Player Skill & Deck Quality
-
It's VLS season yet again. I love VSL for so many reasons, but at this point, the top of that list has to be the shared experience of everyone that I talk to about vintage, watching and seeing the same things. Given that opportunity, I want to look at a fundamental aspect of the game that I have come to believe to be overlooked, or at least, vastly under mentioned in discussions about Vintage.
Simply put... these guys are really really good. For the purposes of this post, I will describe my own skill level simply by stating that I would like to think that I'm better than 50% of the players I see play vintage... would like to believe. SO, when I watch VSL, I am in heaven watching not only the sick plays, but the steady tight play turn after turn. Which brings me to my point, quality of play seems to me to be way way underrated when evaluating deck performance in the meta.
We recently had eternal weekend data dropped on us... which was sweet. People went through and tried to tease out what decks were good, bad, otherwise based on those win rates and head to head matchups. I think that's great, and a worthy pursuit... but as one of the people who spent a good deal of time doing that, I've also become convinced that there is a big problem. The main determiner of what wins, in my opinion, is the player rather than the deck. (Do this experiment, if I was playing really hateful Shops against Reid Duke on Storm... would you bet 5 dollars on me because of the match-up, or would you take the Duke of Storms?... I would not bet on me... now what if it was Montolio or Roland Chang on the Shops... you get the point.)
As I watch the play in, I'm reminded yet again, just how good these players are... and that the player pool is small enough that if a handful of the top players move from one deck to another, that would be enough to shift the win rates that show up in these data sets we see. Heck, Rich Shay moving online from Landstill to Mentor represents a tectonic shift in Landstill win rates alone. Reid Duke put up a sizable portion of the total combo wins at champs by himself, if you include Steve Menendian and Kevin Cron on that, those three players accounted for 16% of combo wins in the tourney, while accounting for only about 10% of combo games played... simply put, the one biggest factor influencing the success of a vintage deck is the player piloting it. Yet discussion almost never centers on particular plays that get made, but rather cards that get included or not, and very general win/loss trends between archetypes.
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
Your story, while nice, ignores deck selection as a skill itself. Did Landstill decline online because Shay stop playing it or did Shay stop playing it because it was poorly positioned?
You also haven't provided any criteria for analyzing skill besides past performance, and even that at a small sample size. Did Duke, Cron, and Menendian play better than other combo players, play better decks, have better pairings, or have better luck? It might be that they performed below expectation, when all these factors are accounted for. You already think of these players as skilled, so their performance confirms a preexisting idea you have of them, but it isn't in itself enough to be evidence of skill.
I'm very uncomfortable with the blanket statement that the technical and strategic play last night was "really really good." That's quite an assertion which you haven't supported at all. If you were blown away by the level of play last night, then for the sake of the format I really hope you are not better than 50% of Vintage players. It was fine, but certainly not something to gush on TMD about.
-
@wappla I did leave that out, but for brevity only. (Yeah, look at the size of that post and believe it.) But there isn't any doubt that what I'm saying easily holds up when you consider deck selection even very heavily, because many of the players with the very best records over years don't change the decks they play very much at all. Reid Duke, Sullivan Brophy, Roland Chang, Brian Kelly, Andrew Markiton... and to a lesser extent, Cron, Menendian, Shay... etc.
If the best players tended to fluctuate dynamically between decks depending on what seemed good at the time, then yes, we could say that the best players gravitate to the best decks and then use that as a solid data to determine what the best deck is... but that isn't what happens. In fact the players with the best win rate move only a very little bit between archetypes. Instead they tend to learn one or a few archetypes and stick there, posting above average win rates compared to the average player at that archetype. That's just exactly what the data we have shows...
Now in fairness, within archetypes, single card selection is probably a little more important than archetype selection, and I at least am willing to believe that better players make better choices about those details... but there is no reason to think that they make better choices in terms of archetype... or let me put it to you this way
Andy Markiton - Shops - 22 Swiss points
Brian Kelly - Oath - 21 Swiss points
Rich Shay - Stax/Shops - 21 Swiss points
Reid Duke - Storm - 21 Swiss points
Ryan Eberhart - Gush Mentor - 19 Swiss Points
Kevin Cron - Outcome/Mentor/Combo - 18 Swiss points
Steve Menendian - Outcome/Mentor/Combo - 15 Swiss points
Average.dec - 11 Swiss pointsI mean... come on. What is the most relevant factor here?
