the most hate drafting I'll do is taking a late pick which would be good for another deck. I'll only do this if their is nothing that I would consider playing in my colors.
While I agree with the OP in general, I don't necessarily agree that taking a card that would make your deck over hatedrafting is always the correct pick. If your deck is extremely vulnerable to a specific card it may be correct to hatedraft that card rather than picking a marginally better 20th-23rd card.
Put in mathematical terms it would be like:
1/7*(HD-O1) > MD-O2
where
HD = value of hatedrafted card against your deck
O1 = value of card that would have been replaced by hatedrafted card
MD = value of maindeckable card in pack
O2 = value of card that would have been replaced by maindeckable card in pack
This is of course a simplification, but you may get the idea.
I guess I should note that the main point of this thread is that most players vastly overrate hatedrafting, and so I want to properly explain why it's in general a LOT worse than they think it is. There are exceptions to this rule as well, of course, but this thread was intended to make this thread to clear up a common misconception that many players struggle with, and for simplicity I didn't want to bring up too many complicated exceptions. Most experienced players are already familiar with this anyway, and don't really need to read it.
Magic is never absolute, but I think it's a net win for players if they don't hatedraft at all rather than doing it as much as they currently do.
Very useful discussion. I 'learned how to draft' in IPA block at a LGS (at the time) where it was exclusively team draft, normal style.
You NEVER hate draft in that format, as noted.
Anyway, I thought I might try to give a bit more quantitativeness to the discussion. To that end, I built a spreadsheet that takes certain inputs and outputs the likelihood of winning a draft (3 rounds of single-elimination) with Hate drafting / NonHate drafting.
The input parameters are:
General game win% of your deck.
General game win% of the deck of whom you are hatedrafting.
Hatedrafting factor.
The assumptions of this model are as follows:
1) Hatedrafting hurts your deck's performance against a general opponent.
2) Hatedrafting helps your deck's performance against the hatedrafted opponent.
3) Hatedrafting hurts the hatedrafted deck's performance.
The Hatedrafting factor mentioned above is used to calculate the impact of the hatedraft itself. The higher the factor, the greater the detriment to your deck and the hatedrafted deck's performance, and the greater your deck's improvement over the hatedrafted deck. (This is one factor for the sake of convenience. The model can be easily modified to decouple the factor.)
For example, given the inputs of
Your general game win%: 0.7
The hatedrafted deck's general game win%: 0.7
Hatedraft factor: 0.02
The probability that you will win the draft without hate-drafting is 0.44, while the probability that you will win the draft with hate-drafting is 0.43.
The most interesting result I found is that as the Hatedraft factor goes up, the Hatedraft performance goes down. The reason is from assumption 3). Basically, if you hurt another deck's performance by hatedrafting, you also decrease the likelihood of facing that deck in the draft.
Anyway, if anyone wants to take a look at my spreadsheet, they are more than welcome to do so.
Ulthwithian- If I understand your model correctly, you are setting the magnitude of effect 1) and 3) equal. A greater hate-draft factor means you are hurting your opponent more, but also yourself more. Of course a greater hate draft factor is going to lower your chances of winning, because 1) is much more relevant than 3)
Not a bad model, since it's actually optimistic in most cases. As noted already It's difficult to get the magnitude of 3) greater than 1), since the hatedrafted deck is going to get a potentially useful pick where you aren't.
I think there is some interesting work to be done if you give separate variables to 1) and 3).
P.S. Were we arguing about milling in some thread some time ago?
Guruu: At the least, the factor is scaled by a constant for 1) and 3).
Currently in the model:
1) Your general win% is reduced by the factor if you hatedraft.
2) Your opponent's general win% is reduced by twice the factor if your hatedraft.
3) Your win% against the hatedrafted opponent is reduced by 10% of your win% if you don't hatedraft.
4) Your win% against the hatedrafted opponent is increased by twice the factor (from 3)).
I'll note that the above, to you, is very optimistic, as I assume that you always hurt the opponent's deck by twice the amount you hurt your own.
Using separate factors is quite easy. What I really need to do is Monte Carlo simulation or some other method that can be statistically tested. Academically, I would be reamed if I presented these results.
I made a 'quick and easy' model, but the question I hope to answer with it is, 'How bad do the win%s have to become (from my perspective) before hate drafting becomes profitable?'
I probably should create a Markov model to get better win%s. Information from posters here would help.
I posted quite a few statistics arguments in the Trepanation Blade thread. Perhaps that is where you saw it?
mordath: The actual key to whether you should hatedraft is the improvement in the performance of your deck against the hatedrafted deck vs. the decrease in the performance of your deck in general. Because of this, Neon-chan's argument is correct.
