Drafting the Right Way

Drafting and Limited in general are not easy to play. You need to be knowledgeable about a whole set of cards – from chaff commons to mythics and in some cases – even fringe cards on some elusive bonus sheet. But you also need to be a proficient deckbuilder – building as best as you can during the draft already and then being able to make last few cuts. And you need to be able to read the signals during the draft to navigate your path as best as possible with confusing noise mixed together with meaningful signals. And if that was not enough, you need to play the resulting creation against decks that could have literally anything in them from the set, unlike constructed formats where most metagame decks vary by only few cards from each other.

The effort is well worth it as what you get by playing draft is the access to the creative, improvisational jazz of Magic as opposed to impressive but rigidly-structured and technique focused constructed formats, which are more akin to symphonic orchestra performance than a jam session in some noisy New Orleans basement. But don’t be fooled by the improvisational character of draft. This doesn’t mean you can’t learn to do it – the process of getting there is perhaps different but improvement in draft is relatively easy to achieve, especially the initial steps. The further you get, the harder it is to make further improvements.

But I think many, even heavily invested players, could improve their win rates and this article will describe a very simple tool that can help you diagnose your drafting style. And since knowledge is half of the battle, you will be in a good position of correcting small issues with your play style thanks to the advice you will find here. But before we can do that – we need to go back in Magic history. And go back quite a long way. A whole decade.

Drafting the Hard Way

In 2013 and Ben Stark wrote his seminal piece called “Drafting the Hard Way”. This laid out a new way of looking at the draft strategy. Ben’s concept is simple at its base: if you are flexible in your draft and pick the best card in the pack in your first picks, you will be able to pick almost any lane later in the draft when you see it open.

But that simplicity, despite contributing to how resonant the article was in Magic community, is an illusion. The concept of drafting the hard way is laden with complex questions and spawned a whole sub-genre of content articles. At the heart of it is a central question: how hard should you draft. To formulate it differently: how agnostic should I approach each pick. Should I evaluate card power in vacuum? Should I weight it against the limited metagame?

Drafting with Preferences

This led to many excellent takes, one I will single out comes from Ryan Saxe in the form of “Maximizing For Your Own Preferences” article, which looks at the perils of evaluating cards in vacuum, without taking format specifics into account. A trap that many players drafting the hard way fall into. The gist of the concept is that given a choice between two cards – you can chose the better card in vacuum or a card that might be weaker in vacuum but leads to a potentially better deck, or a deck that comes together more frequently.

Buy on TCGplayer $0.04 Buy on Card Kingdom $0.35 view card details

Buy on TCGplayer $0.05 Buy on Card Kingdom $0.35 view card details

To give an example from LCI: if you have a choice of Chupacabra Echo and Oaken Siren. In vacuum, Echo is stronger card. It has a Game in Hand win rate of 59.5% compared to 58.2% of the Oaken Siren. But decks with Chupacabra have a win rate of only 52% if they don’t draw it, while decks with Oaken Siren have a significantly higher win rate if they don’t see Siren: 57%.

In this scenario you are faced with a choice of better card or better potential deck. And it is hard to determine what to do. Sometimes one pick will turn better, sometimes the other will. This is where your personal preference and play style might come to rescue – if you like playing with slower, grindy, controlling decks, Chupacabra may behave better in your hands than in the hands of an average 17Lands.com user, if you prefer tempo-aggro strategy, Siren may be your pick. More importantly, some players prefer drafting in a way that keeps you open to different lanes – in that situation Chupacabra is a better pick. Some like to draft in a more assertive way, making the decision fast and locking in on preferred archetype as soon as they can.

Preferences like this lead to very specific problems. But before we dive into them, we will go on a tangent. Don’t worry – it hopefully makes sense afterwards.

The Mystery of Trophy Rates

The idea of trophy rates mesmerised me for some time. Trophy rate is the percent of drafts given player trophies with – so in BO1 it means getting to 7 wins, in BO3 draft, going 3-0. As you can imagine, it is very strongly linked with the player’s win rate. The higher the players win rate, the more drafts they trophy. There is a but though. Looking at 17Lands.com leaderboards you can quickly see that players with very similar win rates will sometimes have very different trophy rates. And it always puzzled me why is it so.

The mathematical answer is simple. Win rate is a collection of the results of all your drafts. But in that collection you will have slightly different distributions of trophy rates, because you will have different distributions of draft outcomes. For example, this is the distribution of draft outcomes for a player with a 60% win rate if the results of the games would be perfectly random would look like this:

6.4% of the drafts would end up 0-3, 11.5% – 1-3 and 23.2% would end up with a trophy – a 7-2, 7-1 or 7-0. But if there are differences in the trophy rates between players with the same win rate – that means the distribution doesn’t always look like this, which means this profile will look different for every player. How can they change? I will show you to examples from the opposite sides of the spectrum.

