Sullivan Library - Winning from Losing
As I write this article the calendar still says "2008" and I’m looking forward to an incredible New Year’s Eve. For some people their favorite holiday is Halloween. For others it is Christmas. I’m sure for others it might be anything from Saint Patrick’s Day to Valentine’s Day to Thanksgiving to Easter to Talk Like A Pirate Day to what-have-you. For me though it has always been New Year’s Eve.
New Year’s Eve is the night to kiss a beautiful stranger (or twelve) to enjoy a fantastic night with most of the rest of the planet and to stay out later than you should. It’s a good time to think about the future and reflect on the past. Historically I’ve had a lot of fun on New Year’s and I expect this year to be much the same.
This year I’m actually really proud of a lot of the Magic accomplishments I’ve been a part of. I haven’t really had the time to truly go to the appropriate number of Pro Tour Qualifiers and otherwise do the work that is necessary to get on back on the Tour but nonetheless I did go to two larger invitation-only events and make a number of really great decks including the Elf deck that Sam Black used to qualify for Worlds. There are a number of articles that I’ve put out there that I thought were helpful. My article on created amalgamated decklists might not have explicitly been used by Frank Karsten at Worlds this year but it laid out the concept that he employed. The concepts I laid out in The Strategic Moment have a kind of value that I think will be useful to Magic players for as long as the game is around. Similarly Deck Discovery and Collective Intelligence in Magic does a good job of laying out the reasons for why Magic tournament play is fundamentally changing and what you can do to stay with the rising crest that is quickly washing over the scene. Heck at the beginning of the year I had already begun to bring up questions that challenged "results oriented thinking" in the section "The Problem with Success" in one of my articles.
It’s the converse of that question that is also really interesting. If there is a problem with success then…
What’s Good About Losing
The answer is a lot.
Losing is something that you absolutely don’t want to do some of the time. You don’t want to lose at all if you’re playing for something. When a match with something of value is on the line this is not the time you want to be losing whether it is in FNM or the PT. By this point what you really want to be doing is winning.
The big thing about losing is that losing provides you knowledge. Even if you’re unfortunate enough to be losing at an actual tournament it can provide you information in a way that winning cannot.
Winning can provide more evidence to verify that something is correct. To actually know that a card choice or a play is correct though often takes tons and tons and tons of data. Each little bit can add more to the pile but any individual win doesn’t get you very far. Losing on the other hand is much better at telling you that something is wrong than winning is at telling you something is right. Your conclusions might be wrong and might be corrected by dipping into the question further with more testing but it often doesn’t take very many games to show you that a deck is up to snuff whereas proving that it is takes a lot of wins.
Winning can trick you into thinking that things are "good enough" but losing allows you no such illusions. Losing demands that you get better before the big day when an actual tournament happens.
If all of your testing is doing is providing you with wins this doesn’t ask you to improve the deck you’re working on. This can be a good thing if you’re at the point in testing where you’re finishing polishing up the deck. Usually though decks aren’t at this stage.
It’s called playtesting for a reason. When you’re testing your goal is to discover what the deck is capable of. Your goal is not winning. Your goal is learning. If you get to the point where you’re constantly winning that is good news. But the process requires losing.
If you’re merely trying to verify if a deck is good as is then winning is good. If you’re merely trying to practice then winning is good. But if you’re trying to blaze a new trail you need losing. Losing is the thing that tells you that you aren’t there yet. Losing is the thing that can help provide you with an understanding of your weaknesses.
At a PTQ I spent a short amount of time playtesting with ICBM team member (and Sullivan Solution pilot) Tommy Kolowith. Tommy was beating me to a pulp with Dreadstill and I couldn’t seem to buy a game. Even games that looked like they were going well would just fall apart and swing back to him.
"Man Adrian you’re a what’s the word? What’s the word?" asked spectator Owen Turtenwald. "A masochist! That’s it. You’re a masochist!"
In the end though it’s just Rocky Balboa with Magic cards. You take that beating and you take that beating and you take that beating and in the end if you’ve done what you should when it counts you can win.
