Introduction

This is part 5 of the transcription of the limited level-ups episode 37. “Better drafters mulligan less”, “Playing to the speed of the format” and “Better players are making fewer sideboard moves”. Here you have a list of all parts of the transcription: part1, part 2, part 3, part 4, part 5, part 6.

Transcription

Alright, so those are our are drafting habits, but we’ve also got some playing habits, three playing habits. Have a number five here: Mulligan’s, and man, this one is so interesting to me. You have here: better drafters Mulligan less.

I was a little surprised about this as well. What we’re seeing in the data here is, in best of one, which obviously has a hand smoother, so we’re going to be mulliganing less overall. Our highly effective drafters are mulliganing about 10.5% of their games, whereas the comparison drafters are mulliganing a little over 14 percent of their games. On the best of 3 side of things, they’re a little closer about 21-22% for each of them. But yeah, this sort of illustrates to me the importance of the speed that you get from, in best of one, where games are faster and it’s more punishing to get a mulligan. This is sort of counter to what I would have thought, especially, given the changes with the London Mulligan about a year ago, we actually did an analysis around and, sample sizes weren’t huge because we were smaller at the time, but we didn’t see much movement in the mulligan rates of players overall when the London Mulligan was introduced, despite the fact, in terms of the win rate effect that it has, it’s about 25% less punishing. So I would have thought the better players would have one of adapted to this and start mulliganing more. Perhaps what it means is people are not mulliganing enough to begin with. What are your thoughts on this? Why do you find this surprising?

Yeah, so I mean, you know, I find it surprising because, exactly, you know, the reasons you said. But when I think about it, I almost think, Gil, There’s a lot of things that could be happening here. One, it could be players that are more used to playing constructed were you should just mulligan more hands. But I think, you know, we talked about, you know, risk assessment on the show before and I think… In general, I think players are risk adverse and they see those 2-land hands and with you know, a 3-drop and no 2-drop and they’re on the play and like, “Err. I just got to ship this one”. So I wonder if those kind of hands are being shipped more often. It’s kind of become a running joke on my stream that I keep hands that seem sketchy and I just get there all the time. Like I never don’t have it, right? But a lot of the times like, the hands that just need, you know, one more land to function or you know, just one more good interaction spell or something like that, I tend to keep those hands, I think, more often than people might expect, because… They actually talked about this on constructed resources this week, which, you know, plugging in a constructing podcast on a limited podcast, but they had some really interesting thoughts. And basically they said, you know, when you are mulliganing, they force the idea, that if this hand goes poorly, I might not get to play magic, right? “If I don’t draw that third land, I might not get to Play Magic” And of course the opposite too, “I need to keep this 6 card hand because going to five means I might just not get to play magic”, Right? The game doesn’t care about if you get to play Magic, the game cares about if you’re going to win, so I am more likely to keep a hand with really high upside, that might not get there some of the percentage of the time, but when it does get there, it does incredibly well. Could you try to evaluate just how likely that opening hand is to win you the game, right? Even if the fail case is horrendous, that doesn’t matter, right, what matters is the percentage and that’s obviously a number that you’re never going to know, but don’t look at the lowest of the low as a reason to Mulligan that hand.

Yeah, exactly. And like you said, you want to be thinking about what you need to get there and how likely that is to happen. So if you just need any land by turn 3, and your 75% to hit it or whatever the odds are, then that’s going to be much more likely than if you have a hand full of red cards and white lands and you need to hit a red source to get there, you’re going to be just like less likely to get that because you need something more specific. You want to be thinking about that, and so if you have a 2-lander and only 3-drops, but you have one of each color, then you probably want to be keeping that more often then you might be.

Yeah, I wonder and I’d have to take in and think about this. Maybe, I’m curious, what you think, why do you think that the difference is pretty substantially different in best-of-one and best-of-three?

I haven’t thought about that too much, but my guess would be, just the way that the hand smoother works. You’re given a hand that’s more functional in best-of-one versus a totally random one in best-of-three. But I’m not sure.

Yeah, the handsmoother, you know,that’s kind of what I might have thought as well. I’m new to playing on Arena often, but do we know everything about the handsmoother? Or is there still stuff that Wizards basically hasn’t told us?

