I have added all of the non-masterpiece cards to my sealed pool generator. I will be adding the masterpieces soon. You can see use the Amonkhet Sealed Pool Maker now. As always, please let me know if you encounter any issues.
I feel like there's some seriously unrealistic amounts of duplicates and triplicates being generated this time around. I think something may be off kilter with the generator. I don't feel like in the past that I've seen so much duplicating. Obviously Commons are Common but even then I don't think I've ever gotten multiple copies of as many Commons as I'm getting in these pools, at least not at any actual Sealed event I've ever done. Here's a few of the worst offending pools as I tried generating new ones* to get what I thought was a more realistic representation:
* I generated maybe two or three more than actually linked here, so the sample size isn't large but small enough to show consistency in how unrealistic the duplicate/triplicate cards showing up appears right now
the fat pack (now bundle) still generates 9 boosters instead of 10
Private Mod Note
():
Rollback Post to RevisionRollBack
"The Wheel of Time turns, and Ages come and pass, leaving memories that become legend. Legend fades to myth, and even myth is long forgotten when the Age that gave it birth comes again"
There's a similar problem with Blazing Volley. Or maybe Trial of Zeal? Either way, my pool has them in reverse order. Trial of Zeal being after Blazing Volley.
Hello Ottomund, i would like to give you some 'feedback/improvement suggestions' on your pre-release (and maybe even sealed in general) simulator. As i have been using it many times in preparation for many pre-releases. And seeing i can't find an email on the site, and happen to stumble on this post, i thought i'dd do it here ^^
-------------
There is a definite 'problem' with duplicates, both at the rare and uncommon levels of the card picker. And i think your simulator would be a lot more accurate in it's pool generation if you would include the following restrictions to the picks:
1. The section in the code that picks the rares should check the rares already picked, if a rare is already picked the simulator should pick another rare. As there is pretty much no real chance of there EVER being two of the same rares in a pool. While there might be a 0.00001% chance the promo is the same as one of the rares in your packs, this chance is so small you should probably ignore it... and seeing getting double rares is thus so small, you should make sure they don't happen...
-> You can probably stick the rares picked in an array, check the array for same names, if a name is the same, pick another, add that to the array instead.
2. The same 'somewhat' applies to uncommons, but the chance on getting the same uncommon is most definitely there. Now 'we' all know this is more likely for the 'tier 2' uncommons, and seeing 'we' do not know the print run details it is hard to determine which uncommons are tier1 and which tier2. But to at least get more realistic results for pools you should probably do something similar as described for the rares.
-> You can probably stick the uncommons in an array, with a 'count' linked to them, then a check 'if count is <1 add to the array', if count is 1 flip a coin, roll a die even/uneven add to array or pick another, if count is 2 then pick another (or roll a 20 sided die, if roll = 11 (fools number) add to array, else pick another (if you want to make it possible to have 3 same uncommons, though i feel the chance on that should probably be 0). Maybe the chance on getting two of the same uncommon should be smaller, say 1 in 6 (keeping the chance limited to 1 of the uncommons in the 6 packs to be the same as another in the 6 packs.
*!* I feel like the above SHOULD happen, as more often than not i feel i have to generate another pool simply because the one coming out is just unrealistic. Which i base upon the sealed pools i have had myself, seen online, as well drafts in both. Having 2 same rares never happens, having 2 same uncommons is a rarity (a bit more likely on the tier2 uncommons, but seeing 'we' don't know them, just ignore that).
3. The uncommons are just fine tbh. as there are most definite chances of getting more than 1, upto 3 or sometimes even 4 or 5 (the rule you can have that many in your deck stems from the chance being realistic). And seeing the chance of 'random picking' from the great number of uncommons, already leads to a small chance on getting 4, this works just fine! (it's mainly the smaller pools of rares/mythics, uncommons, that lead to the currently random pick to pick a second, while the printrun used by WotC is set up to avoid that from happening, so you should make sure it doesn't happen in your simulator either)
---------------
I would very much like it to be able to 'sort' the cards i picked for my deck, as i can do with the entire pool. Seeing the converted mana cost is available in your data base (you use it for the mana-curve), sorting based upon mana-cost or color would be so helpful in understanding the build and making decisions.
