This kind of data is infinitely more useful than looking at individual cards in a limited environment like your other articles. I'm sure this will be very useful.
I'm curious about what tournaments you used to gather your data. How many tournaments did you use? What is the date range of these tournaments? This is kind of important information to have when judging the usefulness of these kind of statistics because decks tend to evolve in ways that would change their overall win percentages over time.
I've been playing magic and reading articles for quite a long time now, and I have to say this series is one of the most interesting I've ever seen. You've done a fantastic job gathering data and putting it to good use.
While the descriptive content of the articles is fantastic, I'd like to see you do a little more with the data. Specifically, I'd love to see you take those normalized statistics from the second half of the article and show us what happens if you push the makeup of the metagame in a number of directions. What happens to the data if you shift 5% of the metagame back to Jund? What about 10%? What happens if you shift the metagame in favor of UWR control? I think looking at these numbers can help us quantify exactly how subtle and not-so-subtle shifts in the metagame affect the actual viability of deck choices.
well ill be honest when i opened the article i was looking forwar to Grave Pact and Death Pit Offerings with zombies and such...really a fun deck there, but i like what you did. The deck seems great if not a littlee bit on the pricey side this time around.
Oh, but to be fair with the time thing, I at one point calculated EV/Hour, and it still turns out that it is better to play 84's than 4322's, it just happens sooner, if I recall that, 4322's and 84's become more attractive to someone who is worried about ev and time (something like 5% points I think)
I dunno how I missed them. Basically, you need some sort of artifact or enchantment hate (in the case of leyline), or a pithing needle, or to move very quickly (before they can resolve helm), or Gaea's Blessing to beat Painter/stone.
I disagree with godot on the time thing, 4322's on average are far shorter than swiss, due solely to the fact that they are single elimination.
Assuming on average 40 minutes, rounds, and 25 minutes to draft it works out swiss = 145 min, 4322 = 25+40*loss+80*win*loss + 120*win*win.
since loss = 1-win and win is between 0 and 1, you can see how on average a 4322 can never take as long as a swiss (unless of course you have a 100% chance of winning... but I don't think you'd be complaining then)
Assuming you have something like an unlikely 75 win percent against the filed (which would be high) you save ~30 minutes in on average.
Agree with godot on rare thing.
Agree with godot on the math thing, for handy reference (I don't know if I'm allowed to post external links, please edit it out if I am not)
here's a spreadsheet of your ev in various formats (not taking into account increasing/decreasing difficulty as rounds progress, as it's a spread sheet. maybe I'll program up a terrible gui). I've posted this before but it has been simplified and the number of formats it calculates is larger.
I think time, convenience(ie it was firing), and practice are the only reasons to draft 4322's (well and probably superstition... did you know a university ran the numbers in football, and it is almost never more profitable to punt than it is to just go for it on the fourth down? Yeah, it's kind of like how teams almost always punt (no pun intented))
Time, convenience are self explanatory.
By practice I mean (assuming an increasing level of difficulty as you progress up the queue which as a below post states might be faulty). In that it is harder to go from drafting with "bad" players to "great" players, than it is to go bad->medium->great, so you may be using 4322's to get a feel for the format to gain access to 84's. As well you may be training for a paper tournament, in which case learning the format on a scaling curve might help.
That said I don't think practice constitutes drafting only 4322's... it should just be a temporary thing.
side note: I'm one of those mid 1700 players in the swiss... just for some reasons I trust my math and know that since I'm less than 50% to beat an average 84er
I could be wrong about this and it is why I wanted to elicit some opinions from other posters. I am right around a 1700 player and don't feel I am good enough to draft 8-4s and win consistently enough conserve money. So I stick to the 4,3,2,2 and the Swiss. Reading these post I feel like people might be missing something. I don't think the 4,3,2,2s are necessarily stocked with better players then the Swiss. On the contrary I think the Swiss likely has a higher average ranking among it's players then the 4,3,2,2. I can't be sure obivously since wizards nolonger lets us see our opponents limited rankings, but just from playing them as often as I do I feel the average level of competition is better in the Swiss draft.
I have actually performed better over time in the 4,3,2,2.
As for reasoning, I think that Godot hit it on the head. Swiss pays out better and has a better "EV". This means that alot o top players tyring to go infinite will go for either 8,4 or Swiss. Look a Godots own numbers 17/25/1. If other players of a similiar caliber have similiar splits, then the it seems much more likely you are going to run alot of high level players in the Swiss and less so in the 4,3,2,2.