Looking at it by macroscopic archetype without at least factoring in who is playing... that's like looking at college football and determining that Spread offence has a higher win rate than Ball Control offences, and concluding that Alabama's offence is second tier... it just kinda misses the point. Or to put it another way, shouldn't we at least, somewhere on a site like this, be looking at what choices these people are making that cause them to win so much more than the average player? It clearly isn't deck selection, I can promise you that.
As for the totally subjective appraisal of VSL play... I still think it's wonderful to see, and have had no idea there even was such a thing as an Oliver Tiu... that dude can play. I love watching that stuff, and I won't speak for myself, but I am positive that the sum of the play is well above the Vintage average.
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
Sullivan Brophy - Dredge - 9 swiss points
Roland Chang - Shops - 6
Nick Detwiler - Eldrazi - 6
Matt Murray - Gush - 3
Jason Jaco - Combo - 9
Sean O’Brien - BUG - 6
Mike Solymossy - PO Mentor - 3
Average.dec 11By your logic these players are less skilled than average?
Calling the play in the Vintage Super League higher than average quality is hardly a ringing endorsement.
-
@wappla Ok, you win... player skill isn't a factor... wait do you even believe that?
And what the heck was Jaco doing playing Combo...?
And isn't the top of our collective bell curve still right of the average?
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
@Topical_Island I'm not saying play skill doesn't matter. Did I say that? I don't think I said that. Where did I say that?
My argument is that justing post a name and a number of swiss points is not adequate evidence supporting the player's skill. Similarly, your simply stating that the level of play in last night's VSL episode was "really really good" does not make it true.
-
@wappla But how can that be your argument when it was my second post? I mean, that is the thesis of your argument? Just posting names and swiss points isn't evidence?... ok stipulated.
Now what do you actually think about the factor of player skill on match outcomes... I have come to believe that it's the most important single factor.
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
@Topical_Island My argument is that your argument is very weak, nothing more.
-
...umm cool. Well I really respect your opinion (that isn't sarcasm). And If you wanted to comment on how much of a factor you think player skill is, that would actually be something I am very interested in...
btw, the average swiss points of all the players that you and I collectively listed is 14.67
-
Yes player skill is highly relevant, but good players also theoretically are choosing the deck that they think is best. At the very least a metagame analysis quantifies the popularity of a deck, and the win % of the deck suggest the popularity amongst the best players.
Its kind of hard to do any sort of analysis on individual plays from a large scale standpoint, unless people are collecting that data. Its already quite a bit to ask of community members just to map out what decks people are playing.
When I write articles, I personally try to emphasize how your personal play and playstyle can alter how good a deck is. I feel like that's usually what a good tournament report does too. I'm not sure what exactly you are suggesting beyond this?
Honestly in my opinion some of the VSL players aren't even good Vintage players. They are just good magic players in general. Its fairly common that they make commentary, and lines that suggest a lack of understanding of how certain match ups work.
-
@vaughnbros But, just to test this in a cursory way... shouldn't we first ask ourselves something like, is the top tier of players playing a seemingly wide or a seemingly narrow array of decks?
Clearly, the answer is that the top... idk, 20% of players is playing a wide rather than a narrow array of decks... it seems to me it might be even a little wider than the total field, since the field is much more shoehorned into 1 or 2 archetypes than the top performing players... even when we include goofy rogue decks.
We can of course, dispute whether that is true, and that's where maybe some data gathering would be good, but if it is, wouldn't that indicate that in game choices and detailed single card inclusions are outweighing archetype selection in terms of being linked to winning?... in other words, player skill is more closely linked to winning than general deck quality?
As to the VSL, I don't know what the difference is between being a good "vintage" player and being a player who is good at magic enough to win vintage magic games...