---------------
The site can probably do with an polish, and a more general 'home-page', currently there is still a 'dragons of tarkir' announcement? Giving the home-page just a general look and some general tekst will keep it 'upto-date' (cough for years to come, without having to spend time on it. Which you likely do not have, else you would have kept it upto date, since it isn't, spend some time on a more general main page and call it good.
---------------
Other than the above, of which the duplicates might well be the only thing driving me to another simulator as the unrealistic pools are simply annoying. I really like the simulator in general, the option to deck build, and sample hands/draw even are fantastic. And the main reason i tend to pick your site over others. But again, those duplicates/unrealistic pools are a pain and you should really do something about them !
Also, tnx a lot for continuing to put new sets into the simulator, i genuinely enjoy using your simulator (most of the time) in preparation for the new season. And i hope this feedback/recommendations will help you improve the simulator to lead to more realistic pools, so that my preparation can be more in line with the actual experience.
As there is pretty much no real chance of there EVER being two of the same rares in a pool.
...
Having 2 same rares never happens, having 2 same uncommons is a rarity (a bit more likely on the tier2 uncommons, but seeing 'we' don't know them, just ignore that).
Arghore, I have a slight problem with part of what you said.
The packs you get in a prerelease are random, so there certainly is a chance you will get duplicates of rares (even potentially mythics), and its not a negligible chance.
While ancedotely evidence, I myself opened two copies of Skyship Stalker at the Kaladesh prerelease, both regular versions, neither was my promo.
I would liken this to the Birthday Paradox (where in a room of 23 people the odds of two people sharing the same birthday is 50% as explained on this link). I believe our case does end up more complicated than that though, just due to mythic rares and their distribution making the maths trickier.
If we ignore the odds of a mythic rare and just count them as rares for the moment, then we have the following:
There are 53 rares in Amonkhet and 15 mythics, so 68 total.
At a prerelease, you get 7 rares - 6 from boosters, plus a promo.
So what are the odds of getting two of the same:
1 - (67/68)^(7*6/2) = 0.2673717012...
So just over a quarter of the time.
Now as I said, this isn't taking into account the distribution of Mythics which will probably play havoc with my odds.
But to say that there is no chance to get duplicate rares, is stretching it. Random is random.
I have added all of the non-masterpiece cards to my sealed pool generator. I will be adding the masterpieces soon. You can see use the Amonkhet Sealed Pool Maker now. As always, please let me know if you encounter any issues.
RCRDaretti: Superfriends Forever RCR
WGBDoran: Ent-mootWBG
GGGMultani: Group Bear HugGGG
GB(B/G)The Gitrog Monster: Dredgefall DurdleGB(B/G)
RGWGahiji, the Honored Group Hug MonsterRGW
UB(U/B)Yuriko, Ninja Trinket AggroUB(U/B)
WUBRGAtogatog: Assembling a OHKOWUBRG
http://www.magicdrafting.com/savedpools/32173
http://www.magicdrafting.com/savedpools/32174
http://www.magicdrafting.com/savedpools/32175
http://www.magicdrafting.com/savedpools/32176
http://www.magicdrafting.com/savedpools/32177
* I generated maybe two or three more than actually linked here, so the sample size isn't large but small enough to show consistency in how unrealistic the duplicate/triplicate cards showing up appears right now
(Also known as Xenphire)
-------------
There is a definite 'problem' with duplicates, both at the rare and uncommon levels of the card picker. And i think your simulator would be a lot more accurate in it's pool generation if you would include the following restrictions to the picks:
1. The section in the code that picks the rares should check the rares already picked, if a rare is already picked the simulator should pick another rare. As there is pretty much no real chance of there EVER being two of the same rares in a pool. While there might be a 0.00001% chance the promo is the same as one of the rares in your packs, this chance is so small you should probably ignore it... and seeing getting double rares is thus so small, you should make sure they don't happen...
-> You can probably stick the rares picked in an array, check the array for same names, if a name is the same, pick another, add that to the array instead.
