. On use. And