I think that a useful definition is needed for 'hatedrafting'.
Judging from this thread, and my own thinking, the following seems useful.
Hatedrafting - The act of deliberately making a selection in a draft with a negative opportunity cost for your own success.
By the above definition, Wintersky, any hatedrafting is 'harmful to your chances' by definition.
More generally, the act of passing 'good cards' as a signaling move is better when the marginal utility of the 'good card' is low; i.e., you are not increasing the quality of the opponent's deck very much. This implies that this is best done in an already-deep color.
The guy who "feeds" me pack1 and pack3 is blue black and I am white red. Things go smoothly ( I happen to have opened contagion engine in pack2). Got passed a blade splicer p1p2 because the blue black guy got a foil black sun's zenith.
Pack3pick2 I get sunblast angel, I think he took something like a blue flyer. Together with a sacrifice artifact outlet and razor hippogriff I end up having 3 wrath effects in my deck.
Match3 we play together, and he gets totally destroyed by the multiple one sided wrath effects (more like plague wind). We both agreed that since he knew I was strong in white (with multiple white bombs/one side mass removal and flyers) he should have hatedrafted sunblast angel...
Opinions?
how does one open NPH card(that also had a foil mbs card) in a mbs scars scars draft?
What I find alarming is that for example LSV in his videos often does the following: Once he has arrived at the late picks (say 10th forward), you regularly hear him say "ok, what do I want to play against the least" or something along those lines and he proceeds to take a card that he wouldn't like to see on his opponents board. For example playing UW flyer and taking a Piston Strike 11th pick over very weak blue cards, like Turn the Tide or Fuel for the Cause in a non-infect/proliferate deck. I really can't tell whether this is good or dangerous.
It's good, it's how I draft.
My descision tree is this:
Bomb in my color/colorless
Removal in my color/colorless
Good creature in my color/colorless
Card I don't want to see across from me
Utility card/avergae creature in my color/colorless
I have hate drafted 1st pick if there is nothing meeting my top 3 criteria. Just last week P1P2 and P1P3 I hate drafted Invisible Stalker even though I was solid RWG.
From a purely mathematical standpoint I think it's very obvious why hate drafting instead of taking a card that will improve your deck is a very poor choice.
Think of it this way, the one card that will help your deck you will play many times during the draft while the card you're so frightened of you will probably not see the whole night.
Last night this guy hate drafted a Huntmaster of the Fells playing UW and I was playing RG and I played him first match and crushed him.
From a purely mathematical standpoint I think it's very obvious why hate drafting instead of taking a card that will improve your deck is a very poor choice.
Think of it this way, the one card that will help your deck you will play many times during the draft while the card you're so frightened of you will probably not see the whole night.
Last night this guy hate drafted a Huntmaster of the Fells playing UW and I was playing RG and I played him first match and crushed him.
also: first post yay
I don't think taking Huntsmaster is a great example
Because it's more of a money draft than a hate draft.
Take the example of Grafdigger's Cage. It's essentially a blank card in limited (I guess it's an ok SB card vs flashback decks, but w/e). However, it's worth a non-trivial amount of money and someone might well choose to draft if for reasons other than winning the draft.
So the huntmaster guy *might* have been hate drafting you, or he might have just been money drafting.
Taking Mind Control 2nd pick when you're not in blue in pack 3 is a hate draft. I've had that happen a lot in M12 because I almost always played blue and people knew it. I HATE that. But I know I've done it and will do it as long as I'm not wasting a pick for my maindeck or taking a solid sideboard card. I hate to pass Geist-Honored Monk in Innistrad.
Because it's more of a money draft than a hate draft.
Take the example of Grafdigger's Cage. It's essentially a blank card in limited (I guess it's an ok SB card vs flashback decks, but w/e). However, it's worth a non-trivial amount of money and someone might well choose to draft if for reasons other than winning the draft.
So the huntmaster guy *might* have been hate drafting you, or he might have just been money drafting.
Yea, we rare draft at home so I got the card in the end. But yea, I see the point now, should have made that clear.
hatedrafted a lot of white bombs in my packs... ended up coming 2nd (his topdeck was 1 turn in front of mine, lol - he played)
let me clarify though.
hatedrafted so much white that i ended up going white by the 3rd hatedraft because i was all... white is open, these cards are sick and sac'ing mausoleum guard to feed the pack is amazing
To me, I would only hatedraft when there are no good picks in my colors, or if I know what the players next to me would be running and I get a bomb that helps them a great deal. A great example is one of the players at my FNM drafts is ALWAYS white (mostly mono .... >.< so annoying!), and if he opens an annoying bomb, he would just hate it. A great example was that he hate drafted deadeye navigator pack 1 pick 1 due to he learned how annoying it is, but does not even bother going blue afterwards.