No, so basically all that we know, is, that it draws two random hands for you and biases towards the one that is more average on the number of lands that you get. And obviously with the data that we have, we can sort of simulate, not simulate, but we can look at the data to see how often you get a 2-lander when you have 17 lands in your deck or things like that, but yeah, we don’t know the exact workings of it. So there’s still more to be learned there. I think one other thing that probably plays into this, that we haven’t talked about much, is, that better drafters might be building their decks differently and so they need to Mulligan less. So, I haven’t looked at this specifically, but if they have a lower curve, they’re going to survive on fewer lands.

Yeah. That’s a great point, I’ve met before, I‘ve verbalized those words before, but I wasn’t thinking about that in this conversation. That’s like such fantastic point I think! Allright, habit 6 here! We have: “Playing to the speed of the format”. So talk to us about this one.

Yeah. So this is another one that’s a little weird to quantify, but it became apparent to me, when I was looking at plots of win rates of these drafters depending on the turn of the game that it ended. What we’re seeing, is, that in faster formats, the highly effective drafters have a bigger gap in their win rate over the comparison group early on in the game then they did later on. And then on the other side of things, in a slower format, ones that we thought of as slower formats, they have a smaller advantage in their win rate on the first few turns, but then as the game goes on, that advantage in the winrate goes up. So just to quantify this a little bit, looking at M21 as an example, these effective drafters had about 15 percentage point advantage in their winrate if the game ended before turn 9, but if it had ended on turn 13 or later that advantage had dropped down to about 11 percentage points. And then on the other end of the spectrum with Ikoria, being a little slower, with games ending before turn 9, the good players had about a 9 percentage point advantage, but then it increased to about 15 percentage points if the game went on past turn 12. So yeah, so what that sort of says to me is, people should be taking the speed of the format into account when you’re building your deck, when you’re drafting. So, that, if fewer games are going to be ending in the early turns, then maybe you don’t need to prioritize 2-drops as much as you might in M21, where if you get behind on board that early, whoever builds that advantage early, is going to jump out to a huge advantage. In a faster format, you don’t want to be worried about the late game so much and in a slower format, you probably don’t need to be as worried about the early game. You still want to get something on board if you can but it’s not going to as punishing as lower format.

Yeah, the way you measure this is fascinating to me. I think it’s genius, It’s really, it’s really cool! I get a lot of enjoyment out of that! I think that’s really smart. Yeah, I mean, the way that I always used to view the format is, when you have a format, like Zendikar or Ikoria, or even like Throne of Eldraine, where there’s decks at polar ends of the spectrum, M20 is actually another good example… Honestly, every format except maybe M21 this year, where, there’s decks on polar ends of the spectrum, you have like super aggressive, like mono-red decks, and you have like, super controlling, you know, blue-black decks, the ones that play to late game, that you know… When I say there’s decks on opposite of the spectrum, I mean competitively viable decks on opposites end of the spectrum, right? You don’t want to make too many concessions on your deck because, you know, you never know what you’re going to face in round 1 or round 2, right? If you’re playing best-of-three, you can certainly make concessions in the draft, to pick up cards that are good against those late-game decks and good against those controlling decks, so that you can bring them in. But, if you’re playing best-of-one, I wouldn’t tend to make too many concessions to the speed. But when you have a format like M21, when it’s very clear that it is polarized in the sense that you’re going to be facing aggressive decks, you know, 70-80% of the time, I make huge concessions to that, right? I just, you know, I never once thought in M21, “Oh, I need a little bit more top end!”, you know, like I always, just thought, “Alright, I’m gonna take this 1- drop, I’m going to take this 2-drop” in the draft. And you really have to shift your thinking of what like a “normal” draft format is like, because it’s like all of the things you have kind of learned about limited don’t always apply to specific formats, right? And, I think, that’s just a general point, right? The things you know about drafting and gameplay in limited, they’re not always going to be true. Even things that you know, me or Ben, Ethan or LSV and Marshall have said over and over and over again, they’re not always true! And you really have to critically think and critically listen, to you know, both to what you see in game and what other people are saying of when you should be applying certain heuristics in a format and when you should not be. So, in those formats that are very polarized to one end, I definitely like, to take that into consideration, a ton. And, so yeah, I’m glad to see that that actually matters in some respect.

Yeah. I think we’re coming into a theme here of “context matters”.

Yes, alright and our last habit here, is sideboarding. Give us what you have about sideboarding.