Heck Godot's reference to the Math article is flawed since it assumes better players in the 4,3,2,2 then Swiss. However the fact that the "cat is out of the bag" on the EV thing means that top players are likely avoiding the 4,3,2,2.
Just a thought does anyone else feel the 4,3,2,2s have lower average skill the the Swiss??
Since you have to play the bellow to win, if he blocks the larva with the scorpion (since he's dead if he doesn't anyways) he in fact blows by playing it in response.
though more likely if he had disfigure he would just kill the larva, with the trigger on the stack, and you can pump it...
Ok, what I'm saying it that disfigure is the trickiest card to play around there if you want to win that turn.
But yeah, I agree with your assessment that he likely didn't have removal, so you really only have to play around the non removal tricks.
And so? Is there a bias on some archetype for that? So let's not talk about statistical relevance, just about facts we can witness. Take this as the results of an extensive, days long testing session
* The time thing is overrated. The only round where you are waiting more than you would in single elimination is round two, and even then, sometimes you are in the slowest match.
* I don't feel there is quite the raredrafting gap between Swiss and 43 that you see. While I obviously haven't drafted a ton of Zendikar 43, I end up drafting plenty of 43 in the nix tix events where it is all that is offered, and drafted a ton of 43 before Swiss was available, and don't feel like there is a huge difference between the level of raredrafting in 43 and Swiss. I admit this is just my sense, it would take some data crunching of a large sample size to prove this either way.
* The raredrafting scenario you mention works both ways. You cite the Sphinx you should have had for your blue deck. What about when a guy two seats upstream from you in pack 3 in BR, takes a Sphinx over the Hideous End, and the next guy in WU passes it to you P3P3 for your BG deck? You are more likely to get potent on-color commons and uncommons later than you should because of raredrafting than you will be denied rare bombs, because frequently the raredraft will not be for a limited bomb, but for a fetchland or something.
* re: variance: I don't reach the same conclusions as you do on this front. For one thing, when you advance in rounds in 43/84, it is generally to face better and better decks. If you lose in the first round of Swiss, you generally go on to face worse decks. Also, see Shaterri's article about the value of raredrafting for a statistical backup about the flatter payscale of Swiss not making 43 a better bet. It simply doesn't make up for that missing pack, or for the fact that the road in Swiss sometimes gets easier instead of harder.
It is a very mathy article, but here is a quote regarding the mathematical conclusion drawn about the different queues:
"Comparing EVs for the different queues reveals something interesting: the 4-3-2-2 queues are never the 'best' choice. If your win percentage against the field is less than 50%, then the Swiss queues are your best bet; if it's better than 50%, then 8-4 becomes better than 4-3-2-2. In fact, this still holds if we skew the win percentages to account for different opponent strengths: if you assume that your win percentage against the average Swiss opponent is 5% better than vs. the average 4-3-2-2 opponent, and 5% better against the average 4-3-2-2 than the average 8-4 opponent, then Swiss becomes your best bet up to about a 65% win percentage vs. the field (and thus a 55% chance against the 8-4 opponents), and above that 8-4 takes over again. In fact, Swiss and 8-4 queues still dominate if you assume a 10% strength difference — all that changes is the crossover point."
Finally, on general *principle* people should avoid 43s because of the lower payout. "Sure Wizards! You guys keep an extra pack of profit for this draft, we don't want it. Think of it as a tip from the Magic community!" That's what everyone who enters a 43 queue is saying...
Ironically: the guy with cheap aggro beaters and removal, who's basically playing to lay a few fast threats and keep you from ever generating a blocker, is "okay." The guy who plays a hedron crab and removal, basically trying to keep you from bashing his face in long enough for the crab and trap mills you out... gets the abuse.
But on an objective level, *neither* deck is really any more or less interactive than the other. If you want real non-interactive: play combo.
Which is where the real irony comes in. If your combo involves casting scapeshift to bring in 6 mountains + 2 valakuts, you're "cool" (judging by Scapeshift's popularity in Extended.) But trade those two valakuts for two hedron crabs, and you'll get the hate. Even if the deck is less effective because of that mill approach.
to be honest i was not a big fan of the limited articles but this is great, it so much easier to simply compare w-l records for me and i can understand where the % comes from. Good job!
Just wanted to chime in as another guy with a 1750+ rating (currently 1775) who drafts exclusively Swiss. I like the fact that I get to play 3 matches every time.