Consider how good the card you'd be hating is versus *your deck* not just in general. Say you've got a DII Rg deck composed mostly of vampires and werewolves. Victim of night doesn't do much against you and you should probably pass it , like, p3p7 even if there's only a marginal sideboard card for your deck. That way, the black deck has a better chance against the UW deck with elite inquisitor, slayer of the wicked, etc when they square off round 1.
You want to give the decks that you aren't afraid of an edge against the rest of the field.
I agree, I always follow the Maindeckable->Sideboard->Hatedraft rule. The only time I would take something out the ordinary is if it was worth some deal of money, because hey...I like money.
How is there still discussion on this? "Hatedrafting" shouldn't be happening. Period. Picking a strong card after there's nothing left in a pack you will play isn't really hate drafting. It's just taking the strongest card. And taking money cards isn't hate drafting... that's just cashing in.
Hate drafting only causes you to pass on a good *playable* card for something you won't play. This is the definition of a stupid pick. Don't do it.... ever. End of discussion.
Taking something that has no chance of making your deck is just a wasted pick. You at least get some value by taking a playable card out of the pack in that situation.
This.
Private Mod Note
():
Rollback Post to RevisionRollBack
Green = Creation of life
White = Preservation of life
Black = Death
Red = Destruction
Blue = ...Wizards?
I already had a complete rakdos deck and more rakdos cards to add in that 3rd pack round.
In the end the Mercurial I hatedrafted proved to be about 1 of 3 bomb rares in the whole pool, 1 of them was a Vraska, I could have splashed the Mercurial in my deck but I didn't feel I had the need to.
If you're Rakdos it could make sense as a splash (though you don't really want to splash in Rakdos), but if it was a pure hatedraft, chances are you were dead wrong Need to see the pack first though.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
1/7*(HD-O1) > MD-O2
where
HD = value of hatedrafted card against your deck
O1 = value of card that would have been replaced by hatedrafted card
MD = value of maindeckable card in pack
O2 = value of card that would have been replaced by maindeckable card in pack
This is of course a simplification, but you may get the idea.
For example, if you already have 4 Invisible Stalkers and a few Butcher's Cleavers, hatedrafting Rolling Temblor/Curse of Death's Hold over a marginally playable card seems logical to me.
Magic is never absolute, but I think it's a net win for players if they don't hatedraft at all rather than doing it as much as they currently do.
You NEVER hate draft in that format, as noted.
Anyway, I thought I might try to give a bit more quantitativeness to the discussion. To that end, I built a spreadsheet that takes certain inputs and outputs the likelihood of winning a draft (3 rounds of single-elimination) with Hate drafting / NonHate drafting.
The input parameters are:
General game win% of your deck.
General game win% of the deck of whom you are hatedrafting.
Hatedrafting factor.
The assumptions of this model are as follows:
1) Hatedrafting hurts your deck's performance against a general opponent.
2) Hatedrafting helps your deck's performance against the hatedrafted opponent.
3) Hatedrafting hurts the hatedrafted deck's performance.
The Hatedrafting factor mentioned above is used to calculate the impact of the hatedraft itself. The higher the factor, the greater the detriment to your deck and the hatedrafted deck's performance, and the greater your deck's improvement over the hatedrafted deck. (This is one factor for the sake of convenience. The model can be easily modified to decouple the factor.)
For example, given the inputs of
Your general game win%: 0.7
The hatedrafted deck's general game win%: 0.7
Hatedraft factor: 0.02
The probability that you will win the draft without hate-drafting is 0.44, while the probability that you will win the draft with hate-drafting is 0.43.
The most interesting result I found is that as the Hatedraft factor goes up, the Hatedraft performance goes down. The reason is from assumption 3). Basically, if you hurt another deck's performance by hatedrafting, you also decrease the likelihood of facing that deck in the draft.
Anyway, if anyone wants to take a look at my spreadsheet, they are more than welcome to do so.
Not a bad model, since it's actually optimistic in most cases. As noted already It's difficult to get the magnitude of 3) greater than 1), since the hatedrafted deck is going to get a potentially useful pick where you aren't.
I think there is some interesting work to be done if you give separate variables to 1) and 3).