Yeah, so this one, I think, is definitely somewhere that I probably need to take into account a little more. So, what we see in the data, is that better players are making fewer sideboard moves. And so, the numbers that we see here are, only about a quarter of a card exchanged per game for the more effective drafters in best-of-three versus about a third of a card per game for the comparison group. And, I was actually surprised about how low this number was, myself, like, I’ve often find myself making multiple card changes per game. So, probably a room for growth for me here. So the way that I’m internalizing this is, maybe what happens to me or drafters like me, is, that you feel like you have to make some change, but maybe I’m just trying to be too clever ??? about the deck that I thought was good for my starting 40, is actually what I should present in games 2 and 3.

Yeah, I think, totally, that’s one of the points that sticks out to me is that if my deck is already great, it’s more streamlined than my opponents deck, or just have a fantastic deck that all the synergy pieces work well together with, I’m much less likely to take out or bring in anything. Because the question you always have to ask yourself when you’re sideboarding is: “Is this going to raise my win percentage?” And if it doesn’t, then why are you bringing it in, right? I think, the classic comparison is bringing in a plummet versus a flyer or bringing in a disenchant versus like a Nahiri’s Binding or 2 or whatever, right? And, I’ve been, you know, pretty vocal about not bringing in plummet against flyers when, you know… You have a streamlined deck or you have an aggro deck that just doesn’t care that much about answering your opponent’s cards, like, yes, it is an extra removal spell, but, sometimes my deck just doesn’t need an extra removal spell, sometimes, it’s just as own thing and sometimes the piece you take out, which might be, you know, a pretty, you know, a pretty unassuming piece, something like a random 2-drop wizard in your wizard deck. It seems like that card might not be providing a lot to your main deck. But, if you have five or six wizard payoffs, it is very painful to cut, you know, even a Fissure Wizard from your draft deck. So, I have just found, myself, in general, when I have a better deck, I’m sideboarding less and I don’t tend to bring in the disenchant or the plummet style cards unless it’s like a very high value target. Because if I’m plummeting my opponent’s 3/3 flyer, their living Tempest or something, like, I’m still not that up on the exchange. I Guess I get their card, but that card wasn’t going to affect the game overall. If it’s something like a bomb, like Drana (,the Last Bloodchief), if it’s something like Felidar Retreat, yeah, I’ll bring in the card because I think having access to that card does up my win rate, but I don’t think answering their Expedition Diviner or their other medium flyer, will up my win rate a lot of the time.

Yeah, I think another thing that you have to keep in mind, is that they’re not always going to draw that card that you’re bringing in some sideboard answer to. And so, if you draw your plummet and they don’t draw their air elemental or whatever you are bringing it in for, then it’s not really going to be there. Because, like, you said, you’re not bringing in some sideboard card to take care of their C-level card just because it can. You need to be bringing it in for something that’s high value and potentially even that they have multiple targets that you’re going to be able to hit with it.

Yeah, of course and I think just overall, something I consider in sideboarding as well is just being able to change my plan or really just to my opponent’s plan. That’s when I’m more likely to sideboard. If I have cards that can really change the texture of my deck to better line up against my opponent’s, as bringing in more 2-drops against the really aggressive deck or bringing in more card draw spells against a grindier deck, right? And those kind of switches are switches that I think are more effective in more games than those one for one swap outs, you know, random 2-drop for the plummet. But at the same time, I’m less likely to change the texture of my deck overall, then If I do have a great wizard deck or a great rogues deck in the first place. So that leads me to sideboarding less than I might have in the past even, when synergy decks were less common and you just didn’t have these streamlined decks. I think in the past, maybe even five years ago, you did end up in a more mid-rangy pile that you could… that were more modular in games 2 and 3, right? But I don’t think that you end up there all that often. Now, that being said, when you do end up in those mid-rangy piles, I do tend to prioritize those cards that can transform my deck in games 2 and 3, because when you’re playing that mid-range role that is your goal, to line up favourably against what your opponent is doing. So, I think that’s an important part of that equation.

Yeah, and in all that data that were looking at here, we’re just looking at draft data, but I’d be interested to see, if we looked at sealed pools, how often people are making those sort of full on swaps like you talked about, to have a whole deck picture that matches up much better against what your opponent is doing.

Yeah, especially because your decks are less streamlined in sealed too, right? So you kind of… it really fits into that story really well.

Yeah. So those are the seven habits that I found that there were differences that I wanted to talk about here today. I think one other thing that’s interesting to me is to also look at a few of the things that there wasn’t a huge difference between these two groups and curious to hear your thoughts on these…

Continue reading

Go to part 6