I would note that I've drafted ZZZ about 15 times or so, and I've only once played against someone with an equivalent rating (as noted by the fact that my rating changed 8 points after the games. Typically I gain 3-5 points for a win and lose 11-12 for a loss).
I imagine this means that my skill level won't be increasing all that much, but I feel it's worth it, because I really hate drafting, losing in the 1st round, and then not being able to play anymore.
My experience: I am 17-9 with Vamps on MODO (Standard queues and one 4-0 daily event), 50 tickets net gain. Against Jund my record shows 2 wins and 5 losses but I don't feel that the matchup is so bad, I got quite unlucky at least twice and I think that the record will improve in time.
White weenie was good in Mirage/Tempest Standard with shadow guys, cataclysm, tithe, and Empyrial Armor. Was that really the most recent time it was good? Sad.
At the time of this draft, the ZEN Duals had already declined in price enough that the major buy bots wouldn't even buy them from me anymore. I did end up getting about a pack's worth of value out the Verdant Catacombs, but it took some time to find the right buyer. At the time, I seem to remember the Scalding Tarn not being as valuable (although I could be wrong about that - currently, they are valued at MTGOTraders equally); furthermore, since there was a strong pick in its place (Burst Lightning), I never really considered it.
I like the inclusion of test of hypothesis and really appreciate the confidence intervals for the win statistics. Out of curiosity, do you think it is important for the audience to understand the use of both CIs and win rankings to evaluate the winning-ness of decks?
Second, the factors some of you are leaving out in your evaluation of 84 vs. 4322 vs. Swiss are time, variance, and opportunity. Yes, the total payout in 4322 is less. Yes, swiss can be very valuable in providing a larger sample size to evaluate a given draft/drafters skills. However, in terms of total value 4322s have some benefits:
Time - Zendikar is a pretty fast format, but draft + 3 rounds still often takes 2.5 hours, and some formats take much longer. If a draft goes badly awry - actually MORE likely in swiss where everyone rare and hate drafts more than they should - I would rather not be committed to fumbling around trying to pick up an extra pack for the next couple of hours, and be able to jump in another queue and try my luck and skill at balancing out my losses. Of course, I refuse to multi-queue out of common courtesy.
Variance - Variance is usually argued as a reason to play swiss. After all, you can draft an amazing deck and proceed to keep a two lander that just needs to draw one land and never get there, and then follow it up with a mull to 4 when none of your 18 land show up. This happens, it is true. But exposing even the best decks to additional rounds increases their odds of losing greatly, and swiss does not reward exposing a winning deck to additional rounds with increasing gains because it takes those packs and deepens the prize pool. Players that honestly feel that they have a very good chance to start 1-1 (or better) most of the time in a swiss pool will usually be better served in a 4322 since then they'll have the chance to get more out of their best decks.
Opportunity - I touched on this earlier, but you don't get *anything* worth more than 2 tix passed passed the second pick in swiss queues, at least in my experience. This means that your ability to "pad" your draft via hawking an occasional rare is entirely eliminated further diminishing the potential total value of your draft. I don't advocate (or participate) in flagrant rare drafting (extreme examples aside), but potentially getting a sphinx as late as pick 3-4 should be one of the rewards of feeding your fellow drafters good signals. *Not* getting a sphinx when the mono-red guy on your right rare drafts it over burst lighting is just aggravating. Obviousl,y 4322 queues are not exactly idyllic PT tables where people draft solely to build the best deck possible... but the swiss queues are just lousy with rare drafters.
This is not to say that swiss is bad per se, and it is GREAT that we have them now, but they do not make sense for everyone. Are many people's perceptions of the "success" in 4322 clouded? Yes. But it actually is the "right" fit for some drafters... but probably not for nearly as many as draft it ... and unfortunately this level of popularity is ultimately the reason why wizards continues to deprive us of a 5322.
What I meant was, considering a tarn is like winning a match anyway, he might as well have gone for it. He even said as much himself earlier when he picked the catacombs.
This kind of data is infinitely more useful than looking at individual cards in a limited environment like your other articles. I'm sure this will be very useful.
I'm curious about what tournaments you used to gather your data. How many tournaments did you use? What is the date range of these tournaments? This is kind of important information to have when judging the usefulness of these kind of statistics because decks tend to evolve in ways that would change their overall win percentages over time.