P.S. Were we arguing about milling in some thread some time ago?
Currently in the model:
1) Your general win% is reduced by the factor if you hatedraft.
2) Your opponent's general win% is reduced by twice the factor if your hatedraft.
3) Your win% against the hatedrafted opponent is reduced by 10% of your win% if you don't hatedraft.
4) Your win% against the hatedrafted opponent is increased by twice the factor (from 3)).
I'll note that the above, to you, is very optimistic, as I assume that you always hurt the opponent's deck by twice the amount you hurt your own.
Using separate factors is quite easy. What I really need to do is Monte Carlo simulation or some other method that can be statistically tested. Academically, I would be reamed if I presented these results.
I made a 'quick and easy' model, but the question I hope to answer with it is, 'How bad do the win%s have to become (from my perspective) before hate drafting becomes profitable?'
I probably should create a Markov model to get better win%s. Information from posters here would help.
I posted quite a few statistics arguments in the Trepanation Blade thread. Perhaps that is where you saw it?
mordath: The actual key to whether you should hatedraft is the improvement in the performance of your deck against the hatedrafted deck vs. the decrease in the performance of your deck in general. Because of this, Neon-chan's argument is correct.
Judging from this thread, and my own thinking, the following seems useful.
Hatedrafting - The act of deliberately making a selection in a draft with a negative opportunity cost for your own success.
By the above definition, Wintersky, any hatedrafting is 'harmful to your chances' by definition.
More generally, the act of passing 'good cards' as a signaling move is better when the marginal utility of the 'good card' is low; i.e., you are not increasing the quality of the opponent's deck very much. This implies that this is best done in an already-deep color.
how does one open NPH card(that also had a foil mbs card) in a mbs scars scars draft?
It's good, it's how I draft.
My descision tree is this:
Bomb in my color/colorless
Removal in my color/colorless
Good creature in my color/colorless
Card I don't want to see across from me
Utility card/avergae creature in my color/colorless
I have hate drafted 1st pick if there is nothing meeting my top 3 criteria. Just last week P1P2 and P1P3 I hate drafted Invisible Stalker even though I was solid RWG.
But when i do i make sure to wreck a spesific archtype for the heck of it.
From a purely mathematical standpoint I think it's very obvious why hate drafting instead of taking a card that will improve your deck is a very poor choice.
Think of it this way, the one card that will help your deck you will play many times during the draft while the card you're so frightened of you will probably not see the whole night.
Last night this guy hate drafted a Huntmaster of the Fells playing UW and I was playing RG and I played him first match and crushed him.
also: first post yay
I don't think taking Huntsmaster is a great example
uhh why
Take the example of Grafdigger's Cage. It's essentially a blank card in limited (I guess it's an ok SB card vs flashback decks, but w/e). However, it's worth a non-trivial amount of money and someone might well choose to draft if for reasons other than winning the draft.
So the huntmaster guy *might* have been hate drafting you, or he might have just been money drafting.
JAMMIT DIM! I'm a DOCTOR not a DECKBUILDER!
Yea, we rare draft at home so I got the card in the end. But yea, I see the point now, should have made that clear.
let me clarify though.
hatedrafted so much white that i ended up going white by the 3rd hatedraft because i was all... white is open, these cards are sick and sac'ing mausoleum guard to feed the pack is amazing
I really want Ajani Vengeant sleeves. And foil planeswalkers. Check my trade thread!
You want to give the decks that you aren't afraid of an edge against the rest of the field.
Modern SkRed
Current Decks:
Pauper: UUDelver BlueUU
Hate drafting only causes you to pass on a good *playable* card for something you won't play. This is the definition of a stupid pick. Don't do it.... ever. End of discussion.
This.
White = Preservation of life
Black = Death
Red = Destruction
Blue = ...Wizards?
I think he confused hate draft with rare draft.
I would be playing... anything and the only thing I would maybe take over a huntmaster is a foil huntmaster*...
Maybe Sorin or one of the other 15-20 dollar cards from the set.
Also how do you hate draft Huntmaster? Dka is the first pack in Isd block draft.
I already had a complete rakdos deck and more rakdos cards to add in that 3rd pack round.
In the end the Mercurial I hatedrafted proved to be about 1 of 3 bomb rares in the whole pool, 1 of them was a Vraska, I could have splashed the Mercurial in my deck but I didn't feel I had the need to.
If you're Rakdos it could make sense as a splash (though you don't really want to splash in Rakdos), but if it was a pure hatedraft, chances are you were dead wrong Need to see the pack first though.