If you look at a deck like Sullivan Solution (or pretty much any innovative deck) it doesn’t just spring off the vine full-grown. One of the first things that has to happen is that germ of an idea throwing it against something that you know is actually a good deck and seeing what works. As I was prepping for Hollywood I would throw down again and again against Owen in many matches but would almost inevitably lose. His Elf deck (the beatdown monster that would provide the shell to Yurchick’s virtual Top 8 finish at Hollywood and my own Elf list that Sam Black took to U.S. Nationals) kept hitting me square in the face. It became my default test deck because it was clearly so brutal even if I was aware that there were aspects of it that weren’t standard. In the end it made me discard many many decks as not up to snuff but it also helped me refine my Red deck into one of the top Red decks in Hollywood. (An aside: heck I even gave everyone a two-month preview of the deck that Sam would play into the National Team.)
This is one of the things that you have to do when you’re losing: you have to understand whether or not a deck is worth pursuing or not. You can lose again and again and again and again in testing but still have this be the deck to play if you are getting a decent EV against your expected field. Depending on the prevalence of what it is you are losing to that losing can give you an idea of how much you should expend on trying to win a matchup. If the EV is not as good as simply playing the "best deck" that you otherwise have you generally better abandon ship (unless you’re looking for payback in some form other than wins – fun for one). Also if you’re losing enough it can also tell you whether or not it is simply time to give up on a matchup and tweak the main and/or board to have more slots for more winnable matchups.
The larger the field the less important it is that you lose to even a prominent matchup in game 1. A 30% game 1 win only requires a 60% game 2/3 to turn around to better than a coin flip; in a matchup like that if that deck is only twenty percent of the field (like say Five-Color might be) trying really really hard to win a game 1 is just silly. If on the other hand that deck is Faeries during Block season you’re going to have a lot more value in fighting harder in that game 1.
If you listen to your losses these are the opportunities to improve your game your deck choice and your chances to win when it counts.
In innumerable MTGO drafts I’d been forcing Five-Color Esper/Grixis lately. I usually would destroy most of my opponents. Unless they were an Exalted beatdown deck. Then nearly always they would largely destroy me. I wasn’t 100% confident that this was a bad matchup but the evidence looked good. Then at the recent Minneapolis PTQ that Brian Kowal took down I lost to his eventual finals opponent Kevin Delger. Kevin had the best deck in the draft a deeply aggressive Exalted deck that splashed into its Blue to be Bant-y.
Kevin ran over me. I think I died on turn 6 in both games. My own deck was a weaker build of the archetype (though still quite strong against a less fast opponent). I knew that there was a weakness in the archetype to Bant and this loss certainly showed that weakness. Perhaps more importantly though the loss highlighted one of the things that I had failed to take into account when I had been preparing for events.
I didn’t have an appropriate curve to handle an aggressive deck. Normally when I’d be drafting on MTGO I’d have that wonderful screen up showing me my curve. I would know that I needed to give more weight to a card like Tidehollow Strix. In this draft though I was just relying on my memory. Because of how busy the rest of my life had been I really didn’t have the time to do in-person-drafting and the very very few that I had we had definitely played peeksies on our deck in the midst of the draft.
Now obviously if you pause to think about it this isn’t good prep for a PTQ or GP but it is often fun to look at your cards to show a spectating friend and just see how things are going. But what I didn’t have was the practice of really remember what my deck looks like and knowing intimately what I would need pick-after-pick without reminders. In the past when I had applied myself more to the draft format in question I would be so familiar with the draft and what I felt like I should do that I could afford to not have every fun live draft be practice. I would know intimately where the deck should go. But this was not the past. This was now when I’d spent much much less time drafting over the past several months and I really didn’t have that intimate knowledge.
Even if I may have been able to get away with less rigorous drafting in the past and have it suffice for my preparations I learned that I could not do that now. In my next live drafts I now knew that I couldn’t do that. I had to be more prepared to not look. Where in previous formats I might have opened hundreds of packs I had to realize that I was nowhere near that for Alara and not kid myself into thinking I was at a level that I might have been in previous formats (where I was that deeply prepared).