I've been playing magic and reading articles for quite a long time now, and I have to say this series is one of the most interesting I've ever seen. You've done a fantastic job gathering data and putting it to good use.
While the descriptive content of the articles is fantastic, I'd like to see you do a little more with the data. Specifically, I'd love to see you take those normalized statistics from the second half of the article and show us what happens if you push the makeup of the metagame in a number of directions. What happens to the data if you shift 5% of the metagame back to Jund? What about 10%? What happens if you shift the metagame in favor of UWR control? I think looking at these numbers can help us quantify exactly how subtle and not-so-subtle shifts in the metagame affect the actual viability of deck choices.
Fantastic work, keep it up!
well ill be honest when i opened the article i was looking forwar to Grave Pact and Death Pit Offerings with zombies and such...really a fun deck there, but i like what you did. The deck seems great if not a littlee bit on the pricey side this time around.
Oh, but to be fair with the time thing, I at one point calculated EV/Hour, and it still turns out that it is better to play 84's than 4322's, it just happens sooner, if I recall that, 4322's and 84's become more attractive to someone who is worried about ev and time (something like 5% points I think)
I dunno how I missed them. Basically, you need some sort of artifact or enchantment hate (in the case of leyline), or a pithing needle, or to move very quickly (before they can resolve helm), or Gaea's Blessing to beat Painter/stone.
I disagree with godot on the time thing, 4322's on average are far shorter than swiss, due solely to the fact that they are single elimination.
Assuming on average 40 minutes, rounds, and 25 minutes to draft it works out swiss = 145 min, 4322 = 25+40*loss+80*win*loss + 120*win*win.
since loss = 1-win and win is between 0 and 1, you can see how on average a 4322 can never take as long as a swiss (unless of course you have a 100% chance of winning... but I don't think you'd be complaining then)
Assuming you have something like an unlikely 75 win percent against the filed (which would be high) you save ~30 minutes in on average.
Agree with godot on rare thing.
Agree with godot on the math thing, for handy reference (I don't know if I'm allowed to post external links, please edit it out if I am not)
here's a spreadsheet of your ev in various formats (not taking into account increasing/decreasing difficulty as rounds progress, as it's a spread sheet. maybe I'll program up a terrible gui). I've posted this before but it has been simplified and the number of formats it calculates is larger.
http://sites.google.com/site/whatisfgh/spreadsheet
I think time, convenience(ie it was firing), and practice are the only reasons to draft 4322's (well and probably superstition... did you know a university ran the numbers in football, and it is almost never more profitable to punt than it is to just go for it on the fourth down? Yeah, it's kind of like how teams almost always punt (no pun intented))
Time, convenience are self explanatory.
By practice I mean (assuming an increasing level of difficulty as you progress up the queue which as a below post states might be faulty). In that it is harder to go from drafting with "bad" players to "great" players, than it is to go bad->medium->great, so you may be using 4322's to get a feel for the format to gain access to 84's. As well you may be training for a paper tournament, in which case learning the format on a scaling curve might help.
That said I don't think practice constitutes drafting only 4322's... it should just be a temporary thing.
side note: I'm one of those mid 1700 players in the swiss... just for some reasons I trust my math and know that since I'm less than 50% to beat an average 84er
I could be wrong about this and it is why I wanted to elicit some opinions from other posters. I am right around a 1700 player and don't feel I am good enough to draft 8-4s and win consistently enough conserve money. So I stick to the 4,3,2,2 and the Swiss. Reading these post I feel like people might be missing something. I don't think the 4,3,2,2s are necessarily stocked with better players then the Swiss. On the contrary I think the Swiss likely has a higher average ranking among it's players then the 4,3,2,2. I can't be sure obivously since wizards nolonger lets us see our opponents limited rankings, but just from playing them as often as I do I feel the average level of competition is better in the Swiss draft.
I have actually performed better over time in the 4,3,2,2.
As for reasoning, I think that Godot hit it on the head. Swiss pays out better and has a better "EV". This means that alot o top players tyring to go infinite will go for either 8,4 or Swiss. Look a Godots own numbers 17/25/1. If other players of a similiar caliber have similiar splits, then the it seems much more likely you are going to run alot of high level players in the Swiss and less so in the 4,3,2,2.
Heck Godot's reference to the Math article is flawed since it assumes better players in the 4,3,2,2 then Swiss. However the fact that the "cat is out of the bag" on the EV thing means that top players are likely avoiding the 4,3,2,2.