2. The same 'somewhat' applies to uncommons, but the chance on getting the same uncommon is most definitely there. Now 'we' all know this is more likely for the 'tier 2' uncommons, and seeing 'we' do not know the print run details it is hard to determine which uncommons are tier1 and which tier2. But to at least get more realistic results for pools you should probably do something similar as described for the rares.
-> You can probably stick the uncommons in an array, with a 'count' linked to them, then a check 'if count is <1 add to the array', if count is 1 flip a coin, roll a die even/uneven add to array or pick another, if count is 2 then pick another (or roll a 20 sided die, if roll = 11 (fools number) add to array, else pick another (if you want to make it possible to have 3 same uncommons, though i feel the chance on that should probably be 0). Maybe the chance on getting two of the same uncommon should be smaller, say 1 in 6 (keeping the chance limited to 1 of the uncommons in the 6 packs to be the same as another in the 6 packs.
*!* I feel like the above SHOULD happen, as more often than not i feel i have to generate another pool simply because the one coming out is just unrealistic. Which i base upon the sealed pools i have had myself, seen online, as well drafts in both. Having 2 same rares never happens, having 2 same uncommons is a rarity (a bit more likely on the tier2 uncommons, but seeing 'we' don't know them, just ignore that).
3. The uncommons are just fine tbh. as there are most definite chances of getting more than 1, upto 3 or sometimes even 4 or 5 (the rule you can have that many in your deck stems from the chance being realistic). And seeing the chance of 'random picking' from the great number of uncommons, already leads to a small chance on getting 4, this works just fine! (it's mainly the smaller pools of rares/mythics, uncommons, that lead to the currently random pick to pick a second, while the printrun used by WotC is set up to avoid that from happening, so you should make sure it doesn't happen in your simulator either)
---------------
I would very much like it to be able to 'sort' the cards i picked for my deck, as i can do with the entire pool. Seeing the converted mana cost is available in your data base (you use it for the mana-curve), sorting based upon mana-cost or color would be so helpful in understanding the build and making decisions.
---------------
The site can probably do with an polish, and a more general 'home-page', currently there is still a 'dragons of tarkir' announcement? Giving the home-page just a general look and some general tekst will keep it 'upto-date' (cough for years to come, without having to spend time on it. Which you likely do not have, else you would have kept it upto date, since it isn't, spend some time on a more general main page and call it good.
---------------
Other than the above, of which the duplicates might well be the only thing driving me to another simulator as the unrealistic pools are simply annoying. I really like the simulator in general, the option to deck build, and sample hands/draw even are fantastic. And the main reason i tend to pick your site over others. But again, those duplicates/unrealistic pools are a pain and you should really do something about them !
Also, tnx a lot for continuing to put new sets into the simulator, i genuinely enjoy using your simulator (most of the time) in preparation for the new season. And i hope this feedback/recommendations will help you improve the simulator to lead to more realistic pools, so that my preparation can be more in line with the actual experience.
Arghore, I have a slight problem with part of what you said.
The packs you get in a prerelease are random, so there certainly is a chance you will get duplicates of rares (even potentially mythics), and its not a negligible chance.
While ancedotely evidence, I myself opened two copies of Skyship Stalker at the Kaladesh prerelease, both regular versions, neither was my promo.
I would liken this to the Birthday Paradox (where in a room of 23 people the odds of two people sharing the same birthday is 50% as explained on this link). I believe our case does end up more complicated than that though, just due to mythic rares and their distribution making the maths trickier.
If we ignore the odds of a mythic rare and just count them as rares for the moment, then we have the following:
There are 53 rares in Amonkhet and 15 mythics, so 68 total.
At a prerelease, you get 7 rares - 6 from boosters, plus a promo.
So what are the odds of getting two of the same:
1 - (67/68)^(7*6/2) = 0.2673717012...
So just over a quarter of the time.
Now as I said, this isn't taking into account the distribution of Mythics which will probably play havoc with my odds.
But to say that there is no chance to get duplicate rares, is stretching it. Random is random.
Normally, I use tappedout.net, but their card database seems to be having some major issues right now.
Any other options like it?