Losing preps you for tomorrow far more than winning might. You can get a lucky win and not get a takeaway from it. Losing (and the push to not lose tomorrow) is what arms you to win. If you are afraid to lose in playtesting perhaps because you are afraid to be embarrassed by the loss or for other reasons you rob yourself of the chance to become great.
Winning is a path to being good. As Patrick Chapin said though if you really want to get to that Next Level you can’t be afraid of failure.
For me the path forward this year is probably going to involve a lot of Magic. I hope to go to as many Grand Prix events as I can afford. I plan on trying to actually get a bunch of cards for MTGO (I never really built up my collection and it is so full of holes I could strain pasta with it). I have a pretty large hole in my "real life" schedule; if I get into grad school in one of the programs I like things won’t start until the fall and so I’ll have a lot of time to spend casting magical spells and summoning fantastic creatures. There’s a whole lot of Magic I want to do. I’m excited to see what happens when I’m actually expending the kind of effort I used to expend on Magic ten years ago.
From the Mailbag
Beginning this year I’m planning on bringing up some of the letters that I get in the mail each week for further discussion. Sometimes I just get such great mail that I think it would be great if more people could read it than me.
From Chris forum user Noodles2375:
First of all I really enjoyed this article. In particular I thought your analogy on the use of polar/Cartesian coordinates was particularly apt. I wanted to share a little anecdote with you about this but I didn't think it warranted a forum post.
During World War Two Richard Feynman and a team of physicists he was working with were at Los Alamos (before they were actively working on the Manhattan project). His team was working on a mobile artillery piece and to avoid annoying interjections by one of the Lieutenants who oversaw their work they feigned ignorance about the overall project and answered only questions about the details they in particular were working on.
While working they had decided quite reasonably that in terms of measurable variables it would be easiest to work out the firing tables/aiming mechanisms in terms of a polar coordinate system with the artillery piece at the origin. Making the assumption that the gunner and spotter were both at the same location (the center). One day the Lieutenant came around and asked "What happens if the spotter and gunner are displaced?"
The question nearly dropped their jaws. Computing firing data from a displaced observer is really annoying in a polar coordinate system and while the Cartesian system was more difficult on the design end it allows you to account for the displaced observer rather trivially.
At any rate your comment in the forum reminded me of this story and I thought it was a neat illustration of the point you were making. Keep up the great writing!
First of all thanks Chris I definitely appreciate the kind words.
I think that the thing about your comment that is really illustrative is that there are often more than one way to approach a problem in general even in Magic. I’m not privy to the design-making process that might go on in the heads of some of the very very top minds in Magic play of all time. I’ve probably played many hundred hours or more with Bob Maher and I know that some of his skills have rubbed off on me (some small small part maybe) but I also am fairly confident that his process is probably a very different one than mine. I might not have the right kind of brain to be able to actualize his process in a way that can be effective.
Similarly I thought about this e-mail as I talked to one Michigan player Eric "Dinosaur" Taylor and as I reflected on a conversation with another Patrick "I-Man" Chapin. With edt we had gotten into a heated conversation on the nature of empirical play and theoretical play and the strengths and limitations of each. While I think we both agreed that each had value edt put a much stronger weight on empirical play. I’ve already written about my own views on the difference in value between the two but at essence what it boiled down to was that with both elements one could find the way to the right play though in certain situations one or the other would provide the better shortcut. The Chapin conversation was a fundamental disagreement on when to count card advantage – ah theoretical technical disagreement! – but in the end it all boiled down to the same thing the difference between Cartesian and polar coordinates both finding the right answer but having a different process (with different strengths and weaknesses) to get there.
Thanks for the anecdote Chris. It’s a pretty awesome real world example of what I’m talking about.
A quick parting question for everybody: do you think that I should share reader mail every week (or nearly so) or just at times? Thanks for your input!
Until next week…