Just a thought does anyone else feel the 4,3,2,2s have lower average skill the the Swiss??
Since you have to play the bellow to win, if he blocks the larva with the scorpion (since he's dead if he doesn't anyways) he in fact blows by playing it in response.
though more likely if he had disfigure he would just kill the larva, with the trigger on the stack, and you can pump it...
Ok, what I'm saying it that disfigure is the trickiest card to play around there if you want to win that turn.
But yeah, I agree with your assessment that he likely didn't have removal, so you really only have to play around the non removal tricks.
very useful article, clear explanations
no mention on the mill combos ? (helmvoid, painterstone...) Are they currently out of the meta ?
And so? Is there a bias on some archetype for that? So let's not talk about statistical relevance, just about facts we can witness. Take this as the results of an extensive, days long testing session
Neither the builds nor the pilots are standardized. Please stop calling these results statistically significant.
Wouldn't Phyrexian Rager be better than SIgn in Blood? For 1 more mana you get a nice 2/2 blocker instead of a card, and you lose less life.
Double post, sorry, first one gave me an error.
* The time thing is overrated. The only round where you are waiting more than you would in single elimination is round two, and even then, sometimes you are in the slowest match.
* I don't feel there is quite the raredrafting gap between Swiss and 43 that you see. While I obviously haven't drafted a ton of Zendikar 43, I end up drafting plenty of 43 in the nix tix events where it is all that is offered, and drafted a ton of 43 before Swiss was available, and don't feel like there is a huge difference between the level of raredrafting in 43 and Swiss. I admit this is just my sense, it would take some data crunching of a large sample size to prove this either way.
* The raredrafting scenario you mention works both ways. You cite the Sphinx you should have had for your blue deck. What about when a guy two seats upstream from you in pack 3 in BR, takes a Sphinx over the Hideous End, and the next guy in WU passes it to you P3P3 for your BG deck? You are more likely to get potent on-color commons and uncommons later than you should because of raredrafting than you will be denied rare bombs, because frequently the raredraft will not be for a limited bomb, but for a fetchland or something.
* re: variance: I don't reach the same conclusions as you do on this front. For one thing, when you advance in rounds in 43/84, it is generally to face better and better decks. If you lose in the first round of Swiss, you generally go on to face worse decks. Also, see Shaterri's article about the value of raredrafting for a statistical backup about the flatter payscale of Swiss not making 43 a better bet. It simply doesn't make up for that missing pack, or for the fact that the road in Swiss sometimes gets easier instead of harder.
http://puremtgo.com/articles/ev-raredrafting-and-you
It is a very mathy article, but here is a quote regarding the mathematical conclusion drawn about the different queues:
"Comparing EVs for the different queues reveals something interesting: the 4-3-2-2 queues are never the 'best' choice. If your win percentage against the field is less than 50%, then the Swiss queues are your best bet; if it's better than 50%, then 8-4 becomes better than 4-3-2-2. In fact, this still holds if we skew the win percentages to account for different opponent strengths: if you assume that your win percentage against the average Swiss opponent is 5% better than vs. the average 4-3-2-2 opponent, and 5% better against the average 4-3-2-2 than the average 8-4 opponent, then Swiss becomes your best bet up to about a 65% win percentage vs. the field (and thus a 55% chance against the 8-4 opponents), and above that 8-4 takes over again. In fact, Swiss and 8-4 queues still dominate if you assume a 10% strength difference — all that changes is the crossover point."
Finally, on general *principle* people should avoid 43s because of the lower payout. "Sure Wizards! You guys keep an extra pack of profit for this draft, we don't want it. Think of it as a tip from the Magic community!" That's what everyone who enters a 43 queue is saying...
Ironically: the guy with cheap aggro beaters and removal, who's basically playing to lay a few fast threats and keep you from ever generating a blocker, is "okay." The guy who plays a hedron crab and removal, basically trying to keep you from bashing his face in long enough for the crab and trap mills you out... gets the abuse.
But on an objective level, *neither* deck is really any more or less interactive than the other. If you want real non-interactive: play combo.
Which is where the real irony comes in. If your combo involves casting scapeshift to bring in 6 mountains + 2 valakuts, you're "cool" (judging by Scapeshift's popularity in Extended.) But trade those two valakuts for two hedron crabs, and you'll get the hate. Even if the deck is less effective because of that mill approach.
well this is why swiss is easier i think, players just rare draft making it far easier to win multiple packs and raise ratings.
to be honest i was not a big fan of the limited articles but this is great, it so much easier to simply compare w-l records for me and i can understand where the % comes from. Good job!
