Introduction

This is part 1 of the transcription of the limited level-ups episode 37. It covers the question of the week. Here you have a list of all parts of the transcription: part1, part 2, part 3, part 4, part 5, part 6.

Transcription

Hey everybody. Welcome to limited level-ups.

I’m Alex and we are sans Abram from this week. But I’m not alone. We have a very special guest joining me, Robert, otherwise known as Viralmisnomer. The main Dev for 17lands.com. Robert, how’s it going? Good. How are you doing today Alex? I’m fantastic, I’m so glad you could join me this week. I think we have a really good show today. Robert is analyzing a bunch of data from 17Lands.com and we’re going to share it with you and we’re calling this 7 habits of highly effective drafters. Robert brought a lot to the table here and I’m really excited to talk about well all the habits that good drafters have so that everyone at home can potentially get into these habits. So, before we get into that, Robert, why don’t you just tell everybody a little bit about yourself and maybe a bit about 17Lands.com for those who’ve never used it before.

As Alex mentioned. I’m the lead developer at 17Lands.com. So with 17lands we track data from people’s Arena logs. People who want to use a tracker to record their drafts and share their decks either or their records all the time things like that. And a big premise for us is that we also want to make the aggregated data from that available to the broader limited community. To kind of write a new angle for how that can help us improve help us learn the new format. See what’s working. What’s not. So yeah, so we started 17lands about a year and a half ago and been growing users ever since and the more users we get, the more data we have, the more detailed insights we can get. With that I’m excited to have a lot of detail you can go into today to kind of help people level up.

Yeah. I’m a huge fan of what you guys are doing over there. Not just the kind of data you collect but the way you present it and make it readily available for everybody. You give tools to people to allow them to teach themselves to get better or you give themselves some feedback. I think it’s super important. I think tracking your draft logs and having records of those just really helps and I think that your goals are very aligned with our goals here on the show, too.

Yeah, I appreciate it. And I think it’s through people using the tools and trackers and what not. Giving feedback is a big part of what helps us improve and make a better product.

Awesome. Well, I mean, you know, you guys are doing a fantastic job over there. So just keep it up. Before the games and meat of the show, we do want to shout out the patreon quickly: https://www.patreon.com/limitedlevelups. You know, we got a lot to cover today. So we’re not going to spend too much time on this, but if you feel like you’ve gotten something out of the show and you want to give back, you can head over there. We’ve got a bunch of reward tiers over there to get back to you. And you know, it really does help the ship to keep running, you know, we love doing this every week. Each week we shoutout our new patreons, so this week we welcome: Jurgen, Jonathan, Aaron and Axapong. Welcome, welcome and thank you very much for your patronage. We really appreciate it. And of course thank you as always to all the continued patreons. All right, let’s just dive in as always. Of course, we have our question of the week and, Robert, I would love if you could weigh in on this question. You know, I would be lying if I said I didn’t pick this question specifically because you were going to be on this week. Our question really comes from Dylan, and Dylan says:

How much do you guys feel like you need to play with the card to figure out if a card is good or bad enough that you should bump it up or down in your pick orders? This can extend to archetypes as well. I find that I can form an opinion of a card that I am often impressed or unimpressed with after a few times playing with it only to have other people online come to the opposite conclusion. For example, I thought that Skyclave Pick-Axe was a really good card when I play with it, but it seems to have underperformed for others. I don’t want to be results oriented and draw conclusions from too small of a sample size. But I also like to be able to rely on my own findings to guide my drafts. Alternatively, how much stock do you plan to looking at raw data from something like 17lands. I’ll sometimes be occurred that may be fine, nothing special, like Glacial grasp, have a high winrate on 17lands, despite it not being a card I’ve been too impressed with.

Okay. So just to address the first part of the question, you know, from my perspective, at the beginning of a new format, I’m very much the kind of player who wants to try those, you know, the hard to evaluate cards, like Skyclave pickaxe, you know, the kind of the offbeat ones. I really want to get those reps in to give me at least two to three baseline evaluation reps, you know. Kind of, the feel perhaps, where your brain is really good and I think telling you: “Oh, this card was good or this card is bad”. Even if you can’t exactly put into words why it was good or bad, the important thing overall, if you really are trying to get down to the nitty-gritty of it is, you know, once you’ve got that baseline feeling it’s also much the numbers of how many reps like, you know, five reps, I figured it out. It’s more like, after you got that Baseline try to put into words. Try to explain why you feel a certain way about a card and I think that will really help inform your decisions and that helps you, you know, give almost a thesis statement for a certain card. And then when you compare that to other peoples evaluations, you can really come to like a middle ground compared to where what they thought of it. You know, it’s a saying like, oh, this is a good car or this is bad card. You can explain the why you think it is a good or bad card. So you just have you know, kind of a he said she said approach to it when you’re talking to other people and they’re also sharing their evaluations. Like, who do you trust if one person says one thing and another says another. What about you, Robert? What part of this question speaks to you?