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
@Topical_Island
I think the top players do play a variety of decks. You bring up Rich, but Rich plays pretty much every archetype. You also bring up the VSL, but the guys that do the best in the VSL from season to season are the ones that actually play every deck, not the ones that play only 1 or 2 decks.Deck quality is an incredibly important component to winning. Take a modern or legacy deck and try to beat a Vintage deck... Or hey try even brewing your own Vintage build and see how it does compared to a build that one of the top players is on. You should be able to see the difference to some degree in how a poorly tuned deck functions compared to what is a standard build.
The reality is that choosing and building a deck is part of playskill. I for one am a terrible actual magic player in terms of the litany of errors I commit in a game. I've had marginal success though because of my deck building. Yet when people pick up my decks they usually have no idea what they are doing because they are so tailored to my playstyle. I bring this all up because its almost impossible to separate the skill of a player with the deck they are playing in terms of success.
The difference between a Vintage expert and a random Pro is they don't actually understand the decks at a high level. They just understand how to play magic at high level. Its like an Oncologist vs a general practice Doctor (or any other specialist vs general skill). Its relevant in terms of this conversation since the VSL deck lists are notoriously poorly built and their lines can be really sketchy, but one of them has to win. Using VSL as an indication of a good deck, or good play is kind of just silly. Some of the VSL guys were or have become Vintage experts, but they are in the minority in that league.
Again, I'll ask what are you actually suggesting?
-
@vaughnbros Again, (This is a little frustrating I have to admit, because this will be repetition of thesis statement number 2 now, and I feel like... well I'll just leave it there.) I think that the most important predictor of wins in Vintage is player skill. That is the thesis. And I for one wish this were factored into discussion around here... at all? or at least I can say more. And I wish there were actual discussion of tactics within the game, or analysis of the thought processes of top players. Because I want to get better, and whimsical discussions the subtext of which is that the DCI are all morons just doesn't get there for me.
Saying, what, that I could beat Rich Shay if I was on Vintage Mentor and he had Legacy Death and Taxes... um, ok? I believe that is true, and would be thrilled to run the experiment. Does that mean that deck quality is more important than player skill? I don't even consider it evidence. Let's try this. Give me and Rich the exact same decklist and turn us lose on MODO for 100 games. Seriously, what do you think the result would be?
How about playing dredge, parallel to Brophy?
Oath parallel to Kelly?
You get the point.Lastly we could just let everyone pick any deck they like and track those results... which is of course, what is actually done, and we see top players winning again and again without changing the decks they each play very much, but at the same time playing a wide array as a group, which should really tell us something about archetype importance vs skill. The names at the top of the leaderboard tend to stay pretty static. The decks those names play tend to stay pretty static. The field-wide win % of various archetypes moves up and down like the sea.
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
@Topical_Island That thesis is not a suggestion of how to change our metagame analyses or approach to the game. Its just a general statement. So what if player skill is the most relevant thing to success in Vintage? Isn't that a good thing if its true?
Rich's lists are tailored to Rich. Sullivan's lists are tailored to Sullivan. Brian's list are most definitely tailored to Brian. I've worked with all of these guys more times than once building decks, and I can tell you that not one of these guys net deck. They all work on their lists, and tailor this lists to be what they think is best.
Rich and Brian are both extremely versatile Vintage players. Did you know that Brian actually first rose to Vintage fame playing hatebears? That Rich has top 8'ed worlds playing Workshops? They both play Gush now because they believe its the best archetype/card in Vintage. If you restrict Gush, they will go find the next thing that they think is good.
You are acting like deck and card choices are independent of play, but that's simply not true.
-
@vaughnbros I don't know if this reference will mean anything to you or not, but have you ever been on 2+2 poker site? It's all about how to play better and make more money. That's pretty much it.
There is way more to poker btw. Like macroscopic stuff like bankroll management, where and when to play, stamina... blah blah blah. But pretty much all anyone talks about is just tactics of maxing out your EV in myriad situations and taking down money. It's all about improving play.