Just wanted to chime in as another guy with a 1750+ rating (currently 1775) who drafts exclusively Swiss. I like the fact that I get to play 3 matches every time.
I would note that I've drafted ZZZ about 15 times or so, and I've only once played against someone with an equivalent rating (as noted by the fact that my rating changed 8 points after the games. Typically I gain 3-5 points for a win and lose 11-12 for a loss).
I imagine this means that my skill level won't be increasing all that much, but I feel it's worth it, because I really hate drafting, losing in the 1st round, and then not being able to play anymore.
My experience: I am 17-9 with Vamps on MODO (Standard queues and one 4-0 daily event), 50 tickets net gain. Against Jund my record shows 2 wins and 5 losses but I don't feel that the matchup is so bad, I got quite unlucky at least twice and I think that the record will improve in time.
White weenie was good in Mirage/Tempest Standard with shadow guys, cataclysm, tithe, and Empyrial Armor. Was that really the most recent time it was good? Sad.
At the time of this draft, the ZEN Duals had already declined in price enough that the major buy bots wouldn't even buy them from me anymore. I did end up getting about a pack's worth of value out the Verdant Catacombs, but it took some time to find the right buyer. At the time, I seem to remember the Scalding Tarn not being as valuable (although I could be wrong about that - currently, they are valued at MTGOTraders equally); furthermore, since there was a strong pick in its place (Burst Lightning), I never really considered it.
I like the inclusion of test of hypothesis and really appreciate the confidence intervals for the win statistics. Out of curiosity, do you think it is important for the audience to understand the use of both CIs and win rankings to evaluate the winning-ness of decks?
First, well done article. Thanks for the content.
Second, the factors some of you are leaving out in your evaluation of 84 vs. 4322 vs. Swiss are time, variance, and opportunity. Yes, the total payout in 4322 is less. Yes, swiss can be very valuable in providing a larger sample size to evaluate a given draft/drafters skills. However, in terms of total value 4322s have some benefits:
Time - Zendikar is a pretty fast format, but draft + 3 rounds still often takes 2.5 hours, and some formats take much longer. If a draft goes badly awry - actually MORE likely in swiss where everyone rare and hate drafts more than they should - I would rather not be committed to fumbling around trying to pick up an extra pack for the next couple of hours, and be able to jump in another queue and try my luck and skill at balancing out my losses. Of course, I refuse to multi-queue out of common courtesy.
Variance - Variance is usually argued as a reason to play swiss. After all, you can draft an amazing deck and proceed to keep a two lander that just needs to draw one land and never get there, and then follow it up with a mull to 4 when none of your 18 land show up. This happens, it is true. But exposing even the best decks to additional rounds increases their odds of losing greatly, and swiss does not reward exposing a winning deck to additional rounds with increasing gains because it takes those packs and deepens the prize pool. Players that honestly feel that they have a very good chance to start 1-1 (or better) most of the time in a swiss pool will usually be better served in a 4322 since then they'll have the chance to get more out of their best decks.
Opportunity - I touched on this earlier, but you don't get *anything* worth more than 2 tix passed passed the second pick in swiss queues, at least in my experience. This means that your ability to "pad" your draft via hawking an occasional rare is entirely eliminated further diminishing the potential total value of your draft. I don't advocate (or participate) in flagrant rare drafting (extreme examples aside), but potentially getting a sphinx as late as pick 3-4 should be one of the rewards of feeding your fellow drafters good signals. *Not* getting a sphinx when the mono-red guy on your right rare drafts it over burst lighting is just aggravating. Obviousl,y 4322 queues are not exactly idyllic PT tables where people draft solely to build the best deck possible... but the swiss queues are just lousy with rare drafters.
This is not to say that swiss is bad per se, and it is GREAT that we have them now, but they do not make sense for everyone. Are many people's perceptions of the "success" in 4322 clouded? Yes. But it actually is the "right" fit for some drafters... but probably not for nearly as many as draft it ... and unfortunately this level of popularity is ultimately the reason why wizards continues to deprive us of a 5322.
What I meant was, considering a tarn is like winning a match anyway, he might as well have gone for it. He even said as much himself earlier when he picked the catacombs.
because he's in a swiss que each win only means one pack thus picking a card worth on pack ='s the same as a game win.