Yeah, so looking at just that first part of the question about when you get a few reps with a card and its the first time that you’re playing with it. The first few times that you’re playing with it, it performs differently from what you expected. I think exactly what you were saying there, where you want to think about the context of that situation that you’re looking at because you’re only seeing it in the context of the other 39 cards in your deck and the 40 cards that your opponent is playing with or just the 5 or 10 or whatever number of cards that are on the battlefield on their side and on your side and that is just a small part of the picture. And so I feel I should be wondering why is it good here? Why is it bad here? Better or worse than what I initially thought? And I can try to internalize that but then, also, I think it’s really important to take that and have that discussion with other people to sort of try to increase your sample size a little bit by looking in saying hey, I saw a Skyclave pickaxe in my red green deck and it worked really well because my creatures had trample or something like that, but then you go on Discord or Twitter or talking to someone at your LGS and they say, oh, I had actually pickaxe in my green white deck and it didn’t do so well. What are the differences between these situations and even playing 10 20 drafts you’re not going to get the full picture of where a card is going to be you’ll get more and more context but it’s still understanding the context around where you played it. And then going to the second half of the question looking at data from 17lands or anywhere else that’s aggregating a lot of data, like this you get a much larger sample, but you’ve often don’t have that full picture that the card was played in. So you’ll see people asking questions like: “Oh, this card is rated as a D by all of the content creators, but it’s the fifth highest win rate of all the commons. What are we missing here?” Like what you’ll often see is that the sample sizes are lower for that than the other cards around it, because people are only playing it where the situation warrants it and so you can thing about it as may be a higher rated card in these specific deck that people are putting it in. or its filling a hole because you had more other powerful cards in the deck. So you have to get both sides of the picture. You can get larger sample size but have less specific context around it. But then you also want to have those discussions that include the context to understand why it’s good or why it’s bad.

Yeah, as always context just plays such a huge part in limited in magic in general. Kind of you know when you’re saying that, the card that comes to mind for me is, Allied Assault, the two and a white instant trick, an up to 2 target creatures get +X/+X until end of turn where X is the number of creatures in your party. So, like I, don’t have the win rate of that card offhand. But, you know, I don’t tend to play that card, but the times that I do play that card, when I have a white aggressive deck that has the potential to get, you know, get three or four party members, like it’s insane in my deck. Like the times I’ve put that card in my deck and it’s been good, oh man, it’s been fantastic. I think it’s like, you know when we talk about card evaluations in general, there’s like kind of two things. With the MDFCs this topic has come up a lot of how good is the actual effect of the card and how often is that effect relevant? Right? And, so like, a more narrow card can still be powerful, something like Beyeen Veil, the negative 2 to all creatures your opponents control or, you know, Vastwood Fortification, that can still be a very powerful effect, but in a narrow context, right? And so those cards that are good in those narrow context will have their win rates boosted a bit more than you might expect.

Yeah, exactly, like you are not gonna put these cards that are combat tricks or anthems or whatever in your 10 creature control deck but in the decks that have 18 creatures in it and that often have a full party. Its gonna be a great card in there. And so yeah, context really matters.

And I think this ties back into the first part of the question to, where it’s like, you know, when you see a card that you didn’t think that was good when you’re evaluating the set and you see your opponent casting, you’re like, “WOW, that was really good there”, right? Ask yourself why it was really good. Get down to the fundamental mechanics of what happened in that game to make that card good, to make it better than I thought. Sometimes the answer is easier. Sometimes just like “Oh, you know a 3/4 lines up with the stats of the set better than I thought It might.” But sometimes it’s much harder and It’s like, Oh well, Glacial grasp, lets take that for an example, in their rogue deck where they’re leaving mana up a lot of time and they have a few trixie threats they want to get in like that’s the context that card is great, right? So you can really swing that back around and really almost like reverse engineer it in your own games. I think almost more than you know, not everybody’s going to be able to get 20 reps with the single card, right? And not everybody’s gonna be able to outsource that to other people. Maybe some people just prefer to play on their own, prefer not to engage in that conversation, but, what you can do, is like critically ask yourself “Why was that card good?” and sometimes you’re not going to come up with the right conclusion, but I think even just getting that thinking going, right? I think that really, really helps. What would you say, Robert, for people that are looking on 17Lands.com at those win rates of cards. Like, how would you suggest people apply those or how would you say that they should take it into consideration when drafting or when choosing to put a card in your deck?

Yeah. All right, I would definitely not use a pick order of ordering cards by their win rate or anything like that. I wouldn’t use it during the drafting portion at all. More, I would go look at that when I’m not playing maybe after I played and say “Hey, this card that’s high on the list asking why is it there?” What do we think that might mean. For example if you look back, in Ikoria, there were some cards that were pretty bad in a vacuum, but pretty high on the win rate list there and it got people asking the question. Why is this here? And then you can either jump to the conclusion yourself or look at the data a little more closely and if you look at a card, like Sleeper Dart, which was much higher on the list then you might expect, even jump to the conclusion that that is played more closely with Lurus as the companion which actually is what we found in the data. So yeah, use it as a tool to think about what that means and why that might be.

Yeah, that’s awesome. I think that that really provides a lot of insight. All right. Thank you, Dylan, for your question. You’ll of course be entered into our monthly giveaway for the question askers of the week. And if you would like to be entered into our giveaway for a free drop from the platform of your choice, you can do so by sending your own question to limitedlevelups@gmail.com or sending to us on our Twitter.

Continue reading

Go to part 2