I'm not at all trying to criticize anyone's work on the mana drain at all with this thread. That is, very certainly, not the thesis. For one, I am a total Vintage junky. If somebody wanted to talk about the optimal time to change out their decks sleeves, I'm sure I would be right there. And I love data. I love the data from Eternal Weekend and Extravaganza and all the MODO stuff. There is so much to learn there.
But I ask you, as one self professed medium player, if you want to get better like I do, don't you think there is a lot more that could be said about tactics that isn't? Don't you think there is a ton to be said about tactics that isn't... and more to the point, don't you think the data from Eternal Weekend and Extravaganza and MODO indicates, albeit tangentially, that tactics must be incredibly important to winning, since the same humans keep doing it, but with a variety of decks? That's where I'm focused. I think the deck analysis is great. It's what lead me to this conclusion in the first place.
Brian Kelly Freudian self-analysis of the day:
I like to express the id that a player would feel without the superego of being like, "oh you have to be nice." -
@Topical_Island Of course discussing lines is a good thing. Rich coaches you through nearly every play he makes on stream. So do most streamers... VSL discuss lines of play all the time. Ive seen multiple threads about how to play this opener. So there is definitely content on playing. There just isnt enough Vintage content in general...
If you want true success you need to tailor a decklist to YOUR strengths and weaknesses though. Dont just grab a stock list and expect it to perform as well. None of the good players do that.
-
Hey guys I've been playing Vintage for over a decade and there is only one player whom I consider to be strictly better than me and that is Rich Shay. While other players' success have ebbed and flowed with the changing dynamics of the metagame, Rich has consistently been the best performing player over the past decade and it's not even close. He's Tom Brady and the next guy on the list is Jay Cutler. And if you go back and look at the decks that Rich has played; he always plays the deck that he thinks is the best deck. Control Slaver, GAT, Gush Tendrils, Mentor, Landstill, Stax, Storm, Oath, "Young Americans", and Ravager Shops to name a few. To label him a Shops pilot is wrong. I feel he's most comfortable with Mana Drains/Blue, but obviously he's been playing Shops recently because he thinks it's the best deck and the results speak for themselves.
-
@wappla said in Player Skill & Deck Quality:
Sullivan Brophy - Dredge - 9 swiss points
Roland Chang - Shops - 6
Nick Detwiler - Eldrazi - 6
Matt Murray - Gush - 3
Jason Jaco - Combo - 9
Sean O’Brien - BUG - 6
Mike Solymossy - PO Mentor - 3
Average.dec 11By your logic these players are less skilled than average?
Calling the play in the Vintage Super League higher than average quality is hardly a ringing endorsement.
Pretty sure at least half of that list dropped after being out of contention for top 8. Which I think was 2 losses. I don't play events for planeswalker points, if I'm out of contention for top 8 I drop and go eat (and drink in many cases). Not saying I'm good but be aware those numbers don't mean much.
Edit: And I'm not insinuating that the folks that stick it out until the end are in anyway wrong. It's a noble effort, I just lose my will to win when I in fact can't win anymore.
-
@desolutionist You realize that I'm going to think of you as Jay Cutler for all time now right?
-
@Topical_Island I hear ya. A lot of tactical play that is critical to Vintage success but doesn't really get a lot of treatment (on the forefront of my mind right now is mana sequencing).
A few rambly thoughts:
- Definitely some inevitable feedback and entanglement between tactical play and deck design. For instance, the mana sequencing in my true 3-color Gush decks is leading to some amazing dual selection.
- TMD doesn't have a lot of this discussion of tactics but it's not totally missing. @Smmenen wrote entire books on it for instance.
- One suggestion for everyone is to add some comments on tactical implications of more unique card choices. For instance @wappla (great, great designer if a bit salty) posted a recent Delver list where it took me a while to think through the lack of basic Island (still playtesting it on the draw and trying to figure out whether I choose T1 Delver or T2 Pyromancer). Anyways you can tell I have mana on my mind right now.
- This applies to more than just tactics but I've found myself looking outside Vintage to learn lessons applicable to Vintage. Topics such as mana curve and rate (deck design) and combat, baiting and sequencing (tactics) are just covered so much deeply for Standard and Modern.