This hypothetical drafter also has a win rate of exactly 60%. But they only trophy 6.7% of the time – once every 15 drafts. But they never have a terrible draft – never goes 0-3 or even 1-3.

This hypothetical drafter has a win rate of 60% too, but they trophy 31.3% of the time – almost once every 3 drafts end with a trophy. But there is a cost – roughly every 3 drafts they will also go 0-3 or 1-3.

Those are extreme cases and real data differences will be slightly less spectacular, but extreme examples are here for a reason. What is the difference between those two players? They both win the same but they get there in a very different fashion.

The Hypothesis

I have been thinking about this phenomenon for some time already but never could have put my finger on it, never found a good way to put my ideas in a cohesive form. Until quite recently. I was looking at personal data of Pasta Pirate, a prolific drafter and streamer and noticed that his trophy rate was higher than I would expect by chance – finally I could analyse private data of someone who was on one end of the spectrum you see above. And my instant hypothesis was – he forces a lot. And in this particular case, the data was almost instantly confirmed. Out of ~1400 games he played in Wilds of Eldraine, 1000 were with RB Rats archetype. Many times more that you would expect from a balanced draft strategy.

And when you think about it – it makes sense. If you force a color pair, some amount of time the color pair will be open. When it is – you are ready for it from pack 1 pick 1. This is a big advantage – you will be more likely to trophy in that scenario. But if your color pair is not open and you force it without a good plan B, it is a recipe for disaster – so you will be much more likely to go 0-3 or something along those lines. Your win rate can still be good, but you will have wildly different outcomes of drafts.

On other hand, staying open for a significant portion of a draft will also carry a penalty. At least some of your important early picks are by design wasted – you will not play them in your deck for the sake of future opportunities. And in a league system like on Arena, when you are playing outside of your pod and against people with similar record to yours, your good deck, drafted the hard way, will be pitted against those lucky decks that were forced from P1P1. So you will not trainwreck often, but you will also trophy less.

The hypothesis then is: there is an optimal ratio of staying open and forcing. And the more you stray from it, the more you lose – either in win rate or on EV of the drafts you play.

Diagnosing your drafting skill

How to know if you are forcing too much or delaying the decision to pick a lane for too long? I would start with an introspection. Think about it – what do you suspect? Understanding ones own shortcomings is a key part of improvement. I always knew I tend to err on the side of forcing and am painfully aware of it. Relearning Limited during the bot phase of Arena certainly was not helpful there. But over time I learned to draft the hard way and currently I do much better than before in this department. Ask yourself, where do you think you lie on this spectrum and good chance you already know without the need of confirming that with any data.

But of course after self-diagnosis it is good to validate your findings. In order to make it possible, I plotted top 500 most prolific drafters on Arena that use 17Lands.com and plotted their win rates against their trophy rate. And the trend is quite clear.

The plot shows that the more you win, the more you trophy and that there is a clear central trend. The more above the trendline you are, the more on the forcing side of the spectrum you are. The more below the trend line you are – the more equity you lose by staying open for too long. If you are a 17Lands.com user – you can easily check your all time win rate in BO1 and your all time trophy rate in BO1 and put yourself on this graph. Where do you land?

“But I am a BO3 drafter!” I can hear you scream out in irritation. Fear not. Just do the same but on the graph below:

And if you are wondering who that solitary point with a win rate over 80% and a trophy rate deep into the 50% territory is, well that is the current World Champion, Jean-Emmanuel Depraz – and as you can see he is perfectly on the trendline, suggesting he is capable of judging exactly when to stay open and when to make a decision. And doing it while having a win rate inaccessible to others.

What next?

If you managed to diagnose your position on the graph – what comes next? Hard work, if you want to improve your drafting style. Actively stopping yourself from delaying the decision if that seems to be the problem. Making sure not to draft with blinders on, if forcing is your problem. Start actively looking for alternative picks and consider them more thoroughly – that certainly helped me to reduce that habit. And remember that each draft is individual – sometimes you will stay by your first pick because that is how the pod is aligned, just make sure that you don’t dig yourself into car crash drafts by sticking to the initial picks too much.

If you are drafting too hard way, remember to assign weighting to the colors based on their power in the format. I noticed this complaint a few times already – where people who stay open in the draft mention that they often end up drafting the color that is considered to be the weakest in the set – of course you will if your priority is to just find the open lane. The lane with the weakest cards will be open most frequently because people tend to avoid it. Adding some additional bonus for being in the right color in your individual card evaluation is a good way of avoiding this trap.

And most importantly – be inquisitive and thorough, but don’t put unrealistic expectations on yourselves. Self-applied pressure lead to burnout of many players. Tis has nothing to do with mu curves and drafting theory, keep your mental wellbeing at the forefront of your actions – in the end we play Magic to have fun and not to be miserable.

If you want some extra information on the topic or just want to listen to my dulcet tones – it is available as a seminar on youtube, with some additional tips on how to wheel cards and draft specifics findings in LCI: